Foreign Agricultural 
Service, United States 
Department of Agriculture
Benin McGovern-Dole Project
FINAL EVALUATION 
September 2018
DISCLAIMER: This publication was produced at the request of the United States Department of 
Agriculture. It was prepared independently by Advisem Services Inc. The author’s views expressed in 
this publication do not necessarily reflect the views of the United States Department of Agriculture or 
the United States Government. 
Endline Evaluation of Catholic Relief Services’ McGovern‐Dole
International Food for Education and Child Nutrition Program
Agreement Number: FFE-680-2014-046-00
Funding Year: Fiscal Year 2014 
Project Duration: 2014-2018 
Implemented by: Catholic Relief Services
Evaluation Authored by: Advisem Services Inc.
 
 
 
Endline Evaluation of Catholic Relief Services’ McGovern‐Dole 
International Food for Education and Child Nutrition Program 
FINAL REPORT 
Presented to Catholic Relief Services 
Submitted by Advisem Services Inc. 
Ottawa, Canada 
September 2018 

Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   i
CONTENTS 
Abbreviations, Acronyms and Symbols ........................................................................................................... iii
Executive Summary ........................................................................................................................................... v
1. Introduction ........................................................................................................................................... 1
2. Introduction ........................................................................................................................................... 2
2.1 Context ........................................................................................................................................ 2
2.2 Project Description ...................................................................................................................... 2
2.3 Evaluation Purpose ..................................................................................................................... 4
3. Methodology ......................................................................................................................................... 6
3.1 Design Framework ...................................................................................................................... 6
3.2 Lines of Enquiry ........................................................................................................................... 6
3.3 Sampling Strategy ....................................................................................................................... 9
3.4 Data Analysis ............................................................................................................................. 22
3.5 Challenges, Issues and Limitations ........................................................................................... 25
4. Evaluation Findings .............................................................................................................................. 29
4.1 Relevance .................................................................................................................................. 29
4.1.1 Strategies to Improve Children’s Literacy ..................................................................................... 29
4.1.2 Strategies to Improve Enrollment and Attendance ...................................................................... 32
4.1.3 Strategies to Improve Community Participation and Engagement ............................................. 33
4.1.4 Stakeholder Satisfaction ............................................................................................................... 34
4.2 Effectiveness ............................................................................................................................. 37
4.2.1 Overall Attainment of Targets ...................................................................................................... 38
4.2.2 Results in Improving the Literacy of School‐Age Children ........................................................... 38
4.2.3 Results Increasing the Use of Health and Dietary Practices ......................................................... 53
4.2.4 Evidence from FGDs ...................................................................................................................... 55
4.3 Efficiency ................................................................................................................................... 56
4.3.1 Timely Achievement of Project Objectives ................................................................................... 56
4.3.2 Implementation Efforts and Coordination Between Project Partners ......................................... 57
4.3.3 Monitoring of Project Activities .................................................................................................... 58
4.3.4 Actioning of Recommendations Made in the MTE Report .......................................................... 58
4.4 Sustainability ............................................................................................................................. 59
4.4.1 Activities/Outcomes Most Likely to be Maintained ..................................................................... 59
4.4.2 Contributing Factors to Sustainability .......................................................................................... 64
4.4.3 Government Contribution ............................................................................................................ 64
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
CONTENTS (CONTINUED) 
ii 
4.5 Impact........................................................................................................................................ 65
4.5.1 Quantitative Evidence on Desired Effects .................................................................................... 65
4.5.2 Discussion ...................................................................................................................................... 68
5. Conclusion and Lessons Learned ......................................................................................................... 70
5.1 Conclusion ................................................................................................................................. 70
5.2 Lessons Learned ........................................................................................................................ 71
6. Recommendations ............................................................................................................................... 73
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
iii 
ABBREVIATIONS, ACRONYMS AND SYMBOLS 
ACE  Agent contractuel de l’État [government contractor] 
AME  Association des mères d’élèves [school mothers’ association] 
APE  Association des parents d’élèves [parent‐teacher association] 
CCS  Head of school district [chef de circonscription scolaire] 
CFAF  African Financial Community franc 
CI  Cours d’initiation [introductory class] 
CP  Cours préparatoire [preparatory class] 
CRS  Catholic Relief Services 
DAC  Development Assistance Committee 
DANA  Direction de l’alimentation et de la nutrition appliquée [Food and Applied Nutrition 
Directorate] 
DAS  Direction de l’alimentation scolaire [School Feeding Directorate] 
DEP  Direction de l’enseignement primaire [Primary Education Directorate] 
DID  Difference‐in‐difference 
DIP  Direction de l’inspection pédagogique [Pedagogical Inspection Directorate] 
EGRA  Early Grade Reading Assessment 
FFW  Food for Work 
FGD  Focus group discussion 
GoB  Government of Benin 
INFRE  Institut national pour la formation et la recherche en éducation [National Institute for 
Training and Research in Education] 
INGO  International non‐governmental organization 
IPTW  Inverse probability weighting 
KII  Key informant interview 
M&E  Monitoring and evaluation 
MEAL  Monitoring and evaluation, accountability and learning 
MEMP  Ministère de l’enseignement maternel et primaire [Ministry of Preschool and Primary 
Education] 
MGD  McGovern‐Dole 
MTE  Mid‐term evaluation 
NGO  Non‐governmental organization 
ODK  Online Data Kit 
OECD  Organization for Economic Cooperation and Development 
PMF  Performance measurement framework 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
ABBREVIATIONS, ACRONYMS AND SYMBOLS (CONTINUED) 
iv 
PMP  Performance monitoring plan 
PTA  Parent‐teacher association 
PTR  Pupil‐teacher ratio 
RNG  Random number generator 
ROA  Rapid organizational assessment 
SILC  Savings and internal lending community 
SO  Strategic objective 
THR  Take‐home ration 
ToC  Theory of change 
ToRs  Terms of Reference 
UN  United Nations 
UNDP  United Nations Development Programme 
UNICEF  United Nations Children’s Fund 
US  United States of America 
USAID  United States Agency for International Development 
USDA  United States Department of Agriculture 
WASH  Water sanitation and hygiene 
WEI  World Education, Inc. 
WFP  World Food Programme 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   v
EXECUTIVE SUMMARY 
The McGovern‐Dole International Food for 
Education and Child Nutrition Program 
(hereafter referred to as McGovern‐Dole) is a 
four‐year project that aims to improve literacy 
and alleviate hunger for approximately 43,804 
primary school‐age children attending 144 
primary schools in four communes in the Alibori 
and Borgou Departments of Northeast Benin. 
The USD 19 million project is funded by the 
United States Department of Agriculture (USDA). 
Catholic Relief Services (CRS) implements the 
McGovern‐Dole project, in partnership with 
World Education, Inc. (WEI), and coordinates 
with government ministries and departments, 
principals and teachers, and parent groups. 
The project’s theory of change (ToC) proposes to 
contribute to a productive Beninese citizenry by 
making all girls and boys healthy and providing 
them with a quality and complete primary 
education. It posits that if the project promotes 
the importance of education to parents, 
improves teacher incentives and capacity, 
advocates for policy application, 
decentralization, and parent involvement, and 
provides inputs (school infrastructure, meals), 
then all children, boys and girls, will have a 
positive, healthy, quality learning experience and 
complete their education to be productive 
citizens. Based on this theory, McGovern‐Dole’s 
principal aims are: (i) to improve the literacy of 
school‐age children, (ii) to improve the quality of 
literacy instruction (iii) to improve student 
attendance by providing take‐home rations 
(THRs), (iv) to increase government capacity and 
ownership; (v) to improve the physical learning 
environment; and (vi) to improve health and 
dietary practices and raise awareness of the 
importance of education. 
This report presents the findings of an endline 
evaluation of the McGovern‐Dole project 
conducted by Advisem Services Inc. The 
objective of the evaluation was to assess 
whether the project had achieved its expected 
results and to measure the impact of its 
interventions, to assess project design, 
implementation, and management, and to 
provide lessons learned and recommendations 
for CRS, USDA, project participants and other 
key stakeholders for future food assistance and 
capacity‐building programs. 
The evaluation used a quasi‐experimental, mixed 
methods design with five lines of enquiry: a 
document and data review, surveys, key 
informant interviews (KIIs), focus group 
discussions (FGDs), and field observations. The 
approach was participatory, reviewing and 
assessing evidence from 21 participants in key 
informant interviews (KIIs), 217 participants in 
focus group discussions (FGDs), and 
2,849 participants in four surveys conducted in 
mostly rural, but also urban settings. Field 
observations were also made in schools to 
assess the use of hand‐washing stations, the 
physical environment, and student 
attentiveness. The evaluation sampled the same 
48 treatment schools and 49 control schools for 
the surveys and FGDs as were used at baseline, 
reaching the target of 48 schools in each 
category. Although qualitative and quantitative 
methods were used, the primary source of 
information for the evaluation was statistical 
evidence used to inform key indicators identified 
in the Terms of Reference (ToRs) for this 
evaluation. 
As described more fully in this report, this 
evaluation has faced several challenges. Key 
among them were a teacher’s strike that 
occurred during the project, which not only 
delayed and curtailed field work, but also had 
implications for learning and capacity results. 
Findings 
To facilitate comparisons with other McGovern‐
Dole programming, the endline evaluation 
assessed findings using standard USDA criteria 
under five categories: relevance, effectiveness, 
efficiency, sustainability, and impact. The 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
EXECUTIVE SUMMARY (CONTINUED) 
vi 
following are the key findings under each 
category. 
Relevance 
Strategies to improve children’s literacy –
McGovern‐Dole had a significant McGovern‐
Dole effect on whether students, by the end of 
two grades, could read and understand the 
meaning of grade‐level text. This strongly 
suggests that McGovern‐Dole is improving 
students’ ability to read and write French. At 
endline, students at treatment schools had 
significantly higher average scores than control 
students—a direct contrast to the baseline. This 
notable change suggests McGovern‐Dole 
interventions have had a positive impact on 
EGRA scores of children in treatment schools. 
This likely stems from increased teacher training 
after the mid‐term evaluation (MTE). However, 
some challenges remain, as teachers continue to 
struggle with basic good practices, addressing 
items in the syllabus, and having enough time to 
complete their daily activities. Some classrooms 
have had a high pupil‐teacher ratio, making one‐
on‐one reading and using picture booklets 
difficult. 
Strategies to improve enrollment and 
attendance – Targets for increased school 
enrollment have been met for boys and almost 
met for girls. However, the evaluators saw no 
observable decrease in the number of health‐
related absences, which was the other measure 
of the relevance of these strategies. Children in 
treatment schools were somewhat less likely 
than children in control schools to have had 
stomach ailments or diarrhea, which could be, in 
part, a result of improved hygiene, such as 
washing their hands at school. According to the 
parent survey, around 60 percent of children 
were sick in May 2018 due to malaria, signifying 
that, unfortunately, the project’s hygiene 
interventions can have little effect on the 
primary cause for health‐related absences. 
Strategies to improve community participation 
and engagement – The project’s primary 
strategy in this area has been to support parent‐
teacher associations (PTAs) or similar school 
governance structures. The percentage of active 
PTAs was already high at baseline, with PTAs at 
approximately 92 percent of the schools. 
McGovern‐Dole seems to have had a significant 
impact on the creation of school mothers’ 
associations (AMEs)—almost all treatment 
principals surveyed reported an AME at their 
school, and more treatment school than control 
school principals said the AMEs were “very 
active” and “moderately active” than did 
principals of control schools. AMEs and APEs 
(school fathers’ associations) have reportedly 
become more intrinsic to school functioning, 
supporting management of the canteens, 
providing extra food, mobilizing resources to pay 
for additional teachers, and encouraging 
absentee students to return to class. Almost half 
of savings and internal lending community (SILC) 
members reported spending their SILC income 
on things related to their children’s education. 
Stakeholder satisfaction – In general, the 
qualitative exercises revealed very positive 
responses by all project stakeholders. Students, 
families, teachers, principals and AMEs/APEs all 
expressed that they were very pleased with and 
appreciative of the project. 
Effectiveness 
Overall attainment of targets – Of 50 indicators 
studied, the project met its target for 25 and will 
very likely reach the target for six additional 
indicators at the end of this reporting period. 
Nine indicators that were close to being met and 
nine other indicators that were far from being 
met will unlikely be achieved. 
Results in improving the literacy of school‐age 
children – CRS had the ambitious goal of 
65 percent of boy students and 60 percent of 
girl students who, by the end of two grades, 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
EXECUTIVE SUMMARY (CONTINUED) 
vii 
could read and understand the meaning of 
grade‐level text. Only 13 percent of boys and 
12 percent of girls met the goal. Still, increased 
teacher training since the MTE seems to have 
had a positive effect on EGRA scores. Even 
though EGRA targets were not met, there was 
significant improvement. Other results related 
to literacy include: 
 Students in 793 classrooms were assessed 
using the EGRA tool, exceeding the target. 
 Just over 50 percent of teachers devoted at 
least an average of 45 minutes a day to 
literacy instruction, short of the 75 percent 
target. 
 CRS’ data suggests average teacher 
attendance decreased at endline, although 
principals reported attendance close to the 
95 percent target. McGovern‐Dole planned 
to reduce teacher absenteeism by 
constructing housing, which has not yet been 
completed; it is therefore unclear how the 
project at this time could have had any 
impact on this indicator. CRS is seeking a one‐
year extension to complete infrastructure 
development, including teacher housing. 
 The target of 95 percent of teachers using 
the national literacy curriculum and related 
instructional materials was met. 
 Targets for teacher training and increased 
skills and knowledge of teachers and school 
administrators have been met and will be 
greatly surpassed if CRS continues to 
implement training this semester. 
  McGovern‐Dole provided 1,034,518 
textbooks and other teaching/learning 
materials, surpassing its target. 
 The evaluation revealed little increase in 
attentiveness and, in fact, a slight decrease. 
However, the principal measure was hunger, 
and there are many other variables that can 
affect attentiveness. 
 The percentage of parents who said their 
children were “hungry” during the school day 
was well under the 20 percent target at 
7 percent. This is a significant improvement 
that suggests school canteens have affected 
this variable considerably. 
 The project surpassed its target of 95 percent 
of students consuming school meals daily. 
However, the number of meals provided per 
semester and over the project are well below 
target. 
 A total of 98,212 take‐home rations were 
provided, surpassing the target of 80,703. 
 In 2017, the project met its target of 
80 percent of male and female students 
regularly attending school and will likely meet 
it in 2018. McGovern‐Dole also aimed for a 2 
percent decrease in students reporting 
health‐related absences, but such absences 
at endline were more than 6 percent, up 
from 5 percent at baseline; this target would 
be difficult to meet because the baseline 
value was already very low. 
 More than 55 percent of parents in target 
communities could name at least three 
benefits of primary education, short of the 
60 percent target, but a significant 
improvement over 40 percent at baseline. 
Results increasing the use of health and dietary 
practices – Almost 74 percent of school‐age 
children were receiving a minimum acceptable 
diet at endline, up from 67 percent at baseline, 
but below the 85 percent target. Around 
8 percent of parents and 6 percent of students 
passed a test of good health and hygiene 
practice, below the targets of 40 and 50 percent, 
respectively. The project trained 20 females and 
105 males in child health and nutrition, greatly 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
EXECUTIVE SUMMARY (CONTINUED) 
viii 
surpassing targets. However, only around 
80 percent of food preparers have passed a test 
of safe food preparation and storage, short of 
the 100 percent target. All treatment schools 
have increased access to food preparation and 
storage tools and equipment. 
The project has also increased access to clean 
water and sanitation services. The target of 
87 schools using an improved water source is 
close to being met. CRS reports that 121 of the 
target 144 schools have adequate latrines, 
although principals and teachers reported fewer 
“functional” latrines, and many girls said they 
were dirty and unpleasant to use. At endline, 
almost 65 percent of schools had soap and 
water at the majority of hand‐washing stations 
and students were commonly using them, 
indicating the project played a significant role in 
increasing availability of the services and 
improving the hygiene of students. 
Overall, 28 nutrition or health initiatives or 
activities have been pursued in partnership 
between government and community groups, 
still below the target of 71. The document 
review identified six government staff involved 
in canteen and commodity management 
training, surpassing the target of four. 
Qualitative evidence – Evidence from FGDs 
confirms the project’s inputs are seen as 
mutually reinforcing and constructive. The 
teacher training and regular meetings, student 
reading groups, and reading materials were 
viewed very positively, as were school meals, 
infrastructure improvements and awareness‐
raising of parents. Capacity building, especially 
around the canteens, has driven mobilization of 
parent associations. Broadly, there was a sense 
that access and learning objectives are being 
achieved. Most qualitative evidence from 
parents, including those who did not send 
children to school and who were themselves 
illiterate, recognized and valued the importance 
of education. 
Efficiency 
Timely achievement of project objectives –
Approximately 70 percent of objectives were 
achieved or close to being achieved at endline. 
The project surpassed some targets and 
continues to provide services that could 
enhance results. Timeliness was not always 
consistent, however. For example, an eight‐
month delay prevented WEI from implementing 
planned training. Implementation of project 
activities and coordination between project 
partners were challenging at first but were 
successfully addressed over the course of the 
project. Some local government officials seem to 
expect further progress in coordinating and 
communicating project activities, increasing joint 
monitoring visits, and preventing duplication.  
Monitoring of project activities – Some of the 
indicators for the project may not be entirely 
appropriate, such as those for child absences 
due to sickness, attentiveness, and minimal 
acceptable diet. Similarly, some teacher‐level 
indicators should focus more on the CI and CP 
classes that were most involved in project 
activities. Some of the progress markers 
established by the project may also have been 
flawed.  
Actioning of recommendations made in the MTE 
report – The MTE report identified several 
shortcomings in the teacher training program. 
The evidence suggests that positive actions have 
been taken since then, with more focus on 
teacher education and improved practice.  
Sustainability 
Activities/outcomes most likely to be 
maintained – The second phase of McGovern‐
Dole is to be implemented in the same schools 
as in the first; therefore, many activities, 
including the canteens, will continue for now. 
The evaluation also considered the degree to 
which results associated with the various 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
EXECUTIVE SUMMARY (CONTINUED) 
   ix
components of the project will remain 
sustainable after the project has ended. These 
include: 
 Feeding programs – Principals and teachers 
generally had positive attitudes about the 
sustainability of the canteens, with teachers 
being slightly less positive. To CRS staff, it is 
clear that the current format for the school 
feeding program, which provides imported 
food items to school canteens, is not 
sustainable. CRS also procures food in local 
markets, which is a welcome development, 
but financial resources to buy this food are 
still required. Some parents believe school 
gardens and/or production fields could 
supply food to school canteens, but this 
would require testing. CRS’ assessment is 
that substituting procured food with food 
grown in school gardens is not feasible. Some 
schools have set up gardens, but these do 
not seem even close to being productive 
enough. In the evaluation team’s opinion 
schools do not generally have the land to 
grow food. Some individuals have also 
mentioned that produce has been stolen 
from the gardens and animals left to graze in 
them. 
 Learning – Overall, principals and teachers 
were generally positive about the ability to 
sustain improvements in learning compared 
to their attitudes about the canteens, but 
there were still many individuals who 
indicated the results would be little to not 
sustainable.  
 Health and hygiene – Principals and teachers 
indicated there is a desire to put into practice 
methods/systems introduced by the project, 
as well as the will and capacity to sustain 
health and hygiene results, and that children 
will maintain their good hygiene habits. 
Improvements such as hand‐washing stations 
and usage could be sustainable, although 
schools will need sources of water and ash, 
as soap may be too expensive. 
 Capacity – The sustainability of teachers’ 
capacity is difficult to assess without stronger 
indications that pedagogical learning has 
been consolidated at the individual teacher 
level and institutionalized in the relevant 
Ministry units. AMEs/APEs have increased 
their motivation, but the canteen has been a 
major factor. The improvement in EGRA 
scores, even if low overall, shows a positive 
trajectory and, presumably, will motivate 
teachers to continue to participate in training 
and sharing with colleagues.  
Contributing factors to sustainability –
Regression analysis of the EGRA surveys 
indicates that parental involvement could 
contribute positively to sustainability. FGDs with 
best performing students indicated the key role 
of family in academic performance. Principals 
and teachers often noted that for the results to 
be sustainable, government would need to help 
more. There are a series of sustainability 
reflections to be had at a more strategic level 
and a window of opportunity—particularly with 
the second phase of McGovern‐Dole 
beginning—to test a few partnership models 
with the government and other donors 
Impact 
Quantitative evidence on desired effects – The 
evaluation team carried out DID analyses on 
endline indicators with appropriate baseline and 
endline data. Several indicators, such as average 
teacher attendance rate, national curriculum 
usage, active APEs, and time spent on literacy 
instruction had high values at baseline and 
control schools also displayed results similar to 
treatment schools. In contrast, the McGovern‐
Dole project seems to have had a significant 
impact on the creation of AMEs. There is also 
some indication of ownership by the 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
EXECUTIVE SUMMARY (CONTINUED) 
x 
AMEs/APEs, as these groups have played a key 
role in implementing McGovern‐Dole.  
The DID analysis was significant for parents 
reporting whether their children were hungry 
during the school day. Overall, treatment school 
parents were much less likely than control 
parents to report their child being hungry. The 
effect of the McGovern‐Dole project on 
minimum acceptable diet was also close to being 
significant. Certainly, the school feeding 
program has had an impact, changing (at least 
for the period of the project) the food security 
of students in participating schools and 
influencing health and attendance. However, the 
project does not appear to have a statistically 
significant influence on attentiveness. The link 
between not being hungry, paying attention and 
learning should not be drawn too finely. While 
children having meals at school correlates 
positively with attentiveness, there were clearly 
other factors connected with being a McGovern‐
Dole school that could also be increasing 
students’ engagement.  
McGovern‐Dole had a significant effect on 
whether parents could name three or more 
benefits of primary education. It also had a 
significant effect on whether students, by the 
end of two grades, could read and understand 
the meaning of grade‐level text, illustrating that 
McGovern‐Dole is improving the ability of 
students to read and write French. Although the 
project has not achieved its anticipated results in 
improving literacy, students at target schools 
outperformed those at control schools and 
demonstrated improvement from the baseline. 
Factors beyond the project’s control may well 
have had negative impacts on these results.  
Gender is a major consideration related to 
project impact. Overall, it was difficult to draw 
strong conclusions about the project’s gender‐
differentiated impacts. The quantitative data 
provided some indication, but the surveys and 
FGDs did not shed much light on whether and in 
what ways girls and women may benefit 
differently than boys and men. It may be the 
case that stakeholders themselves are not clear. 
It will be important for future programming to 
track how girls are being affected, how they are 
being counted, and how equally they are 
participating in activities compared to boys. 
Ideas about achieving greater impact for girls 
and women in future programming are 
discussed in the report.  
Conclusions and lessons learned 
The McGovern‐Dole project was relatively 
successful in meeting or almost meeting its 
targets. Since the mid‐term, and as a result of 
adjustments made since then, there appears to 
have been an improvement in children’s literacy, 
even though the target was not met. Parents in 
the treatment area also seem to be much more 
aware of the benefits of education, likely as a 
result of McGovern‐Dole activities. Continued 
focus on raising parents’ awareness and 
increasing their involvement in their children’s 
learning could further enhance results. The 
AMEs/APEs are also energized and engaged in 
the project and are a critical factor in successful 
project implementation and bridging the 
relationship between the schools and 
communities. 
Infrastructure projects have also been better 
supported by technical staff since mid‐term. 
Infrastructure improvements and hand‐washing 
stations were well received by parents and 
community stakeholders and have likely had a 
moderate effect on children’s well‐being. The 
canteens also appear to have been successful in 
improving children’s nutrition and reducing their 
hunger; future efforts should focus on increasing 
the sustainability of canteen services. 
Feedback from all stakeholders was very 
positive. There were some suggestions for 
improvement and better coordination with 
partners, but the main comments were about 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
EXECUTIVE SUMMARY (CONTINUED) 
   xi
expanding the project. This is itself is a 
testimony to stakeholders’ satisfaction with the 
project. The SILCs are also very valued, although 
their contribution to children’s literacy and 
school attendance is indirect and complex to 
assess.  
Some of the less impressive results of the 
project could be due to the fact that several of 
the indicators may not have provided an 
adequate assessment of program achievement. 
There are also important factors that have 
affected the project, but that were beyond its 
control. Key among these are the high number 
of students enrolled in treatment schools, and 
the rotation of teachers.  
Recommendations 
The evaluation has identified some 
30 recommendations for future programming, 
around topics including improvements to 
collecting data and defining and using indicators; 
improving the internal consistency of the 
project’s design and the theory of change; use of 
the EGRA tool; securing effective government 
involvement; retaining and motivating teachers; 
improving prospects for sustainability, 
replicating best practices, and expanding the 
scope of the project; delivering and assessing 
teacher training, and eliminating physical 
discipline of students; and various 
improvements to hygiene programming, school 
meal programs and gardens, and infrastructure. 
More detail is provided in the body of the 
report. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
xii 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   1
1.  INTRODUCTION 
This report presents the findings of the endline evaluation of The McGovern‐Dole International Food for 
Education and Child Nutrition Program (hereafter referred to as McGovern‐Dole) in Benin, an initiative 
carried out by Catholic Relief Services (CRS) in partnership with World Education, Inc. (WEI), with funding 
support from the United States Department of Agriculture (USDA). The evaluation was performed by a 
team of international and national consultants from Advisem Services Inc., a consulting firm based in 
Canada.1
As indicated in the Terms of Reference (ToRs) issued in September 2017 (attached as Appendix 12
), the 
endline evaluation aims to assess whether the McGovern‐Dole project has achieved its expected results 
and to measure the impact of project interventions, while assessing areas of project design, 
implementation, management, lessons learned and replicability. The evaluation also provides lessons 
learned, and recommendations for CRS, USDA, project participants and other key stakeholders for future 
food assistance and capacity‐building programs. To facilitate comparisons with other McGovern‐Dole 
programs, the endline evaluation reports on findings using standard USDA criteria on the following 
dimensions: relevance, effectiveness, efficiency, sustainability, and impact. 
The remainder of this report is structured as follows: 
 Chapter 2 gives an overview description of the CRS McGovern‐Dole project in Benin, including 
the context in which the project is set, and provides a brief overview of the purpose of the 
endline evaluation. 
 Chapter 3 describes the methodology used to conduct the endline evaluation. 
 Chapter 4 discusses findings drawn for each of the five dimensions examined by the evaluation 
team. 
 Chapter 5 provides a general conclusion and outlines lessons drawn from the McGovern‐Dole 
project experience. 
 Chapter 6 lists a series of recommendations inspired by key findings of the study. 
The contents of this report are derived from a review and cross‐examination of information supplied by 
sources canvassed over the course of the endline evaluation. The opinions are strictly those of the 
evaluation team, as informed by a large body of evidence gathered from these various sources. 
For reasons outlined in this report, the endline evaluation of the McGovern‐Dole project has proven to be 
very challenging. Despite these difficulties, the evaluation team is grateful for advice and support 
provided throughout the evaluation by CRS, WEI, USDA and their dedicated staff who not only helped the 
consultants grasp the essence of the project but also organized field visit schedules, accommodations and 
                                                            
1
 The team consists of Louis‐Pierre Michaud (Team Leader), Anne Bernard (International Consultant), Julie Helson 
(International Consultant), Hussein Faruque Aly (International Consultant), Nicaise Satoguina (National Consultant), 
and Dr. Issaou Gado (National Consultant). 
2
 Due to their size, appendices to this report are supplied in a separate document. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
2 
transport. The team also wishes to acknowledge the contribution of students, teachers, parents, school 
staff, members of government and partners who agreed to take part in the evaluation. 
2.  INTRODUCTION 
2.1  Context 
As set out in the rationale CRS developed for the project’s 
theory of change (included with the ToRs for this assignment), 
Benin is one of the world’s poorest countries, with an annual 
per capita income of US$365, below the sub‐Saharan Africa 
average of US$470.3
 The United Nations Development 
Programme (UNDP) ranks it 159 out of 174 countries on the 
Human Development Index. Poverty is highest in northern 
Benin, and significantly above average (32‐36%) in all four of 
the north’s rural departments (Borgou, Alibori, Atacora, Donga;
see Image 1). 
Social services are lacking in the north, as exemplified by an 
insufficient number of primary schools. Although the four 
departments represent almost 75% of Benin’s land mass, fewer 
than half (48%) of the country’s primary schools4
 are located 
there, making schools very remote for many school‐aged 
children to attend. Local customs and beliefs place little value 
on education, and parents—most of whom are poorly educated 
themselves—are generally not involved in their children’s 
schooling. Moreover, poor nutrition—including a severe, 
annual hungry season—affects children’s ability to learn. 
Borgou and Alibori, particularly, receive little assistance from 
either the government or non‐governmental 
organizations (NGOs) for their school canteens. 
2.2  Project Description 
CRS Benin is implementing a four‐year, US$19,080,649 United 
States Department of Agriculture (USDA) funded McGovern‐Dole International Food for Education and 
Child Nutrition Program (hereafter referred to as McGovern‐Dole) to improve literacy and alleviate 
hunger for approximately 43,804 primary school‐age children attending 144 primary schools in four 
communes in the Alibori and Borgou Departments of Northeast Benin. 
                                                            
3
 The statistics provided in this section are taken directly from Appendix A of the ToRs. Sources are not provided for 
all these data, and they will undoubtedly have changed since CRS undertook research to support the project’s 
theory of change. For example, the latest UNDP Human Development Report (2016) ranks Benin 167 out of 188 
countries; it also assigns Benin a gross national income per capita of US$1,979 and a gross domestic product per 
capita of US$1,986 (both statistics 2011 at purchasing power parity dollars).  4
 Benin’s primary school system has three cycles featuring six types of introductory, preparatory, elementary and 
middle‐level classes labelled CI, CP, CE1, CE2, CM1 and CM2, respectively. 
Image 1 
Departments and Communes of Benin
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   3
The project, which started on September 30, 2014, and continues through September 30, 2018, 
coordinates activities with several education‐related ministries and departments, and works with 
212 teachers, 97 parent‐teacher associations (PTAs, or APEs in French) and 97 school mothers’ 
associations (AMEs, in French) to implement the project, including the distribution of 4,420 metric tons of 
commodities (soy‐fortified cornmeal, vegetable oil, lentils and milled rice). 
The main aims of the project are: (i) to improve the literacy of school‐age children, (ii) to improve the 
quality of literacy instruction by training teachers in early grade reading methodology, (iii) to improve 
student attendance by providing take‐home rations (THRs) for students, (iv) to increase government 
capacity and ownership by working with the Ministry of Preschool and Primary Education (MEMP, in 
French); (v) to improve the physical learning environment by building or rehabilitating kitchens, 
storerooms, latrines and water stations; and (vi) to improve health and dietary practices and raise 
awareness of the importance of education through radio broadcasts. 
The project’s theory of change (ToC) is grounded in a desire to contribute to a productive Beninese 
citizenry by making all girls and boys healthy and providing them with a quality and complete primary 
education. It posits that if the project promotes the importance of education to parents, improves 
teacher incentives and capacity, advocates for policy application, decentralization, and parent 
involvement, and provides inputs (school infrastructure, meals), then all children, boys and girls, will have 
a positive, healthy, quality learning experience and complete their education to be productive citizens. 
The project was implemented in partnership between World Education International (WEI) and Catholic 
Relief Services (CRS). In this partnership, CRS was the lead partner, and generally responsible for the 
management of the school feeding program (manage food items, distribution and storage in schools, 
food preparation and nutrition practices), the water and sanitation (WASH) activities, initially some of the 
infrastructure elements (water stations and systems, and latrines) and the formation of savings and 
lending groups (SILCs). WEI was responsible in general for the educational components of the project. 
These included capacity building, distribution of school supplies and materials, enrollment campaigns, 
training teachers, and training parent‐teacher associations. WEI was initially also responsible for 
building/rehabilitation of storage rooms and kitchens. After some challenges with the infrastructure 
elements of the project, CRS assumed the implementation of all infrastructure elements of the project. 
The McGovern‐Dole project was officially launched at the end of March 2015. CRS rehabilitated the main 
warehouse in Cotonou in April ‐ May 2015 whereas, in the field, construction/rehabilitation of storerooms 
actually started in the last quarter of FY15 due to delayed approval to begin activities. Between August 
and September, CRS setup of the Canteen Management Committees and identified cooks at all target 
schools. They were trained in food preparation and storage practices. Canteen activities began 
October 26, 2015 in some schools in the intervention area. The project has involved several governmental 
and non‐governmental stakeholders, including MEMP, the School Feeding Directorate (DAS, in French), 
the Food and Applied Nutrition Directorate (DANA, in French), the National Institute for Training and 
Research in Education (INFRE, in French), the Primary Education Directorate (DEP, in French), the 
Pedagogical Inspection Directorate (DIP, in French), and the Ministry of Health. 
Despite some challenges, including strikes5
 and impacts of political elections, overall, the project has 
operated in a favorable social and economic context with the support and involvement of local 
authorities and the community. The evaluator team’s interviews with the Ministry of Pre‐primary and 
                                                            
5
 The most disruptive strike occurred from January 15th, 2018 to April 25th, 2018. This was at the same time as the 
McGovern‐Dole final evaluation. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
4 
Primary Education (MEMP), and the Departmental Directorates of Pre‐primary and Primary Education 
(DDEMP), the regional pedagogical councils (CRPs), school directors and teachers show that there was a 
very receptive environment for the project. Similarly, the reinforcement of the commitment of the 
Government of Benin towards school feeding programs across the country show an overall support for 
the approach. For instance, for the 2016‐17 budget, the allocation for school feeding programs increased 
from 1 billion XOF (in 2015‐16) to 7 billion XOF (2016‐17), with the goal of reaching 51% of schools with a 
school feeding program by 2021 (probably to be reached by 2018‐19, with the assistance of partners, CRS 
included). There were some punctual challenges during the implementation of the project that affected 
the project’s performance, but overall it operated within a politically enabling environment. To realize its 
ToC, the project carried out a series of activities, led by either CRS or WEI, including: 
 The building/rehabilitation of kitchens, latrines, storerooms and water stations and systems; 
 Capacity‐building efforts at the local, regional or national level; 
 The creation of water sanitation and hygiene (WASH)‐friendly schools; 
 The distribution of school furniture and equipment or school supplies and materials; 
 Enrollment campaigns; 
 The establishment of school gardens;  
 The creation of savings and internal lending communities (SILCs); 
 Efforts to increase girls’ attendance at school; 
 The promotion of teacher attendance; 
 The provision of school meals; 
 Raising awareness of the importance of education; 
 The distribution of THRs; 
 Training in food preparation and storage practices or good health and nutrition practices, and 
training of PTAs and teachers; 
 Early Grade Reading Assessments (EGRAs). 
2.3  Evaluation Purpose 
The purpose of the endline evaluation was to assess the state of completion as compared to expected 
results and to measure the impact of McGovern‐Dole project interventions. It was also intended to 
identify strengths, challenges, and lessons learned and to capture the knowledge of the various 
stakeholders to leverage them for future projects on the same theme. It also aimed to look at whether 
the McGovern‐Dole project had exposed any new dynamics in learning, the quality of teaching, 
attendance and attention of students, the school environment and the importance of education. The 
lessons learned and recommendations of this study will promote the improvement of the quality and 
impact of future interventions. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   5
The endline evaluation was to assess the following specific indicators:  
 Percentage of students who, by the end of two grades of primary schooling, demonstrate that 
they can read and understand the meaning of grade‐level text (male); 
 Percentage of students who, by the end of two grades of primary schooling, demonstrate that 
they can read and understand the meaning of grade‐level text (female); 
 Percentage of students in target schools who are identified as attentive during class/instruction; 
 Percentage of parents in target schools who indicate that their children were “hungry” during 
the school day; 
 Percentage of students who report a decrease in health‐related absence; 
 Percentage of parents in target communities who can name at least three benefits of education; 
 Number of classrooms assessed using the Early Grade Literacy Assessment instrument; 
 Percentage of schools with soap6
 and water at a hand‐washing station commonly used by 
students; 
 Number of government staff in relevant ministries/offices implicated in canteen/commodity 
management training; 
 Number of teachers who devote at least 45 minutes a day to literacy instruction; 
 Number of teachers using the national literacy curriculum and the related instructional 
materials;  
 Number of school administrators and officials in target schools who demonstrate use of new 
techniques or tools as a result of USDA assistance;  
 Number of teachers/educators/teaching assistants in target schools who demonstrate use of 
new and quality teaching techniques or tools as a result of USDA assistance; 
 Percentage of school‐age children receiving a minimum acceptable diet. 
The endline evaluation was also to assess the level of achievement of McGovern‐Dole monitoring 
indicators, as evidenced by CRS’s monitoring data, to review the degree to which the project had 
achieved targets for all of the project indicators, and to identify the reasons behind any 
underachievement or overachievement. In addition, the endline evaluation aimed to measure the 
completion of the CRS management responses to recommendations from the mid‐term evaluation (MTE). 
The evaluation focused on the five criteria put forth in the “principles for evaluation of development 
assistance” endorsed by the Development Assistance Committee (DAC) of the Organization for Economic 
Cooperation and Development (OECD), namely relevance, effectiveness, efficiency, sustainability, and 
impact. It also identified key lessons learned in a cross‐cutting manner to capture systematic learning that 
would inform future strategic decision‐making for CRS, WEI, and USDA. The full set of evaluation 
                                                            
6
 The project also uses a locally‐accepted substitute (ash) for soap in schools. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
6 
questions, along with performance indicators, data and information sources, and data collection 
methods, is provided in Appendix 2. A list of definitions and terminology relevant to this study is found in 
Appendix 27. 
3.  METHODOLOGY 
3.1  Design Framework 
The endline evaluation adopted a quasi‐experimental, mixed methods design that focused on observing, 
measuring and describing project‐related phenomena as they occur or exist, rather than attempting to 
manipulate specific variables to “scientifically” demonstrate the existence of causal relationships between 
activities and outcomes. It used several lines of enquiry to triangulate data collected and provide solid 
findings and evidence to address the evaluation questions. It also drew some comparisons between the 
original project baseline, the MTE survey, and fresh quantitative and qualitative data gathered by the 
evaluation team. 
The evaluation also used a participatory approach, from identification of key learning questions to 
design of data collection tools, and validation of data analysis findings. This process granted key 
stakeholders—including CRS, WEI, USDA, local government bodies, community members and school 
support groups, and schools—an opportunity to provide their input at all stages of the evaluation 
process. 
3.2  Lines of Enquiry 
In line with the ToRs, the methodology featured five lines of enquiry: a document and data review, 
surveys, key informant interviews (KIIs), focus group discussions (FGDs), and field observations. 
Document and Data Review 
This line of enquiry consisted of a thorough examination of program documentation supplied by CRS 
relevant to McGovern‐Dole project activities in Benin. This evidence informed the development of the 
inception report and data collection tools, fed into the assessment of progress towards project objectives, 
and formed the basis for answering the key evaluation questions. Sources are listed in Appendix 3 and 
include, but are not limited to, the McGovern‐Dole agreement with USDA, the performance 
measurement framework (PMF), and semi‐annual log reports and accompanying narratives, as well as the 
baseline report and MTE report and supporting documentation (such as the work plan, databases, 
surveys, workshops, FGD guides, and EGRA).  
Surveys  
The evaluation team used four questionnaires/surveys, designed to gather mostly quantitative data, but 
also some qualitative data evidence from project beneficiaries helped and trained through the project to 
assess the success of: 
 Literacy training (for instance, percentage of students who, by the end of two grades of primary 
schooling, demonstrate that they can read and understand the meaning of the grade‐level 
text – males and females); 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   7
 School and THRs (for instance, decrease in hunger levels, better health and nutrition, and fewer 
days of school missed); 
 Improved health and dietary practices (for instance, better health and nutrition, and fewer days 
of school missed); 
 Increased awareness of the importance of education (for instance, fewer days of school 
missed). 
The four different survey tools used were an EGRA survey, a parent survey, a principal survey and a 
teacher survey. These are found in Appendices 4 through 7 respectively, in their original French version.7
An extensive description of the sampling strategy associated with the surveys is supplied below 
(Section 3.3). 
Key Informant Interviews 
KIIs were conducted by one member of the evaluation team, either remotely (via Skype) or face‐to‐face 
over the course of a field mission in Kandi, Parakou, Cotonou, and Porto Novo. As shown in Table 1, the 
consultant interviewed a total of 21 people outside of schools or in schools, including representatives of 
CRS, WEI and USDA, officials from government departments and regional authorities, representatives of 
local partners, and school directors. A full list of participants in the KIIs (except school Directors) is 
supplied in Appendix 8. Protocols used to conduct interviews with various categories of informants are 
featured in Appendices 9 through 11. 
During the field mission, the consultant who conducted the KIIs had to adapt to a number of unforeseen 
constraints. As a result, it was not always possible to conduct the KIIs as planned. For instance, a couple of 
interviews had to take place during the break of an activity that was taking place in Kandi, and as such 
were limited to approximately 20 minutes of conversation. Other interviews had to be cancelled, such as 
one that was planned with INFRE personnel who happened to be travelling at the time of the consultant’s 
visit. One interview with MEMP became a joint conversation with DAS, as the person selected by MEMP 
to speak with the consultant was also the head of DAS. This meeting turned out to be particularly 
productive, as other team members of DAS joined and contributed to the discussions. Finally, it was 
originally planned to conduct three interviews with CRS officials, but as there were relevant sources of 
information to be gathered by various staff members, the consultant ended up conducting a total of six 
interviews with CRS representatives, in Kandi and Cotonou. 
Focus Group Discussions 
FGDs were led by the same member of the evaluation team who conducted KIIs over the course of a field 
mission in Kandi, Parakou, Cotonou, and Porto Novo. As shown in Table 1, there were a total of 40 FGDs 
in schools, involving 217 participants (97 males, 120 females), including teachers, students, best 
performing students, parents, and members of PTAs and other networks. Guides developed to lead the 
discussions are featured in Appendices 12 through 17, respectively. 
                                                            
7
 All four surveys were conducted in French. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
8 
Table 1 
Number of Participants in Key Informant Interviews and Focus Group Discussions 
Organization/type of stakeholders involved  Number of 
locations 
Number of 
KIIs or FGDs 
Number of participants 
Male  Female  Total 
KIIs outside of schools (see Appendix 8) 
WEI officials (Parakou) [1]  ‐ 2 ‐  ‐ 3
USDA officials (US) [2]  ‐ 1 ‐  ‐ 2
MPEM/DAS  ‐ 1 ‐  ‐ 1
Heads of school districts (Kandi)  ‐ 1 ‐  ‐ 1
Regional circle officials (Kandi, Malanville, Gogonou)  ‐ 1 ‐  ‐ 3
Local partners (UNICEF)  ‐ 1 ‐  ‐ 1
CRS staff (Headquarters, Cotonou, Kandi)  ‐ 6 ‐  ‐ 6
Total  ‐ 13 ‐  ‐ 17
KIIs in schools 
School directors  ‐ 4 3  1 4
FGDs in schools 
Students [3]  4 8 35  41 76
Teachers [3]  4 8 13  8 21
Parents of students [3]  4 8 15  29 44
Members of PTAs and other networks [3]  4 8 15  28 43
Parents of children not attending school  4 4 4  1 5
Best performing students [4]  4 4 15  13 28
Total  ‐ 40 97  120 217
Notes: 
1. One interview was conducted over skype, whereas the other was a joint interview with two participants in Kandi. 
2. Interview conducted at the early stages of the evaluation 
3. Separate FGDs were held with male and female participants. 
4. Observations were also made for best performing students. 
Despite challenges faced in mobilizing participants—particularly parents and members of networks—it 
was possible to conduct all FGDs as planned and to derive satisfactory information from them. 
Specifically, the plan was to set up gender‐disaggregated FGDs with students, teachers, parents of 
students, network representatives, and parents of children not attending school. With the exception of 
the latter category of participants, all FGDs were successfully organized in gender‐segregated groups.  
The mobilization of the families of students who did not attend school was a slightly more complex 
process. Schools and AMEs repeatedly advised the consultant that there were no children who did not 
attend schools, and our first attempts to speak to these families were frustrated. The teachers, on the 
other hand, said that there were students who did not attend school. The evaluators understood this 
discrepancy of perception from the directors and AMEs and the teachers when they were conducting the 
FGDs in the second school (the evaluators returned to the first school to interview families of children 
who did not attend school on day 2). In order to identify the families of students who did not attend 
classes, the consultant had to visit the villages with teachers or representatives of AMEs to identify the 
families of students who did not attend school. This was a challenging process, requiring consultants to go 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   9
from home to home, and only a limited number of adults were identified with whom the consultant could 
speak. One constraint to holding disaggregated exercises was that it would have been difficult to ask 
males to wait while FGDs took place with females, and vice‐versa. As the number of participants was low, 
it was decided to conduct FGDs with parents of children who did not attend schools in a single setting 
(that is, males and females together), and ensure active facilitation would allow all participants to 
contribute evenly. 
Field Observations 
The evaluation team conducted some field observations, including: (1) observing if water and soap (or the 
locally‐accepted substitute ash) were available at hand‐washing stations and whether hand‐washing 
stations were properly used; (2) a general observation of the physical school environment (for 
instance, the state of kitchens, storerooms, latrines, drinking water and school gardens); and (3) the 
attentiveness of students in the classroom. Tools developed to conduct these observations are featured 
in Appendices 18 through 20, respectively.  
Observations of hand‐washing stations and the physical school environment were made by enumerators 
at each of the schools visited. Attentiveness of students during class/instruction was assessed in two 
randomly‐selected classrooms (third to sixth grade, as in the baseline) at each school visited. Within each 
classroom, 10 randomly selected students (selected previously through “points” in the classroom) were 
deemed as attentive or inattentive at the time of observation, following a method similar to that used in 
the baseline. Girls and boys were differentiated in the recordings made, noting that an equal number of 
girls and boys should be observed (treatment school population of CE1, CE2, CM1, CM2 is 47.8% girls and 
52.2% boys, meaning observing 5 girls and 5 boys was appropriate; no detailed control school populations 
were provided before field work began). 
Collection of Survey and Observational Data  
Surveys and observational tools were kept to a maximum length of approximately 30 to 45 minutes and 
primarily consisted of structured questions with pre‐defined answers (with the option of an “other” 
response in relevant cases). Some open‐ended questions were included in the surveys to obtain the 
opinions of the respondents. Surveys were based heavily on those used in the baseline and mid‐term to 
ensure the comparability of results; they also included some specific questions to answer the endline 
evaluation questions from the ToRs.  
Survey questionnaires and observational tools were uploaded to smartphones, making use of some of the 
equipment that CRS had available. The software utilized was Online Data Kit (ODK) and Ona for all 
surveys/tools, with the exception of EGRA, for which the software Tangerine was used. Due to the 
functionality of Tangerine, larger‐sized tablets were required. Electronic surveys increased both the speed 
and accuracy of data collection and assisted with quality control. 
3.3  Sampling Strategy 
The evaluation team drew the sample for all the data collection methods from a database of treatment 
and control schools and associated participants provided by CRS Benin. Based on this database, and 
considering project information, the overall criteria applicable to the final evaluation of McGovern‐Dole in 
Benin include: 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
10 
 Geographic location – All of the CRS intervention schools occur in Northeast Benin, specifically 
in Alibori Department and its communes of Gogounou, Malanville and Kandi and in Borgou 
Department and its commune Kalalé. All of the CRS control schools occur in Northeast Benin, 
specifically in Alibori Department and its commune of Banikoara and in Borgou Department and 
its communes Kalalé and Bembereke. The sampling ensured that schools were selected in each 
of the communes. 
 Treatment versus control – The sample provided adequate control schools for comparison. CRS 
provided the global population of control schools, which CRS confirmed were not receiving 
government assistance or international assistance from any other organization, especially in the 
form of canteen services. Control schools were specifically used for the surveys and 
observational tools. 
 Categories of beneficiaries – The sample provided adequate representation of beneficiaries and 
stakeholders influenced through the project, including students, parents, teachers, principals, 
PTA members, government representatives, as well as key CRS, WEI and USDA personnel. These 
groups of individuals were reached through a combination of quantitative and qualitative 
methods. 
The specific methods for determining the sample for each of the quantitative and qualitative methods are 
discussed below. 
The first step for determining the sample for all of the surveys and observational tools relied on the 
selection of treatment and control schools to be visited. With the number of intervention schools at 143, 
the aim was to visit 50 treatment schools and 50 control schools, as the baseline suggested visiting 
48 treatment schools and 48 control schools. Thus, the evaluation team sampled the same number of 
schools that had been sampled in the baseline and mid‐term. Exact numbers sampled for each of the 
surveys and observational tools are discussed below. 
For selection purposes, control schools should not have received any similar intervention and treatment 
school should have only received the intervention of CRS/WEI. The same control and treatment schools 
as used in the baseline were used for the endline evaluation. This was necessary to undertake difference‐
in‐difference (DID) analyses used to assess the impact of project interventions.8
 The evaluation team 
understood that the control schools had not received any interventions.9
 It was confirmed during the 
fieldwork that none of the control schools had canteens. The evaluation team also recognized that it was 
                                                            
8
 If treatment schools were randomly selected, the evaluation would first be deviating from the assumption of 
parallel trend necessary for DID. It would also need to apply econometric patches (such as inverse probability 
weighting (IPTW)), which would become an extremely complex exercise. Also, econometrics will not be able to 
completely control for the differences between the baseline and endline treatment samples (because these are 
different), and thus we would not know if the overall effect is due to the difference in treatment schools or because 
of the treatment. The IPTW or other econometric approaches could control for some of the differences between the 
baseline and endline treatment samples, but the likelihood of unobserved factors (meaning factors that we cannot 
adjust for) between the two samples is great. Thus, even with econometric adjustment it would still be difficult to 
determine whether the effect measured between the baseline and endline treatment samples is due to the 
treatment. Thus, the potential biases introduced by selecting a different endline treatment sample are far more 
problematic than the any extra effort CRS could have put on the baseline treatment sample.  9
 The evaluation team assumed that CRS contacted all control schools to confirm this, as it did not have the contact 
information to validate this on its own. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   11
unlikely that CRS could have placed extra effort on these 50 treatment schools. Therefore, the findings of 
the endline evaluation would represent CRS’ impact at its best (for instance, this was not a double‐blind 
experiment). 
Within the beneficiary schools, there are 79 rural schools (55%) (20,180 students; 45.2%) and 64 urban 
schools (45%) (24,496 students; 54.8%). The evaluation team adhered closely to the past sampling regime 
(7 urban schools and 43 rural schools) to maintain consistency necessary for DID, resulting in 8 urban 
schools and 42 rural schools.10 Overall, in Alibori Department there are 30 intervention schools in the 
commune of Gogounou, 32 in Malanville and 67 in Kandi, and in Borgou Department and its commune 
Kalalé there are 14 intervention schools (see complete list in Appendix 21). For the control schools, a list 
was provided to the evaluation team that included a total of 104 schools in rural areas and 9 in urban 
areas. Additionally, there are 51 schools in Alibori Department and its commune of Banikoara, and 62 
schools in Borgou Department and its communes of Kalalé (28 schools) and Bembereke (34 schools; see 
Appendix 21). 
Ultimately, the endline evaluation used the same population of treatment and control schools as was 
used in the baseline study. Table 2 shows the overall population of treatment and controls schools in 
each department and commune, the number of treatment schools and control schools that were used in 
the baseline and were again selected to be used in the final evaluation, and the final distribution of 
schools that were sampled. One control school was removed from the list, as in March 2018 the 
evaluation team was informed that it now has a canteen. Three control schools that were previously one 
school during the baseline have now split into two schools. A replacement control school was selected 
from these schools. Selection between the other split schools was done using a random number 
generator (RNG) application. 
High standards of data quality were ensured by piloting the tools (this occurred once in February 2018 for 
all tools, with the exception of EGRA, and again in June 2018, with all tools including EGRA), making any 
necessary adjustments to the tools, training the data collectors, and monitoring the implementation of 
data collection (for instance, conducting spot checks to observe the enumerators).  
Local enumerators were hired from those in the CRS database. The evaluation team trained these 
enumerators on the questionnaires themselves and how to input the data and provided an opportunity 
for them to practice until they were comfortable with the process. The training had been programmed to 
extend over a period of four days; however, this was reduced to three days because of a shortened 
window for data collection. For the EGRA training, an EGRA expert was brought in on the evaluation 
team. 
The evaluation team expected that the enumerators hired through the CRS database would have 
previous experience conducting surveys and that many would have experience implementing EGRA. 
Unfortunately, a small number of the enumerators hired did not attend the first or sometimes even the 
second day of training. However, the consultants did make time to try to bring these individuals up to 
speed. 
                                                            
10 Therefore, an appropriate stratified sample would have had very similar populations. However, it appears that at 
the beginning of the project there was a limited number of urban control schools. Therefore, for both the baseline 
and mid‐term only 7 urban schools were assessed and 43 rural schools. The evaluation team adhered closely to the 
past sampling regime to maintain consistency but notes that the sample is very biased towards rural schools. 
Furthermore, the number of urban control schools available is still very limited. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
12 
Table 2 
Total Number of Schools, Schools Selected to be Sampled, and Schools actually Sampled, by 
Treatment/Control and Department and Commune 
Department  Commune 
Alibori  Borgou  Gogounou  Malanville  Kandi  Banikoara  Kalalé  Bembereke 
Treatment  
All schools  129  14 30 32 67 ‐  14 ‐
Selected schools  44  6 11 13 20 ‐  6 ‐
Sampled schools  42  6 9 13 20 ‐  6 ‐
Control 
All schools  51  62 ‐ ‐ ‐ 51  28 34
Selected schools  20  30 ‐ ‐ ‐ 20  9 21
Sampled schools  20  30 ‐ ‐ ‐ 19  9 21
Total to be sampled in 
each Department or 
Commune 
64  36 11 13 20 20  15 21
Total actually sampled 
in each Department or 
Commune 
62  36 9 13 20 19  15 21
Field work had been programmed to occur over a period of 10 days, using 40 local enumerators11 and 
four supervisors for the quantitative data gathering, as well as three facilitators, three notetakers, and 
one supervisor for qualitative data gathering. However, again because of a shortened window for data 
collection, this period was reduced to six days, and a total of 68 enumerators and 10 supervisors were 
hired. The shortened fieldwork period occurred because the consultants deemed it appropriate to wait 
for one month after the strike ended to start data collection; logistical and contractual considerations; 
and because the evaluation team was informed that “white exams”12 would begin June 18th making it 
difficult, and perhaps impossible, to work in the schools. Each day, four enumerators completed all 
surveys, including EGRA, and conducted observations at the school they visited. The qualitative data 
gathering in the field took place over a ten‐day period: four days were used for data collection in schools; 
two days were used for transcription and data cleaning; and the remaining KIIs took an additional four 
days. 
In the end, unfortunately, two treatment schools in Alibori, Gogounou (EPP ALAFIAROU and EPP 
GOUNAROU/B) were not sampled due to miscommunication. However, 48 treatment schools and 
49 control schools were sampled for the surveys and field observations, thus making the goal of sampling 
48 schools in each category (see Table 3). In addition, four schools were randomly selected for qualitative 
data collection activities (see Table 4). 
                                                            
11 Because of the overall number of surveys (4) and observational tools (3) to be implemented in each school, the 
evaluation team had required four enumerators per school. CRS’ EGRA also has a long demographic section that 
children need to answer before moving onto the EGRA modules. 
12 These are essentially practice exams. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   13
Table 3 
Schools Selected for Survey and Observation Tool Sampling 
School  Treatment or 
control 
Rural or 
urban 
Total population 
of students 
School  Treatment or 
control 
Rural or 
urban 
Total population 
of students 
Alibori, Gogounou 
Alafiarou (not sampled)  Treatment  Rural  139 BODJECALI/B  Treatment  Rural  229
Gounarou/B (not sampled)  Treatment  Urban  308 BODJECALI/C  Treatment  Rural  193
Lafiarou  Treatment  Rural  144 GAROU/A  Treatment  Rural  455
Ouèrè/A  Treatment  Rural  277 GOUN‐GOUN Treatment  Rural  402
Ouèrè/B  Treatment  Rural  228 GUENE QUARTIER  Treatment  Rural  340
Soukarou/A  Treatment  Rural  264 GUENE/A  Treatment  Rural  369
Soukarou/B  Treatment  Rural  235 GUENE/B  Treatment  Rural  307
Wara/A  Treatment  Rural  377 MADECALI Treatment  Rural  757
Wara/B  Treatment  Rural  314 MONKASSA  Treatment  Rural  461
Zougou‐P/A  Treatment  Rural  389 MVILLE QUARTIER/A  Treatment  Urban  527
BODJECALI/A  Treatment  Rural  454 TASSI‐TEDJI/C  Treatment  Urban  789
Alibori, Kandi 
WOLLO  Treatment  Rural  331 PODO‐LAFIAROU  Treatment  Rural  257
ABOUKOTO  Treatment  Rural  127 SAKATOUSSA  Treatment  Rural  100
ALFAKOARA  Treatment  Rural  428 SAM A (split into 
A & B [note]) 
Treatment  Urban  373
DEOU  Treatment  Rural  98 SAM B Treatment  Urban  166
DONWARI‐A  Treatment  Urban  316 SINAWONGOUROU/A Treatment  Rural  233
KAMA  Treatment  Rural  118 SINAWONGOUROU/B Treatment  Rural  112
MADINA / B  Treatment  Urban  363 TANKONGOU  Treatment  Rural  385
MONGO  Treatment  Rural  190 THUY Treatment  Rural  363
PADE  Treatment  Rural  434 THYA / A  Treatment  Rural  364
PEDE‐A  Treatment  Rural  257 TISSAROU  Treatment  Rural  295
PEDE‐B  Treatment  Rural  277 YEGOUROU  Treatment  Rural  148
Alibori, Banikoara 
EPP ANGARADEBOU  Control  Rural  55 EPP MONDARA  Control  Rural  85
EPP BADAGOU  Control  Rural  111 EPP MOUROUSSE  Control  Rural  68
EPP BONHANROU/A  Control  Urban  221 EPP PARIS  Control  Rural  70
EPP BONHANROU/B  Control  Urban  138 EPP SIRIKOU  Control  Rural  182
EPP DOMBOURE‐PEULH  Control  Rural  87 EPP SOMBI‐KEREGOU Control Rural  116
EPP DONDANGOU  Control  Rural  90 EPP SONWARI  Control  Rural  161
EPP GABE‐GBEROU  Control  Rural  105 EPP SOUMABOUGOU  Control  Rural  177
EPP KOGNINKOU  Control  Rural  128 EPP TISSOUA  Control  Rural  77
EPP KPESSANROU  Control  Urban  106 EPP TOUMAROU Control Rural  138
EPP MAMA GBÉROU  Control  Rural  136  
Borgou, Bembereke 
EPP BEREKE/B  Control  Urban  272 EPP MANI  Control  Rural  114
EPP BEREKE‐SUD A  Control  Urban  267 EPP NASSA Control Rural  283
EPP BEREKE‐SUD B  Control  Urban  206 EPP PEDAROU/A  Control  Rural  270
EPP DONWARI  Control  Rural  41 EPP PEDAROU/B  Control  Rural  256
EPP GAMARE Control  Rural 195 EPP SANSE  Control  Rural  56
EPP GANRO  Control  Rural  294 EPP SAORE Control Rural  411
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
14 
Table 3 
Schools Selected for Survey and Observation Tool Sampling 
School  Treatment or 
control 
Rural or 
urban 
Total population 
of students 
School  Treatment or 
control 
Rural or 
urban 
Total population 
of students 
EPP GUERA N’KALI  Control  Rural  348 EPP SOMBOUAN 1 Control Rural  235
EPP GUERE  Control  Urban  314 EPP SOMBOUAN 2  Control  Rural  164
EPP KENEKOU  Control  Rural  155 EPP SONGOURA  Control  Rural  358
EPP KPREGOUROU B  Control  Rural  193 EPP WANRAROU/A  Control  Rural  319
EPP KEROUKPOGOH (EX 
WARANKEROU) 
Control  Rural  255 EPP PEDAROU 
GANDO 
Control Rural  34
Borgou, Kalale 
DERASSI  Treatment  Rural  740 EPP DANTCHA  Control  Rural  111
KIDAROUKPEROU/A  Treatment  Rural  342 EPP DIAMBADI Control Rural  94
KIDAROUKPEROU/B  Treatment  Rural  206 EPP DUNKASSA/A  Control  Rural  255
NASSICONZI/A  Treatment  Rural  224 EPP DUNKASSA/B  Control  Rural  278
NASSICONZI/B  Treatment  Rural  489 EPP KOPRE  Control  Rural  127
PEONGA  Treatment  Rural  735 EPP LOU Control Rural  253
EPP BESSASSI/A  Control  Rural  280 EPP SONKOROU  Control  Rural  75
EPP BESSASSI/B  Control  Rural  264  
Note: This school has been split into two different schools, both urban. 
  
Table 4 
Randomly Selected Schools for Qualitative Data Collection Activities 
Commune  School  Rural or urban  Total number of 
students 
Borgou, Kalale  KIDAROUKPEROU/A   Rural  342 
Alibori, Mallenvile  TASSI‐TEDJI/C   Urban  789 
Alibori, Kandi  DONWARI‐A   Urban  316 
Alibori, Gogonou  Soukarou/A   Rural  264 
General sample information is provided below for all the survey and observational tools. Details on any 
missing surveys can be found in Appendix 22. 
EGRA Tool  
During the baseline and for the data used in the MTE, students in both first grade and second grade were 
sampled. However, because the indicator is specifically “by the end of two grades of primary schooling, 
demonstrate that they can read and understand the meaning of grade‐level text,” the evaluation team 
only sampled students in second grade. This is especially pertinent considering students may have missed 
schooling due to a teacher strike that occurred in Benin, which greatly influenced the conduct of this 
evaluation (see Section 3.5).  
Students were to be selected randomly using school rosters and the RNG application. This was to ensure 
that the students selected to complete EGRA were representative of the second‐grade student 
population within the schools visited. Since these schools were also selected randomly, it can be inferred 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   15
that the EGRA scores are representative of the second‐grade populations in all the intervention and 
comparison schools. 
During the first three days of field work it appears that enumerators used stratified sample selection and 
several different methods for selecting the first student that did not include RNG (details are included in 
Section 3.5). Starting the fourth day of data collection, RNG and school rosters were solely used for 
selecting students. Regression analyses were done post‐field work to assess if any potential bias resulted 
from this; however, results indicated that any bias was unlikely (see Appendix 23).  
At each school, the aim was to test 16 students. Generally, this goal was achieved (see Table 5), but in six 
schools (5 control and 1 treatment) there was no CP class13 and in some other schools fewer than 
16 students were present in a classroom. Also, the aim was to sample an equal gender ratio, which was 
closely achieved by sampling 49.0% females and 51.0% males (see Table 6). Like the baseline, the 
consultants made use of the program Tangerine, thus undertaking EGRA data collection completely 
electronically. 
Table 5 
Programmed and Performed Sampling for EGRA Test 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  672   128 800 602  114  716 
Control  656  128 784 540 132  672 
Total  1,328  256 1,584 1,142 246  1,388 
Table 6 
Distribution of Students Completing the EGRA Test by Gender 
Gender  Treatment  Control  All Schools 
Number  %  Number  %  Number  % 
Male  365  51.0 317 47.2 682  49.1 
Female  351  49.0 355 52.8 706  50.9 
Total  716  100.0 672 100.0 1388  100.0 
 Chi‐square 1.85, p>0.05, not significant.
Parent Survey 
For the baseline and mid‐term, 5 to 10 parents were randomly selected using the school rosters. This 
ensured that the sample of parents interviewed in each school was representative of the parents of 
enrolled students in each community. Supervisors were asked to do sample selection using RNG and 
students needed to contact their parents; this was to be done before enumerators visited the school. 
Generally, 15 to 20 parents at each school were invited to complete the survey (stratification was done by 
classroom). During the first three days of field work it appears that some convenience‐based sampling for 
                                                            
13 In three schools this was recorded as no consent. Overall, three students appeared to not provide consent to do 
the test and these students were removed from the dataset. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
16 
parents occurred (details are included in Section 3.5) and RNG was not used for sample selection. 
Regression analyses were done post‐field work to assess if any potential bias results from this; however, 
results indicated that any bias was unlikely (see Appendix 23). Starting the fourth day of data collection 
RNG and school rosters were solely used for selecting students who would then ask their parents to come 
for the survey. 
The endline evaluation attempted to sample 10 parents per school and in 93 of the 97 sampled schools 
(nine parents sampled in three schools and seven in one school) was able achieve this goal. Also, in some 
schools more than 10 parents were sampled. Enumerators had been instructed to thank the surplus and 
encourage them to keep following up on their children’s education, but in a few cases, it appears that 
more than 10 parents were surveyed.14 Table 7 below provides the programmed and performed sampling 
numbers. In treatment schools, slightly more females were surveyed than males; whereas, in control 
schools more males were surveyed than females (see Table 8). 
Table 7 
Programmed and Performed Sampling for Parent Survey 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  420  80 500 459 69  528 
Control  410  80 490 422 80  502 
Total  830  160 990 881 149  1,030 
Table 8 
Distribution of Respondents for the Parent Survey by Gender 
Gender  Treatment  Control 
Number  %  Number  % 
Male  237 44.9 347 69.1 
Female  291 55.1 155 29.3 
Total  528 100.0 502 100.0 
Chi‐square 60.59, p<0.05, significant; more female respondents were also 
sampled in the treatment area in the baseline.
Generally, parent respondents had no education (treatment: 61.5%, control schools: 73.3%), or had 
completed primary education (treatment: 17.4%, control schools: 9.56%) or the first cycle of secondary 
education (treatment: 11.2%, control schools: 8.2%), with much lower percentages reporting completing 
any other level of education. Overall, respondents in both treatment and control schools most commonly 
reported working in agriculture (treatment: 51.1%, control schools: 74.1%) and in the household 
(treatment: 26.7%, control schools: 11.0%). 
                                                            
14 The evaluation team feels it is valuable to maintain these extra data points. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   17
Table 9 below provides the average family size, the average number of children per family, and the 
average number of children in the sampled school for parents completing the survey across the various 
strata, with treatment schools tending to have slightly lower averages than control schools. 
Table 9 
Average Family Size, Average Number of Children, and Average Number of Children in the Sampled 
School for Parents Completing the Survey 
Classroom  Average family 
size 
Average number 
of children 
Average number 
of children in the 
sampled school 
Treatment, rural  11.4 6.1 1.7 
Treatment, urban  11.2 5.7 2.0 
Control, rural  14.7 9.1 1.9 
Control, urban  10.9 6.4 1.7 
Treatment  11.4 6.1 1.7 
Control  14.1 8.6 1.9 
Total  12.7 7.3 1.8 
Principal Survey  
The evaluation was planned to survey each principal at each school, so no selection methodology was 
implemented. Table 10 below provides the programmed and performed sampling numbers. 
Table 10 
Programmed and Performed Sampling for Principal Survey 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  42  8 50 40 7  47 
Control  41  8 49 41 8  49 
Total  83  16 99 81 15  96 
Respondents to the principal survey could either be the director of the school or the deputy director. In 
the treatment schools, 78.7% of respondents were the director and 21.3% the deputy director, while in 
the control schools, 95.9% of respondents were the director and 4.1% the deputy director.15 Overall, 
most respondents for the principal survey were males, with 83.0% being males in treatment schools and 
98.0% being males in control schools (see Table 11). 
                                                            
15 From this point onwards in the report, respondents to the principal survey will simply be referred to as 
“principals.” 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
18 
Table 11 
Distribution of Respondents for the Principal Survey by Gender 
Gender  Treatment  Control 
Number  %  Number  % 
Male  39 83.0 48 98.0 
Female  8 17.0 1 2.0 
Total  47 100.0 49 100.0 
Chi‐square 4.69, p<0.05, significant
Teacher Survey  
If possible, all teachers (5‐6 if present; principals may teach one of the classes; one teacher may teach 
several classes) were surveyed at each school, so no selection methodology was implemented. Table 12, 
below, illustrates the programmed and performed sampling. Fewer teachers (335) were sampled than 
programmed (assuming an absolute maximum), but in many cases this was due to the absence of 
teachers at a school or because fewer than six classes were taught. Four teachers did not provide 
consent.16 Teachers surveyed were much more commonly males than females, with 72.1% being males 
from treatment schools and 72.4% being males from control schools (see Table 13). 
Table 12 
Programmed and Performed Sampling for Teacher Survey 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  210‐252  40‐48 250‐300 150 29  179 
Control  205‐246  40‐48 245‐294 124 32  156 
Total  415‐498  80‐96 495‐594 274 61  335 
Table 13 
Distribution of Respondents for the Teacher Survey by Gender 
Gender  Treatment  Control 
Number  %  Number  % 
Male  129 72.1 113 72.4 
Female  50 27.9 43 27.6 
Total  179 100.0 156 100.0 
The percentages of teachers surveyed by class are quite similar between treatment and control schools 
(see Table 14). Please note that the number of classrooms sampled is greater than the number of 
teachers sampled, because sometimes teachers taught two classes and, in one case, four classes. Also, 
the average number of years teachers reported working at their schools was 2.66, with treatment school 
                                                            
16 These are not included in the 335 total. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   19
teachers having worked slightly less time (2.53 years) compared to control school teachers (2.81 years). 
Also, of note is that the distribution of the highest level of education achieved is quite similar between 
treatment and control schools (see Table 15), which could be an important factor related to teaching 
quality. 
Table 14 
Classrooms Taught by Teachers Sampled for the Teacher Surveys 
Classroom  Treatment  Control 
Number  %  Number  % 
CI  41 21.5 39 22.9
CP  43 22.5 37 21.8
CE1  34 17.8 36 21.2
CE2  40 20.9 27 15.9
CM1  28 14.7 27 15.9
CM2  5 2.6 4 2.4
Total  191 100.0 170 100.0
Table 15 
Highest Level of Education Achieved by Teachers Sampled for the Teacher Surveys 
Classroom  Treatment  Control 
Number  %  Number  % 
6e – Secondary School Level 1  2 1.1 1 0.6 
5e – Secondary School Level 1  1 0.6 0 0.0 
4e – Secondary School Level 1  1 0.6 1 0.6 
3e – Secondary School Level 1  30 16.8 15 9.6 
2e – Secondary School Level 2  15 8.4 24 15.4 
1ère – Secondary School Level 2  28 15.6 36 23.1 
Terminale Secondary School Level 2  78 43.6 56 35.9 
University  22 12.3 23 14.7 
Other  2 1.1 0 0.0 
Total  179 100.0 156 100.0 
Overall, the average number of years teachers who reported working at their schools was 2.66, with 
treatments school teachers having worked slightly less time (2.53 years) compared to control school 
teachers (2.81 years). 
Overall, teachers most commonly reported themselves to be government contractors (ACE, in French) or 
communal teachers, regardless of whether they taught in treatment or control schools. Treatment 
schools had a higher percentage of intern or volunteer teachers, while control schools had a higher 
percentage of ACE teachers (see Table 16). 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
20 
Table 16 
Status of Teachers who Undertook the Teacher Survey, by Treatment and Control Schools 
Classroom  Treatment  Control 
Number  %  Number  % 
Government contractor  23 12.8 12 7.7* 
Permanent government worker  60 33.5 70 44.9** 
Unionized worker  1 0.6 0 0.0 
Communal worker  52 29.1 52 33.3** 
Intern or volunteer  43 24.0 21 13.5 
Other  0 0.0 1 0.6 
Total  179 100.0 156 100.0 
*p<0.1, **p<0.05 
Hand‐Washing Station Observation Tool 
This tool was programmed to be implemented at each school surveyed, so no selection methodology was 
employed (see Table 17 for details). Enumerators made observations of the use of hand‐washing stations 
throughout their day at the school and visited each station to determine if they had water and soap/ash 
and evaluate the usage of each station. 
Table 17 
Programmed and Performed Sampling for Hand‐Washing Observation Tool 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  42  8 50 41 7  48 
Control  41  8 49 41 8  49 
Total  83  16 99 82 15  97 
Physical School Environment Observation Tool 
This tool was implemented at each school surveyed, so no selection methodology was implemented (see 
Table 18 for details). Ideally, the enumerator tasked with undertaking the principal survey also undertook 
a tour of the grounds (with the school principal) and completed this survey, examining the presence and 
state of canteens, storerooms, latrines, water systems and gardens. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   21
Table 18 
Programmed and Performed Sampling for School Environment Observation Tool 
Type of school  Number programmed  Number done 
Rural  Urban  Total  Rural  Urban  Total 
Treatment  42  8 50 40 5  45 
Control  41  8 49 41 8  49 
Total  83  16 99 81 13  93 
Student Attentiveness Observation Tool 
For this component of the study, only classrooms from third to sixth grade were selected, as the students 
in second grade were being pulled out for EGRA testing.17 Another reason for using third to sixth grades 
was to maintain consistency with the baseline study. The endline evaluation utilized a methodology like 
the baseline study. Overall, a relatively equal percentage of each classroom was observed between 
treatment and control schools (see Table 19) 
Table 19 
Classrooms Sampled for the Classroom Observation Tool 
Classroom  Treatment  Control 
Number  %  Number  % 
CE1  18 20.0 23 25.6
CE2  33 36.7 23 25.6
CM1  23 25.6 26 28.9
CM2  16 17.8 17 18.9
Other  0 0.0 1 1.1
Total  191 100.0 170 100.0
Overwhelmingly, French classes were observed in both treatment and control schools (see Table 20). 
Other classes observed included mathematics, communication, writing expression, and general revision. 
Table 20 
Subject Observed for the Classroom Observation Tool 
Classroom  Treatment  Control 
Number  %  Number  % 
French  83 92.2 81 90.0
Other  7 7.8 9 10.0
Total  90 100.0 90 100.0
At each school, two classrooms were randomly selected (using the hat method and the RNG application) 
for observation. In some cases, enumerators were not able to complete two classes, but a total of 
                                                            
17 As in the baseline, it is assumed that such classrooms would have sufficient disturbance to negatively bias the 
results, since students would be paying attention to the movement of their peers instead of the teacher. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
22 
180 classes were observed out of a programmed 198 (see Table 21). Within the classroom, 10 students 
were randomly selected to be observed and classified as attentive or inattentive at the time of 
observation. Nine of the classrooms observed had fewer than 10 students present, 12 classrooms had 
fewer than five girls, and eight classrooms fewer than five boys. The random selection of students was 
done using templates with already chosen “points” in the classroom. Gender was also taken into 
consideration so that equal numbers of girls and boys would be observed, which in practice also occurred 
with 449 girls and 451 boys sampled in treatment schools and 451 girls and 434 boys sampled in control 
schools. Overall, a total of 1,763 students were observed with the student attentiveness observation tool 
(see Table 21). 
Table 21 
Programmed and Performed Sampling of Classrooms and Students for the Student Attentiveness 
Observation Tool 
Type of school  Number of classrooms  Number of students 
Programmed  Done  Programmed  Done 
Rural  Urban  Total  Rural  Urban  Total  Male  Female Total  Male  Female Total 
Treatment  84  16  100  78 12 90 500 500 1,000  451  449 900
Control  82  16  98  74 16 90 490 490 980  434  451 863
Total  166  32  198  152 28 180 990 990 1,980  885  878 1,763
If in some cases classrooms were combined because of a lack of enough teaching staff, enumerators were 
instructed to observe students in only the randomly‐selected grade if the two grades could be 
distinguished from each other (for instance, teachers of combined classrooms may assign one grade to sit 
on the right side of the room, and the other grade to sit on the left). On one occasion in a control school, 
CE1 and CE2 students were mixed (recorded as “other” in Table 19 above) and because of the small sizes 
(5 and 8 students) of both classes, both grades were observed together. 
3.4  Data Analysis 
Ultimately, all data analysis had the specific goal of either answering one of the evaluation questions 
stated in the ToRs or addressing the 14 indicators for which the evaluation team was responsible for 
collecting data. Data analyses (such as DID, Chi‐square tests, student t‐tests, Wilcoxon tests, and 
regressions) were primarily focused on documenting the indicators answered by the surveys,18 but were 
also used to assess some of the other questions asked in the surveys. However, DID analyses did primarily 
focus on the McGovern‐Dole indicators. DID analyses were done at the “individual level” and weighted 
values were only used for the attentiveness indicator (issues with the weighting scheme used in the 
baseline are discussed in Section 3.5). 
                                                            
18 Such as percentage of boy/girl students who, by the end of two grades can read and understand the meaning of 
grade‐level text; percentage of students in target schools who are identified as attentive during class/instruction; 
percentage of students in schools receiving USDA assistance receiving a minimum acceptable diet; and indicators in 
the area of short‐term hunger reduction, health‐related absences reduction, and community understanding of 
benefits of education. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   23
Specific econometric models were used in a few cases where deemed necessary, such as for active AMEs 
when accessing “minimum acceptable diet” and “child absences due to illness” (as the presence of an 
active AME may be correlated with health and hygiene measures) and classroom size when assessing 
“student attentiveness” (as classroom size could negatively affect attentiveness).19 Variables such as the 
gender of the respondent and school rural or urban status were always included in the models—bearing 
in mind that overloading the models and running tests to compare various models is never 
recommended, as the model should be developed with a hypothesis in mind and tested. 
Other evaluation questions were answered through figures and tables developed through the survey 
results, as well as in appropriate cases statistical analyses, often providing supporting evidence to 
information derived from the qualitative exercises. For student indicators, values were developed using 
the weights/adjustment factors explained below, which present the inverse probability of selection. No 
weights/adjustment factors were used for principals, teachers or parents (see explanation in Section 3.5). 
All other indicators were informed by evidence provided by CRS through its bi‐annual progress reports. 
Cleaning of the datasets and basic analyses were performed in Excel. Other more complex analyses, such 
as regressions and DID analyses, were performed in R project. 
Adjustment Factors 
With respect to student attentiveness, classrooms were similarly adjusted for the probability of selection. 
The adjustment factor for classrooms is as follows: ܣ݆ܿ=݆݊ / ܿ  
  where ݆݊ is the total number of classrooms eligible for observation in school ݆ (note that only third 
to sixth grades were eligible for selection), and ݆ܿ is the number of classrooms that were observed in 
school ݆. Additionally, this sample is adjusted based on the probability of selecting a student within the 
classroom. This adjustment factor is as follows: ܣ݆ܿݏ=݆݊ܿ / ݏ݆ܿ  
  where ݆݊ܿ is the number of students in classroom ܿ in school ݆, and ݏ݆ܿ is the number of students 
observed in classroom ܿ in school ݆. Thus, the final weight for classroom observation is as follows: 
  ݏ݆ܿܣ∗݆ܿܣ =ݏܹ݆ܿܨ
With respect to the EGRA indicator, the goal was to test 16 students from each school. Because students 
have a different probability of selection depending on the total number of students in each class, a weight 
was calculated as follows: ܨܹ݆݅= ݆݇݅/ݏ݆݅. 
  where, ݆݇݅ is the total number of students in grade ݅ from school ݆, and ݏ݆݅ is the number of 
students sampled from grade ݅ from school ݆. When the number of students is less or equal to 16 the 
sample weight was equal to 1.  
The student adjustment factors could only be used on endline observations, and endline values will be 
reported as both weighted and non‐weighted. 
                                                            
19 For DID the X is where one captures those other factors (such as the percentage of active mothers’ associations) 
that may affect the dependent variable of interest.
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
24 
Calculation of Indicators  
The equations for all indicators in this study can be found in Appendix 24.  
Calculations Made to Baseline Data 
The EGRA test used for the baseline was slightly different20 compared to that which was used by CRS/WEI 
after the baseline and thus also for the endline. Thus, EGRA sections 1 (baseline 17 questions and endline 
15 questions; a proportion was developed), 2 (baseline had 19 questions and endline had 18 questions; 
the one question was removed) and 3 (baseline had 100 questions and endline had 90 questions; the 
extra 10 questions were removed) in the baseline data were adjusted to match the endline data. EGRA 
section 4 was completely missing from the baseline data, even though the baseline report states that 
section 4 on producing letter sounds “was not used” and that “the introduction of the letter sounds is a 
new development in the national curriculum that is beginning to be implemented.”21 The evaluation team 
believes this section should have been done to allow a baseline value to be available. Finally, the indicator 
associated with EGRA—improved literacy defined as obtaining 2 out of 5 points “PASS” (less than 50%) on 
the reading and comprehension parts, which are sections 8, 9 and 10—was devised after the baseline; 
thus, the evaluation team calculated the values using the baseline data. 
Unfortunately, it does not appear that the baseline asked questions of parents for all school children, but 
instead only derived values for one child. Thus, two indicators derived from the parent survey (minimum 
acceptable diet and reported hunger during the school day) do not examine all school children, but only 
one per parent. This must be kept in mind when comparing baseline and endline results. 
Additionally, and as discussed in Section 3.5 below, baseline indicator values were recalculated by the 
evaluation team, using the same method as the indicators were calculated in the endline (no weights for 
principals, teachers or parents). Baseline indicator values developed using both baseline calculations and 
endline calculations are presented later in this report (Section 4.2, Table 25). 
Qualitative Data  
Data generated through the KIIs and FGDs were analyzed through a variation of the constant comparative 
triangulation approach through which the analysis of each set of data was considered when analyzing the 
next. This serves as a means of identifying and confirming patterns and highlighting any inconsistencies, 
which could then be checked.  
These analyses, when coupled with the survey/observational/EGRA data, provided a comprehensive 
picture of what the McGovern‐Dole project was achieving or not achieving, for whom, and why. It 
enabled identification of trends in implementation experience and results based on what had been done, 
how and with whom, in ways expected to inform ongoing project decisions and practice. Where there 
were significant gaps in the emerging picture of the project, it provided enough data to support an 
analysis of the discrepancies as they related to variables such as gender, age, location and role. In all 
                                                            
20 The consultants do not know the reason for this and they were solely provided the two versions. They noticed the 
differences between the two tests and communicated this to CRS before data collection began.   21 Brown, D., and J.C. Guzmán. (March 2015). CRS Food for Education Baseline Study. Notre Dame (Indiana), The 
University of Notre Dame, Notre Dame Initiative for Global Development. p. 35. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   25
cases, a direct link was maintained from evidence (data), through findings to recommendations and 
lessons. 
3.5  Challenges, Issues and Limitations 
The following challenges, issues and limitations have been identified for this endline evaluation: 
 Inappropriate sample and statistical limitations – From the very beginning, in the baseline process, 
there appeared to have been issues with sample selection. Within the beneficiary schools there are 
79 rural schools (55%) and 64 urban schools (45%); therefore, an appropriate stratified sample 
should have been relatively equal. However, it appears that at the beginning of the project there 
was a limited number of urban control schools available.22 For both the baseline and mid‐term only 
7 urban schools were assessed and 43 rural schools. However, the evaluation team did not find the 
rationale explaining the choices made for baseline sample selection and can assume that it was 
based solely on the small number of control urban schools available (June 2017 and endline). The 
evaluation team adhered to the past sampling regime (sampling 8 urban schools and 42 rural 
schools) but noted that the sample was heavily biased towards rural schools.23 However, because 
almost the same population of schools24 were being sampled as in the baseline, changes in the 
indicators and differences between control and treatment schools can be thought of as reliable.  
 Stability of the sample – The evaluation design paid attention to ensuring consistency of the sample 
of intervention schools and control schools and communities from baseline to the present 
assessment. However, it cannot be known with certainty that within‐group stability has been 
maintained. Children regularly enter, drop out and transfer or are absent for long periods, thereby 
missing major portions of the treatment. Teachers similarly move, and family make‐ups change. 
The evaluation team tried, with limited success, to confirm through school directors and files how 
much shifting of sample populations there had been and to discuss this qualitatively. As a result, 
variables measured in the baseline that were deemed important to indicators were examined in 
the endline. 
                                                            
22 The schools to benefit from the intervention were chosen from among “the schools made available by the 
government” (that is, schools in the area eligible for project implementation, without school canteens programmed 
and for which CRS received discharge for intervening). These available schools were 64 urban schools and 79 rural 
schools for intervention. For the control schools, it was found necessary to identify the schools of the surrounding 
communes (in order to ensure a good comparability in terms of nutritional habits, fact which influences the impact 
of the Project). Thus, we obtained only 7 urban schools available against 91 rural schools in the control area. 
For reasons of comparability: (a) 43 rural schools have been selected from the 91 rural control schools versus 43 
rural schools selected from the 79 rural intervention schools; and (b) all the 7 urban control schools have been 
selected versus 7 urban schools chosen among the 64 urban intervention schools. At baseline, within each stratum, 
all theses schools were randomly selected with a probability proportional to the size of the school, and all the 
7 urban control schools were automatically selected. 
23 The statistical weighting cannot correct for the fact that the original sample selected may have bias. And this 
cannot be assessed because it is not known how the 7 urban schools selected in the baseline were representative of 
the actual 64 urban school in the sample. Examining whether these urban schools were representative of the total 
urban school population was not possible under the conditions of the endline evaluation TORs. Also, it should be 
noted that indicators should not be assessed at the “school level” according to the indicator definitions.  24 Several schools split into two schools (A and B), two schools (A and B) merged, one control school was no longer a 
suitable control, etc. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
26 
 Teachers’ strike – The situation with Beninese primary school teachers appeared to be very 
sensitive during this endline evaluation period. On February 7, 2018, the evaluation team was 
informed that many teachers had been on strike for around three weeks and negotiations had not 
been successful. Consultants were informed that not all schools responded similarly to the strike, 
incorporating further variability into the sample set. Teachers/schools were still on strike at the end 
of February 2018, when one of the international consultants arrived in Benin after having received 
the go‐ahead signal from CRS to proceed with the field mission. The schools re‐opened in the week 
of the 5th of March, but the teachers Unions hardened their position allowing for only one day of 
classes per week, instead of the 3 days originally thought by CRS. At this point it was decided that it 
would not be possible to conduct the data gathering for this evaluation with such a short amount 
of school contact hours available every week. Ultimately, the strike went on until the very end of 
April 2018, and a few comments from the principal survey suggested there was still some unrest. 
Data collection occurred in June 2018 (see field mission schedule in Appendix 25), one full month 
after teachers had resumed work, yet the strike undoubtedly had an impact on the data collected. 
The fact that there had been a four‐month long strike means that teaching and project 
implementation may have been affected, and the long‐term absence from school was likely to have 
had an impact on the students’ literacy capacity. 
 Time limitations – The amount of time for field work, the analysis of data and writing the report 
were reduced significantly from the time indicated in the ToRs due to the strike and other delays. 
The evaluation team made all efforts to mitigate for this challenge. Additionally, an extension of 
the original contract with CRS was made to attempt to ensure adequate provision of time for the 
field work, data analysis, report writing and addressing comments made by CRS management to 
the draft report. The evaluation team did suggest waiting one month after the end of the strike to 
begin fieldwork to ensure that students had time to recover from the prolonged absence from 
class. Had they conducted the EGRA examination on the first week of classes, the evaluators feel 
they would have had a very negative result, not reflective of the work of CRS and the schools. Also, 
dates of national and regional white exams communicated late given the revision done in the 
school calendar which narrowed the window for data collection. Overall, even though there were 
many issues that occurred outside of both teams control, the result was the successful completion 
of the final evaluation. 
 Enumerators not hired for first day of training – Stemming from the constraints mentioned in the 
point above, not all required enumerators/supervisors were able to attend the first day of training. 
For example, five enumerators showed up during the second day of EGRA training and three 
enumerators during the third (and last) day of EGRA training, and approximately five showed up 
the second day of training for the non‐EGRA surveys. Ultimately, not receiving the full training 
could increase enumerator error, even though the evaluation team did make the effort to get 
these late arriving individuals up to speed.  
 Convenience‐based sampling in parent survey – During the first three days of field work, it appears 
that convenience‐based sampling for parents occurred. Supervisors were not able to make the 
random selection either for the first day of sampling (Friday) because the day before was the pilot, 
or for the second day of sampling (Monday) as enumerators/supervisors stayed in Kandi on the 
Friday and schools were not open to make random selection over the weekend. On Tuesday there 
were also some issues with not using RNG for sample selection. Thus, supervisors collaborated with 
school directors to choose parents who lived nearby and were available to come to the school for 
the survey. It should be noted that the evaluation team had agreed to redo these surveys; 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   27
however, because of the wrongly communicated sampling dates by CRS (discussed under “time 
limitations,” above) this was not possible. Regression analyses were done post‐field work to assess 
whether any potential bias resulted from this; fortunately, results indicated that any bias was 
unlikely (see Appendix 23).  
 EGRA sample selection methods – CRS raised concerns about the sample selection method of 
students for the EGRA test. The work plan had indicated that RNG would be used for sample 
selection. The evaluation team brought on board an EGRA expert who had also consulted for WEI 
to help with its EGRA implementation. This expert taught stratified sample selection, which is also 
included as a valid selection method in the USAID EGRA Manual. CRS also expressed concern that 
the first student in each class was not being selected randomly; however, the EGRA expert on the 
evaluation method taught two methods to do this: (1) the hat technique, which involved writing 
student names and putting them in a hat or a box, picking one name randomly from the hat or box, 
seeing the position or corresponding number of that student and repeated that step to select the 
other students; and (2) provide numbers to each student in the first group of students (enumerator 
does not see the numbers), pick one number at random (1, 2, 3, etc.), and follow this selection in 
the other groups taking into account the step.25 The evaluation team is confident that, because of 
the training received, these are the methods that enumerators used to make the selection of the 
first student. Nevertheless, sampling mistakes or errors could have been made in the field, 
especially by those enumerators who arrived at the end of the second day or on the final day of 
training. The stratified sample selection method was primarily used over the first two to three days 
of sampling and, at CRS’ request, RNG over the final days of sampling. The evaluation team does 
not believe that the issues mentioned above have resulted in any positive or negative bias on the 
sample selected. Regression analyses were also done post‐field work to assess if any potential bias 
results from this; however, results indicated that any bias was unlikely (see Appendix 23). 
 Sample size for each survey – In some of the surveys the number sampled was slightly less than the 
number programmed to be sampled. For the teacher survey, it was because fewer than six 
teachers were present at a school (some schools having the director teaching in four classes), For 
EGRA, there were sometimes fewer than 16 pupils in the class and on six occasions there was no 
CP classroom at the school. For the parent survey, all parents who were called for the survey could 
not always all come due to the farming season (they had to work in their fields). For the 
attentiveness observation, some schools did not have all the CM1, CM2, CE1 and CE2 classes and in 
a few cases, there were time constraints. Unfortunately, two treatment schools were not sampled 
after a miscommunication between enumerators/supervisors and consultants when CRS called for 
all sampling to stop due to the non‐use of RNG. The evaluation team assumes responsibility for this 
oversight. However, because of the very large sample sizes programmed (for EGRA and parents the 
maximum numbers were programmed to be surveyed instead of the minimum numbers; see 
Appendix 1) and the close achievements of these, the evaluation team has confidence in the results 
from the sample sizes obtained. 
 Multiplicity of data types and sources – The nature of the McGovern‐Dole project, range of 
stakeholders and requirements of the ToRs have led to a multiplicity of varying data perspectives 
and types, both qualitative and quantitative. This presents a challenge to maintaining consistency, 
recognizing patterns and identifying key points of differentiation. However, the evaluation 
mitigated this limitation by ensuring that all tools (KII, FGD and observation protocols and surveys) 
                                                            
25 This addresses the concern of CRS finding “pieces of paper with numbers” during their visit to some schools. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
28 
adhered to the evaluation matrix. Sources were clearly identified in the evidence matrix and 
emerging data were regularly analyzed through triangulation and a constant comparative 
approach. 
 Weighting used in the baseline (and MTE) – After close consultation with CRS headquarters and CRS 
Benin during the endline analyses, it was confirmed that the school weights used in the baseline 
should not have been attached to a sample selected using probability proportional to size 
methodology. Also, it should be noted that the endline study found no consistent trend between 
rural and urban schools (which may have been part of the reason for the development of these 
elaborate weighting schemes), with indicator values sometimes being more positive in rural 
schools and sometimes in urban schools. The baseline consultants also used “adjustment factors” 
(also weighting schemes) for principals, teachers, and parents, which again was inappropriate. For 
teachers and principals, the intent was to survey 100% of those positions in sample schools. If 
many of these individuals were not present on the day of the interview, the baseline weighted the 
observations they did have to accommodate for this. This is again incorrect, as these teachers and 
principals should not be construed as “representing” the teachers and principals who were not 
present. The limitation is that not all principals and teachers were present at the schools during 
sampling for both the baseline and endline. Parents in the baseline were weighted based on the 
school population; however, this is an assumption and the overall population of parents is 
unknown. The endline did not use these adjustment factors. Subsequently, baseline indicator 
values were recalculated by the endline consultants, using the same method as used to calculate 
the indicators for the endline. In Section 4.2, Table 25, baseline indicator values using both baseline 
calculations and endline calculations are presented. Adjustment factors (weighting) were to be 
used for both student‐level indicators (EGRA and attentiveness). Given absenteeism, these values 
likely represent an upper bound, as children who were absent from school on the day of testing 
might also be underperformers. Unfortunately, baseline EGRA data did not have the CP class sizes 
for 1212 schools (the students from these schools had been provided “0s” in the baseline, and thus 
were essentially “deleted” or worse if these “0s” were included directly in the analyses for the sake 
of preserving the weighting scheme) and it was determined that it would be more appropriate to 
use non‐weighted data rather than to eliminate all these data points. Thus, a comparison between 
baseline and endline section values was made using non‐weighted values and DID analysis was also 
conducted with non‐weighted values. For the baseline attentiveness values there were a total of 
6 schools (10 classrooms) for which the associated data for the number of classrooms in the school 
could not be derived from the principal survey. To allow for the weighting and to not delete these 
values, it was assumed that all CE1, CE2, CM1 and CM2 classrooms were present in these schools. 
 Variation in indicator calculations – In the baseline, the parent survey only asked questions, or at
least only reported on, one child per parent, even though at endline parents were found to have 
from one to seven children attending school. Thus, two of the indicators derived from the parent 
survey (increased use of health and dietary practices and reduced short‐term hunger) are based on 
only one child per parent in the baseline survey instead of one parent being counted as many times 
as the number of his/her children (as was done in the endline). Also, the percentage of student 
absences due to illness was calculated based on the total class population instead of the total 
number of absences in the baseline. However, to maintain consistency the calculation was 
performed both as in the baseline and as the indicator definition states in the endline. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   29
4.  EVALUATION FINDINGS 
The following section forms the heart of this report, summarizing the findings of multiple lines of enquiry, 
and melding qualitative and quantitative data. It is organized around the five evaluation criteria laid out 
by the Development Assistance Committee (DAC) of the Organisation for Economic Cooperation and 
Development (OECD) in its DAC Principles for Evaluation of Development Assistance; namely: relevance, 
effectiveness, efficiency, impact, and sustainability. Definitions of these criteria can be found at the 
beginning of each sub‐section, and in Appendix 27 of this report.  
The findings reported in this section respond to the specific questions that were posed in the ToRs for this 
assignment under each of these criteria, and which are listed in their entirety in the evaluation matrix 
provided in Appendix 2. However, many of these questions are inter‐related, and some overlap. 
Therefore, for ease of reading and to maintain good narrative flow, the findings respond generally to all 
the questions under each criterion, rather than specifically to individual questions. These questions are 
summarized at the beginning of each sub‐section. 
4.1  Relevance 
This section of the report discusses whether the McGovern‐Dole project did the right thing with the right 
people in the right place. It provides evidence that speaks to the appropriateness of implementation 
strategies adopted by the project and to the degree of satisfaction of stakeholders involved. 
4.1.1  Strategies to Improve Children’s Literacy 
The evaluation assessed the extent to which implementation strategies were relevant enough to improve 
children’s literacy, measured by improved literacy of school‐age children; the percentage of boy students 
who, by the end of two grades could read and understand the meaning of grade‐level text; and the 
percentage of girl students who, by the end of two grades, could read and understand the meaning of 
grade‐level text.   
EGRA Scores  
The McGovern‐Dole intervention had a significant effect (seen through non‐weighted DID analysis) on 
whether students, by the end of two grades, could read and understand the meaning of grade‐level text 
(calculated using sections 8, 9 and 10 of the EGRA test). This strongly suggests that the McGovern‐Dole 
intervention is, in fact, improving the ability of students to read and write French. 
Table 22, below, shows the results of an analysis of variance between treatment and control EGRA 
section scores at both baseline and endline. Using non‐weighted values, for all EGRA sections at the 
baseline, treatment school students had significantly lower average scores than control school students, 
except for sections 8b and 9 where no significant difference was found. In contrast, for all EGRA sections 
at the endline, treatment school students had significantly higher average scores than control school 
students. This is a notable change and suggests that the McGovern‐Dole interventions have had a positive 
impact on the EGRA scores of children attending the treatment schools. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
30 
Table 22 
EGRA Section Scores, Including the Average Score and Percentage of the Maximum Score for CP Students Surveyed at Baseline and Endline in 
CRS’ McGovern‐Dole Project Treatment Schools and Control Schools 
Test component  Passing 
score [1] 
Max 
points 
Baseline (CP)  Endline (CP) 
Treatment (n=740)  Control (n=591)  Difference [3]/ 
statistical test 
treatment 
Treatment (n=716)  Control (n=672)  Difference [3]/ 
statistical test 
treatment 
Average 
score N =  
Average % 
max 
Average 
score N =  
Average % 
max 
Average 
score N =  
Average % 
max 
Average 
score N =  
Average % 
max 
1.  Vocabulary  10  15 8.62 57.5 9.11 60.8 ‐3.3  ***  10.93 72.9 10.36 69.1 3.8 *** 
2.  Oral comprehension  12  18 12.05 67.0 12.64 70.2 ‐3.3  ***  12.67 70.4 11.69 64.9 5.5 *** 
3.  Letters names  45  90 7.97 8.9 11.60 12.9 ‐4.0  ***  18.12 20.1 12.03 13.4 6.8 *** 
4.  Letters sounds [2]  20  60 ‐ ‐ ‐ ‐ ‐    12.49 20.8 8.18 13.6 7.2 *** 
5.  Initial sounds  5  10 0.51 5.1 0.86 8.6 ‐3.6  **  4.24 42.4 2.65 26.5 15.9 *** 
6.  Familiar words  20  50 0.73 1.5 1.24 2.5 ‐1.0  **  4.36 8.7 2.05 4.1 4.6 *** 
7.  Invented words  10  50 0.47 0.9 1.08 2.2 ‐1.2  ***  3.86 7.7 1.91 3.8 3.9 *** 
8a.  Words in context  15  44 0.71 1.6 1.93 4.4 ‐2.8  ***  6.43 14.6 3.39 7.7 6.9 *** 
8b.  Text comprehension  3  5 0.00 0.0 0.00 0.0 0.0    0.18 3.6 0.07 1.4 2.2 *** 
9.  Oral comprehension  3  5 0.25 5.0 0.26 5.3 ‐0.3    0.55 11.0 0.35 7.1 3.9 *** 
10.  Dictation  15  18 1.06 5.9 2.08 11.5 ‐5.6  ***  5.45 30.3 4.39 24.4 5.9 *** 
Legend: *** p<0.001; ** p<0.01; * p<0.05. 
Notes: 
1. Evaluation des compétences fondamentales des écoliers des basses classes en lecture (EGRA) dans les écoles ciblées par le projet Food For Education Bénin. May 2017. Page 6. 
2. Section 4 was completely missing from the baseline EGRA dataset. 
3. Difference = % max treatment ‐ % max control 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   31
CRS had the ambitious goal of 65% of boy and 60% of girl students who, by the end of two grades, could 
read and understand the meaning of grade‐level text. However, the actual method to measure this 
indicator was not determined until the MTE. Therefore, the evaluation team calculated the baseline 
values (non‐weighted) for this indicator and obtained values of 0.3% for boys, 1.1% for girls and 0.7% for 
all children. These numbers suggest that the target devised at the beginning of the project was far too 
ambitious. Nevertheless, the endline values (weighted) rose to 13.0% for boys and 12.2% for girls (non‐
weighted 14.0% and 12.5%, respectively). 
Teacher Training 
The positive effect on student EGRA scores probably stems from the delivery of increased teacher 
training reading groups, and increased reading material distributions in the wake of the MTE, as the core 
features of McGovern‐Dole has been training teachers in pedagogies of reading instruction, the reading 
groups and use of learning‐to‐read booklets.26 According to the teacher survey, almost 100% of 
participants have found the teacher training excellent or good and some 97% are applying the knowledge 
and skills acquired. The principal survey reported a similar opinion. The students and parents had very 
positive feedback on the reading groups and reading materials, as evidenced through the qualitative data 
gathering exercises conducted.  
While disrupted to some extent by the strike, which affected all 146 schools from January 16, 2018, until 
the last day of the school semester, class visits were conducted in all four McGovern‐Dole target 
communes. The ensuing monitoring report indicates that most of the trained teachers were 
implementing new methods in their classroom practice, “leading to a qualitative change in reading and 
writing behavior among their learners.”27 Teachers are using the national literacy curriculum to a large 
degree. According to progress reporting by the McGovern‐Dole project, the creation of “reading days”28
has fostered better reading habits among pupils, while at the same time encouraging parents to see the 
value of education by witnessing “their children’s reading progress [...] Parents and teachers provided 
vivid testimonies on improvements in the young children’s reading capacity.”29
Notwithstanding the above, some challenges remain. While there were no observational data confirming 
changes in teachers’ behavior, and examples given of new practices were limited (for example, reading 
the booklet to students, demonstrating decoding, and teaching letters), documentary evidence indicates 
that teachers continue to struggle with basic good practices such as maintaining accurate files on school 
                                                            
26 Literacy training delivered to the teachers used a mix of audiovisual materials, and dealt with the physical aspects 
of reading, including aural mechanisms (the process of hearing) and how they relate to language production, 
perception and vision; daily practices to promote reading (activities that use the children’s names, news of the day 
and date of the day); and activities that use reading booklets. The teachers also received training on nutrition, 
hygiene and life competencies, more specifically on hygiene and illnesses, water hygiene, cleanliness of the home 
environment, hand‐washing, nutrition and hygiene when preparing food, and physical and clothing hygiene. Parents 
and AMEs referred to some of this information in the various FGDs conducted by the evaluation team.  27 Karimou Guera, M. Rapport/synthèse de l’observation des pratiques de classe et de validation des résultats de la 
première phase de suivi/coaching pédagogique des pratiques de classe. pp. 6‐7.  28 Reading days are specially organized events at the schools, with the presence of the APE/AME, communities and 
teachers. A reading day includes a competition between children of the CI and CP classes of the same school, with 
the goal of improving the reading level of students. Reading days have been previously organized in schools in all 
4 project communes. 
29 FY18 Semi‐Annual Narrative Report CRS BENIN McGovern Dole Award Reporting Period: October 2017‐
March 2018. p. 5. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
32 
children (not a critical of the training provided to teachers) and sufficiently preparing their lessons. As a 
result, some teachers are struggling to apply items in the syllabus and complain about not having enough 
time to complete daily activities.30  
Further, fewer than 10% of teachers reported their students’ reading and writing had improved, a 
shockingly low outcome‐level change for a key indicator. High pupil‐teacher ratio (PTR) has been a 
problem in some classrooms, making implementation of some one‐on‐one reading practices and the use 
of picture booklets difficult. However, it was not clear from the data gathered by the evaluation whether 
specific strategies for dealing with the numbers at the class or school level were being effectively devised. 
Another troubling fact is that, when surveyed about whether they had “enough time” for the literacy 
class, teachers essentially responded by pointing at the students. For instance, teachers who felt there 
was not enough time blamed the fact that children had limited knowledge to begin with, were 
slow/needed time to comprehend, or had difficulty reading. In contrast, teachers who felt there was 
enough time mentioned that children could keep up and learn what they were being taught. Such 
attitudes are indicative of a lack of recognition by teachers of their central role as primary agents of 
children’s learning. 
Finally, principals surveyed from control schools reported teachers spending more time each day on 
reading with their CI and CP students compared to treatment school principals. This may be an accurate 
assessment and, if so, it would be important to know why—especially since McGovern‐Dole principals, as 
well as teachers, are being trained on the importance of this instruction. However, it may be that 
McGovern‐Dole principals because of their training are simply more rigorous as to what “counts” as 
teaching reading, such as actively engaging with students in reading activities versus sitting at the desk 
while students do silent reading. This would be a positive finding and is supported by the fact that 
principals mentioned they were trained on how to monitor and provide support to teachers regarding 
literacy training. 
4.1.2  Strategies to Improve Enrollment and Attendance 
The evaluation also assessed the extent to which the implementation strategies were relevant enough to 
improve enrollment and attendance among students, particularly girls, measured by reduced health‐
related absence, and increased student enrollment. 
Health‐Related Absences 
To reduce health‐related absences, the McGovern‐Dole project has been actively promoting better health 
and hygiene. The evaluation team saw no observable decrease in the number of health‐related absences. 
However, children in treatment schools were somewhat less likely than children in control schools to 
have had stomach ailments (treatment: 5.0%, control: 10.0%) or diarrhea (treatment: 3.1%, control: 
6.7%), which could be, in part, a result of improved hygiene, such as washing their hands at school. 
According to the parent survey, around 60% of children were sick in May 2018 due to malaria, a 
mosquito‐borne illness that neither better health and hygiene nor better nutrition can effectively 
                                                            
30 Karimou Guera, M. Rapport/synthèse de l’observation des pratiques de classe et de validation des résultats de la 
première phase de suivi/coaching pédagogique des pratiques de classe. pp. 6‐7. Some of the issues mentioned, such 
as the lack of time to accommodate for literacy activities that are outside of the national curriculum like the 
distributed reading materials were corroborated in the FGDs with teachers. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   33
mitigate. Mitigation measures to reduce malaria would include things such as reducing mosquito 
breeding areas, insecticidal sprays, and bed net usage, which were not part of the McGovern‐Dole 
project. In the FGDs, some of the members of the AMEs and APEs mentioned that they had received 
some information on how to keep their home environment clean, particularly in relation to stagnant 
water around the house. 
That said, it was positive that parents were aware of bed nets to manage malaria and were also ready to 
suggest these be included in the McGovern‐Dole project. It was perhaps a gap in the health training that 
more did not appear sufficiently mobilized to manage breeding areas around their homes, which is more 
within their own capacity to manage—although several of the APEs/AMEs recognized that environmental 
hygiene around the house was an important vector. Additionally, they were keenly aware of the hygiene 
and sanitation elements of the project and recognized the importance of hand‐washing for students and 
for parents before cooking.  
School Enrollments 
Using semester values before the strike, targets for increased school enrollment of boys had been met 
and almost met for girls. Since measurement at the baseline, values for both boys and girls steadily 
increased each reporting semester until the October 31, 2017 to March 31, 2018 semester, which is likely 
a result of the teacher strike that occurred. The parent associations, particularly the AMEs that the 
McGovern‐Dole worked closely with, played an active role in mobilizing parents of school‐aged children 
to enroll their children in school. With the AMEs knowledge of their communities and which families had 
school‐aged children, they would actively visit the families and sensitize parents to the need of enrolling 
their children in school. Also, the McGovern‐Dole radio ads explained the importance of education and 
raised parents’ awareness to the need of enrolling their children in school. 
4.1.3  Strategies to Improve Community Participation and Engagement 
The evaluation assessed the extent to which the implementation strategies were relevant enough to 
improve community participation and engagement, measured by increased community understanding of 
benefits of education and increased engagement of local organizations and community groups. 
The primary strategy set by the McGovern‐Dole project in this area has been to support PTAs or similar 
“school” governance structures. The percentage of active PTAs was already quite high in the baseline 
(122 schools in the intervention area had active PTAs, which is approximately 92% of the schools). In 
contrast, the McGovern‐Dole project does seem to have had a significant impact on the creation of AMEs 
(seen through DID analyses discussed in Section 4.5 of this report). 
Based on the principal surveys, all principals surveyed stated that an APE existed at their school. 
Treatment school principals, as compared to control school principals, were more likely to express that 
the APE was “very active” (80.9% versus 57.1%), while control school principals were more likely to 
express that the APE was “moderately active,” as compared to treatment school principals (40.8% versus 
17.0%) (see Table 23 for extrapolated values). Only one surveyed treatment school principal and one 
control school principal responded that the APE was “not active” and the reason in both instances was 
that parents did not have enough time. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
34 
Table 23 
Number (Extrapolated) and Percentage of Active APEs in Treatment and Control Schools  
Active APEs [1]  Treatment  Control  Difference 
Number (extrapolated, by strata)  141 110
Percentage  98.6 97.8 0.8 [2] 
Notes: 
1. Considered “active” if in existence and principal reported the APE to be either very active or 
moderately active (same calculation as baseline). 
2. Significant at 5%. 
In contrast, almost all treatment principals surveyed (97.9%) stated that there was an AME at their 
school, but fewer control principals said there was an AME (44.9%). Thus, it appears that McGovern‐Dole 
activities have had a very positive effect on the existence of AMEs. Also, treatment school principals 
expressed that the AMEs were “very active” (76.1%) and “moderately active” (23.9%), while control 
school principals expressed that the AMEs were “very active” (50.0%), “moderately active” (22.7%), and 
“not active” (22.7%; see Table 24). 
Table 24 
Number (Extrapolated) and Percentage of Active AMEs in Treatment and Control Schools  
Active AMEs [1]  Treatment  Control  Difference 
Number (extrapolated, by strata)  134 36
Percentage  93.6 32.2 61.4 [2] 
Notes: 
1. Considered “active” if in existence and principal reported the AME to be either very active 
or moderately active (same calculation as baseline). 
2. Significant at 1%. 
Based on the FGDs, there has been a significant improvement of community involvement in schools 
engendered by the project. For instance, according to mothers in one community, McGovern‐Dole “has 
allowed for a better collaboration [...] it has changed the vision of the school world and of its importance 
for children in the community.” APEs/AMEs have become a more intrinsic part of school functioning: in 
addition to supporting management of the canteen and provision of extra food, they have mobilized 
resources to pay for additional teachers and encouraged long‐absent children to return to class. Critically, 
they appear to be making more durable commitments as well, as some 47% of SILC members reported to 
be spending “the additional income from SILCs directly on their children’s education, including the 
purchase of school supplies, contributions to school canteens, etc.”31
4.1.4  Stakeholder Satisfaction 
The evaluation also assessed whether stakeholders (including students, PTAs, AMEs, teachers, and local 
authorities) were satisfied with their participation in the project—and why or why not— measured by 
level of satisfaction of stakeholders with their participation in the project. 
                                                            
31 FY18 Semi‐Annual Narrative Report CRS BENIN McGovern Dole Award Reporting Period: October 2017‐
March 2018. p. 3. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   35
Evidence from Surveys 
While the surveys did not specifically ask if individuals were satisfied with their participation in the 
project, some evidence can be found in responses to the surveys on quality. 
Overall, all stakeholders surveyed in the treatment schools were very positive about how they thought 
the McGovern‐Dole project improved the quality of learning in their school, with parents providing the 
most responses, followed by principals and then teachers (see Figure 1).  
Figure 1 
Perception of Surveyed Principals, Teachers and Parents on how the McGovern‐Dole Project Improved 
the Quality of Learning in the School 
Results focusing on how the McGovern‐Dole project improved the quality of health and hygiene in their 
school were even more positive than those for improving learning, with principals providing slightly more 
positive results, followed by parents and then teachers (see Figure 2). 
Overall, of the parents surveyed, the largest percentage said the McGovern‐Dole project had very much 
improved the quality of life of their family (43.8%); however, rural individuals seemed much more positive 
than urban individuals. Moderate percentages of individuals also said their life quality improved “much” 
or “moderately,” with only very small percentages stating “little” or “very little” (see Figure 3). Thus, 
parents generally seem to be happy with how the project has improved their family’s lives. It should be 
noted that for the quality questions, individuals sampled from rural schools consistently gave more 
positive results than those from urban schools; however, because these questions did not have 
associated qualitative comments, the evaluation team cannot ascertain the reason for this. 
Generally, comments made by principals and teachers indicated that they were overall very satisfied with 
the project and its impact. Those individuals who were not satisfied with their participation in the project 
were teachers outside of CI and CP, who would also like to receive training.  
59.6
43.0
62.3
27.7
30.2
24.8
10.6
16.8
8.4
2.1
4.5
1.0
1.7
3.9
3.4
0 10 20 30 40 50 60 70 80 90 100
Principal
Teacher
Parent
%
Very much Much Moderate Little Very little Do not know
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
36 
Figure 2 
Perception of Surveyed Principals, Teachers and Parents on how the McGovern‐Dole Project Improved 
the Quality of Health and Hygiene in the School 
 Figure 3 
Perception of Surveyed Parents on How the McGovern‐Dole Project Improved the Quality of Life of their 
Family 
66.0
54.2
61.5
23.4
26.3
24.6
8.5
13.4
9.3
2.1
3.4
1.1
0.6
2.2
3.4
0 10 20 30 40 50 60 70 80 90 100
Principal
Teacher
Parent
%
Very much Much Moderate Little Very Little
47.8
17.4
43.8
23.9
36.2
25.5
17.1
34.8
19.4
3.7
2.9
3.6
2.2
1.9
5.3
8.7
5.7
0 20406080 100
Rural
Urban
Total
%
Very much Much Moderate Little Very Little Do not know
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   37
Evidence from FGDs and Documentation 
In general, the qualitative exercises have shown very positive responses by the various stakeholders to 
the McGovern‐Dole project. Students, families, teachers and directors were all very pleased and 
appreciative of the project. The most important elements of the project in their view were the 
food/school feeding and the teacher training. The study groups in the communities (held on Wednesday 
afternoons) have also been very appreciated. 
SILC groups are very valued by community members, although it is difficult to find a close connection 
between participating in SILC groups and improved learning outcomes of the project. Participants in FGDs 
have mentioned that participating in the SILC groups has allowed members to borrow and spend on 
educational materials and other education costs. Participants also referred to the SILC’s direct 
contribution of financial resources to schools, although this seems to be anecdotal rather than verifiable. 
The evaluators reiterate that families are very pleased with the SILC groups, and find them very valuable.  
In reviewing McGovern‐Dole project documentation, the evaluation team found that SILC groups were 
active in all four target communes, with a total of 392 groups comprising 8,752 women and 2,210 men. 
Of all these group members, 25.3% applied for credit, which averaged 109.950 CFCA per request, with 
women requesting lower amounts than men. Most loans were made for existing income‐generating 
activities (existing businesses or agriculture) but, on average, SILC group members invested 40.7% of the 
profits from SILC activities into their children’s education, which is an important number to consider. 
The documentation also indicates that the AMEs/APEs being very satisfied, as demonstrated by their 
significant contribution to the project. The AMEs/APEs have played an important liaison role between 
parents and communities and act as intermediaries for problem solving. The AMEs/APEs are a critical 
element in the school feeding program, performing a key role in the management of food items and 
preparation of meals. Furthermore, the AMEs play a role in ensuring children’s attendance and good 
hygiene. 
The APEs/AMEs have received considerable training from the McGovern‐Dole project. Training sessions 
dealt with monitoring and evaluation of activities, sexually transmitted diseases and HIV/aids, monitoring 
of the development of activities that would assist students in retaining better what was learned in school, 
monitoring of the execution of micro‐projects, and financial management. The AMEs also received 
training on nutrition and management of food items, how to prepare enriched foods, instruction about 
basic food groups, how to store foods, how to cook foods, and hygiene. The APEs/AMEs have 
demonstrated the knowledge they have acquired and put it into practice throughout the implementation 
of the McGovern‐Dole project. 
4.2  Effectiveness 
This section of the report discusses whether the McGovern‐Dole project achieved its expected results. It 
provides evidence that speaks to the overall attainment of targets, as well as the degree of achievement 
of specific results associated with strategic objectives to improve the literacy of school‐age children and 
increase their use of good health and dietary practices. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
38 
4.2.1  Overall Attainment of Targets 
Appendix 26 provides information that was given to the evaluation team before field work regarding the 
PMF for the McGovern‐Dole project in Benin, as well as information that was included in the inception 
report regarding the tool or data sources for information, if a specific endline survey was required, who 
was responsible for the information, and comments. 
Table 25, below, is based on the indicators supplied in the ToRs, including the list of specific indicators 
that needed to be addressed in the endline study.32 Examining information later provided by CRS, there 
have been several changes to indicators and specially to targets between the information provided in the 
ToRs and the USDA reporting template. These are reported directly in Table 25 or in notes found at the 
end of the table. 
Overall, of the 50 indicators reported, the McGovern‐Dole project has successfully reached its target for 
25 indicators and will very likely reach the target for six additional indicators at the end of this reporting 
period (September 30, 2018). There are nine indicators that were close to being met but will unlikely be 
achieved by the end of this reporting period and nine other indicators that were rather far from being 
met and will unlikely be achieved. Reasons for this will be discussed below. Additionally, one indicator 
“McGovern‐Dole illustrative: Percent of schools with soap and water at hand‐washing stations commonly 
used by students,” was measured for the first time during the endline and does not have a target. 
4.2.2  Results in Improving the Literacy of School‐Age Children 
It appears that the increased amount of teacher training since the MTE has had a positive effect on 
student EGRA scores. Even though the targets were not met, the evaluation team notes that there has 
been significant improvement. The target devised at the beginning of the project was much too 
ambitious, however.  
Targets were completely met this past semester for the number of individuals benefiting directly (male 
and female) and indirectly from the USDA‐funded McGovern‐Dole intervention: 
 Male direct beneficiaries included enrolled school boys (21,327); the number who have 
received nutrition training (607); the men who attended village general assemblies (2,112); and 
the male teachers/administrators who received training on new reading techniques (298), male 
SILC group members (individuals included above), and males who received food management 
committee training (individuals included above). 
 Female direct beneficiaries included all enrolled school girls (20,057); the females who received 
nutrition training (726) in the communities; the female SILC group members (2,805); the 
community health workers (CHWs) and nurses who were involved in the menstrual hygiene 
management training (23); the female teachers/administrators trained on new reading 
techniques (114); and female members of groups listening to the radio broadcasts, females who 
received food management committee training, who were cooks, and who were members of 
village general assemblies (individuals included above). 
                                                            
32 This is based on the most up‐to‐date information provided to the evaluation team. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
  39 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
McGovern‐Dole  SO 1: Improved literacy of school‐age children 
65% of boy students who, by the end of two grades can read 
and understand the meaning of grade‐level text 
0.3 0 0 0 0 1.2  1.2 1.2 13.0 65 13.0  
(non‐weighted) 
[2]
  (weighted) 
[3]
22,621 males benefiting directly from USDA‐funded 
interventions 
0 0 4,967 22,436 25,877 25,618  28,232 24,344 24,344 22,621 24,344  
60% of girl students who, by the end of two grades, can read 
and understand the meaning of grade‐level text 
1.1 0 0 0 0 1.7  1.7 1.7 12.2 60 12.2  
(non‐weighted) 
[2]
  (weighted) 
[5]
22,879 females benefiting directly from USDA‐funded 
interventions 
0 0 6,421 22,237 27,201 24,385  25,257 23,725 23,725 22,879 23,725  
220,783 individuals benefiting indirectly from USDA‐funded 
interventions 
0 0 2,444 287,270 289,768 309,152  309,280 287,969 287,969 220,783 287,969  
1.1 IMPROVED 
QUALITY OF 
LITERACY 
INSTRUCTION 
70% of teachers who devote at least an 
average of 45 minutes a day to literacy 
instruction [6] 
90.5 / 91.0    50.6 70% 50.6  
1.1.1 More 
consistent 
teacher 
attendance 
75% average teacher 
attendance rate for each 
school and aggregated by 
district 
92 / 93.3 0 0 92 96.7 96.7  85.15 85.15 ‐ 94.3 95% 94.3  
 (target
increased 
from ToRs 
info)
1.1.2 Better 
access to school 
supplies and 
materials 
4,230 textbooks and 
other teaching and 
learning materials 
provided as a result of 
USDA assistance 
0 0 0 10,800 0 343,080  326,498 354,140 1,034,518 1,013,746 1,034,518  
 (target
increased 
from ToRs 
info)
1.1.3 Improved 
literacy 
instructional 
materials 
144 teachers using the 
national literacy 
curriculum and the 
related instructional 
materials/Changed to: 
Percentage of teachers 
using the national literacy 
curriculum and the 
related instructional 
materials 
88.3 / 91.3 0 0 88.3 100 100  100 100 100 96.1 95 96.1  
(CI and CP)
  91.5
(principal)
1.1.4 Increased 
skills and 
knowledge of 
teachers 
144 teachers/educators/ 
teaching assistants 
trained or certified as a 
result of USDA assistance 
0 0 0 249 0 252  0 260
260
288 761  
 (target
increased)
0 0 0 0 62 0  204 0  204 97.4 216 266  
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
40 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
144 teachers in target 
schools who demonstrate 
use of new and quality 
teaching techniques or 
tools 
(CI and CP) (target 
increased)
1.1.5 Increased 
skills and 
knowledge of 
school 
administrators 
144 school administrators 
or officials trained or 
certified as a result of 
USDA assistance 
0 0 0 145 0 155  0 152
155
144 452  
106 school administrators 
in targeted schools who 
demonstrate use of new 
techniques or tools 
0 0 0 0 73 0  56 0
73
91.5 106 129  
1.2 IMPROVED 
ATTENTIVENESS 
75% of students in target schools who are 
identified as attentive during 
class/instruction 
65.79 [7] / 
69.54
0 0 65.79 65.79 80.9  80.9 80.9 ‐ 65.9 75 65.9  
1.2.1 Reduced 
short‐term 
hunger 
Less than 20% of parents 
in target schools indicate 
that their children were 
“hungry” during the 
school day 
55 /47 0 0 55 55 12.09  12.09 12.09 ‐ 7.1 20 7.1  
1.2.1.1 
Increased 
access to food 
(school feeding) 
95% of students in target 
schools consuming daily 
meals at school 
0 0 0 97.46 98.61 97.88  ‐ 98.3 98.3 95 98.3  
22,268 boy students 
receiving daily school 
meals (breakfast, snack, 
lunch) as a result of USDA 
assistance 
0 0 0 20,674 20,464 23,153  21,267 20,935 22,268 20,935  
23,153
21,536 girl students 
receiving daily school 
meals (breakfast, snack, 
lunch) as a result of USDA 
assistance 
0 0 0 19,083 19,785 20,140  19,913 19,739 21,536 19,739  
20,140
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
  41 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
18,756,998 daily school 
meals (breakfast, snack, 
lunch) provided to school‐
age children as a result of 
USDA assistance 
0 0 0 3,280,048 1,868,453 3,495,091  1,667,000 2,566,394 12,876,986 18,756,998 12,876,986  
80,703 take‐home rations 
provided as a result of 
USDA assistance 
0 0 0 28,708 20,303 29,561  19,640 18,619 98,212 80,703 98,212  
13,884 girls receiving 
take‐home rations as a 
result of USDA assistance 
0 0 0 7,750 7,393 8,405  8,623 7,936 40,107  13,884 7,936  
13,017 boys receiving 
take‐home rations as a 
result of USDA assistance 
0 0 0 7,154 8,010 7,486  8,974 7,646 39,270 13,017 7,646  
1.3 IMPROVED 
STUDENT 
ATTENDANCE 
19,607 boy students regularly (80%) 
attending USDA‐supported 
classrooms/schools 
18,893 0 18,893 18,893 20,199 20,199  22,444 0 22,444 19,607 22,444  
80% attendance during normal school 
operating hours during the school year 
19,069 female students regularly (80%) 
attending USDA‐supported 
classrooms/schools 
18,589 0 18,589 18,589 19,216 19,216  20,692 0 20,692 19,069 20,692  
80% attendance during normal school 
operating hours during the school year 
1.3.2 Reduced 
health‐related 
absences 
2% of students who 
report a decrease in 
health‐related absences 
5 / 6.8 0 0 5 5 7.4  7.4 7.4 ‐ 6.6 [8] 2 6.6  
1.3.3 Improved 
school 
infrastructure 
144 school kitchens 
rehabilitated/ constructed 
as a result of USDA 
assistance 
0 0 0 67 50 3  0 0 120 144 120  
252 cabins rehabilitated/ 
constructed as a result of 
USDA assistance 
252 Removed or 
cabins refers to 
latrines 
144 storerooms 
rehabilitated/ constructed 
as a result of USDA 
assistance 
0 0 0 137 1 0  0 0 138 144 138  
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
42 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
141 latrines rehabilitated/ 
constructed as a result of 
USDA assistance 
0 0 0 0 52 62  0 0 114 252 114  
144 wells and water 
stations/systems 
rehabilitated/ constructed 
as a result of USDA 
assistance 
0 0 0 34 110 0  0 0 144 144 144 Likely refers 
primarily to 
hand‐washing 
stations 
1.3.4 Increased 
student 
enrollment 
21,536 girls enrolled in 
schools with USDA 
assistance 
19,002 0 0 19,019 19,871 21,167  21,388 20,057 21,388 21,536 21,388 Will use the 
reporting 
period before 
the strike 
22,268 boys enrolled in 
schools with USDA 
assistance 
19,694 0 0 20,498 20,944 23,064  23,288 2,1327 23,288 22,268 23,288 Will use the 
reporting 
period before 
the strike 
1.3.5 Increased 
community 
understanding 
of benefits of 
education 
60% of parents in target 
communities who can 
name at least three 
benefits of primary 
education 
31.8 / 40.4 0 0 31.8 31.8 5.4  0 0 55.3 60 55.3  
1.4 
FOUNDATIONAL 
RESULTS 
1.4.1 Increased 
capacity of 
government 
institutions 
Students in 144 schools 
assessed using Early 
Grade Reading 
Assessment tool 
0 0 0 264 267 0  262 0 267 288 262 EGRA testing 
has occurred 
numerous 
times. Not all 
schools have CI 
and CP classes. 
1.4.4 Increased 
engagement of 
local 
organizations 
and community 
groups 
22,268 male social 
assistance beneficiaries 
participating in 
productive safety nets as 
a result of USDA 
assistance 
0 0 0 20,674 20,464 23,153  21,267 20,935 22,268 20,935 Refers to boys 
receiving meals
23,153
21,536 female social 
assistance beneficiaries 
participating in 
productive safety nets as 
a result of USDA 
assistance 
0 0 0 19,083 19,785 20,140  19,913 19,739 21,536 19,739 Refers to girls 
receiving meals
20,140
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
  43 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
97 parent‐teacher 
associations (PTAs) or 
similar “school” 
governance structures 
supported as a result of 
USDA assistance 
0 0 0 97 97 97  97 97 97 97 97  
100% of schools in target 
communities with active 
PTAs or similar “school” 
governance structures 
92.4 / 95.7 0 100 100 32 32  76 ‐ 76 97.8 100  76  
(92.2 
extrapolated by 
strata)
McGovern‐Dole SO 2: Increased use of health and dietary practices 
95% of students in schools receiving USDA assistance receiving a 
minimum acceptable diet
67 / 66.8 0 0 67 89.09 89.09  98 98 ‐ 73.7 85  73.7  
 (target has
been 
reduced)
2.1 IMPROVED 
KNOWLEDGE OF 
HEALTH AND 
HYGIENE 
PRACTICES 
40% of parents in target 
schools who achieve a 
passing score on a test of 
good health and hygiene 
practices 
10.09 0 10.09 10.09 10.09 10.09  8.4 8.4 ‐ 40 8.4  
50% of students in target 
schools who achieve a 
passing score on a test of 
good health and hygiene 
practices 
3.02 0 0 3.02 3.02 3.02  6.7 6.7 ‐ 50 6.7  
2.2 INCREASED 
KNOWLEDGE OF 
SAFE FOOD PREP 
AND STORAGE 
PRACTICES 
  100% of food preparers at 
target schools who 
achieve a passing score 
on a test of safe food 
preparation and storage 
N.A. 0 0 50.1 62.66 62.66  92.9 80.08 ‐ 100 80.1  
2.3 INCREASED 
KNOWLEDGE OF 
NUTRITION 
  4 female teachers trained 
in child health and 
nutrition as a result of 
USDA assistance (school 
gardens) 
0 0 0 0 9 11  0 0 20 4 20  
21 male teachers trained 
in child health and 
nutrition as a result of 
USDA assistance (school 
gardens) 
0 0 0 0 31 74  0 0 105 21 105  
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
44 
Table 25 
PMF for CRS’ McGovern‐Dole Project in Benin, including Semester Values, the Total to Date, Baseline and Endline Values, Targets and Final Value 
Result stream  Intermediate 
result 
Expected result  Baseline 
(baseline value / 
endline value) 
[1] 
2015  2016  2017  2018  Total to 
date 
Endline  Target (life 
of project)
Final value
(see 
legend) 
Comments 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
Apr 1‐
Sep 30 
Oct 1‐
Mar 31 
2.4 INCREASED 
ACCESS TO CLEAN 
WATER AND 
SANITATION 
SERVICES 
  100% of schools using an 
improved water source/ 
changed to 87 schools 
using an improved water 
source 
0 0 0 73 73 73  7 0 80 87 80  
144 schools with 
improved sanitation 
facilities (latrines) 
0 0 0 102 12 7  0 0 121 144 121  
2.6 INCREASED 
ACCESS TO 
REQUISITE FOOD 
PREP AND 
STORAGE TOOLS 
AND EQUIPMENT 
  144 target schools with 
improved food 
preparation and storage 
equipment 
0 0 0 144 0 0  0 0 144 144 144  
2.7 
FOUNDATIONAL 
RESULTS 
2.7.1 Increased 
capacity of 
government 
institutions 
4 government staff in 
relevant ministries/offices 
implicated in 
canteen/commodity 
management training 
0 0 0 1 4 0  1 0 6 [9] 4 6 
2.7.4 Increased 
engagement of 
local 
organizations 
and community 
groups 
71 nutrition or health 
initiatives or activities 
pursued in partnership 
between government and 
local community groups 
0 0 0 0 0 0  16 12 28 71 28  
McGovern‐Dole illustrative: Percent of schools with soap and 
water at hand‐washing stations commonly used by students 
[10] 
  64.6 No target 64.6  
Average number of students per hand‐washing station at USDA‐
supported schools 
6,330 0 0 194 46 50  50 50 50 50  
Legend: 
  Target 
reached 
  Target should be reached at final 
reporting period 
  Fairly close to reaching target, but unlikely to be 
reached at final reporting period 
  Far from reaching target, unlikely to be reached at final 
reporting period 
  Inappropriate target. Far from reaching target, unlikely to be
reached at final reporting period 
Notes: 
1. Baseline indicator values have been re‐calculated using the same method and weighting or non‐weighting as used in the endline. However, both the 
baseline calculated indicator value and the endline calculated indicator value are included. 
2. Weighted values for the baseline data are not available, as classroom size was not available for 12 schools. It was decided to report non‐weighted data, 
rather than deleted data point. 
3. Non‐weighted value is 14.0% for boys. 
4. For the indicators that have “double‐counted” this refers to the fact that the total indicator value should really refer to value obtained during the last 
reporting semester. For example, for the indicator “22,621 males benefiting directly from USDA‐funded interventions” the total to date would better read 
5. Non‐weighted value is 12.5% for girls. 
6. This indicator seems to have been removed from the USDA reporting template Excel spreadsheet. 
7. This was reported in the baseline report as 74.0%. 
8. This indicator value is calculated as in the baseline as number of absences due to illness divided by all students; however, 
the actual indicator value (35.9%) calls for as number of absences due to illness divided by all absences. 
9. This number is from the document review and could not be confirmed with government officials. 
10. To be considered a “yes,” 50% or more of hand‐washing stations must have soap and water in each school and the 
enumerator must provide the answer “yes, very much” to whether there is evidence that students are using the stations. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
  45 
“24,344” rather than “131,474,” which is derived from adding all semesters together. There is double‐counting because each semester many of the same 
students would be counted as enrolled and receiving feeding services from the canteen. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
46 
Improved Quality of Literacy Instruction  
Regarding the target of 70% of teachers who devote at least an average of 45 minutes a day to literacy 
instruction, the baseline found that 91.0% of teachers devoted at least an average of 45 minutes a day to 
literacy instruction; however, the endline study found this number has decreased and that only 50.6% of 
teachers devoted at least an average of 45 minutes a day to literacy. In the surveys, treatment group 
principals reported that 86.4% of CI and 86.7% of CP teachers devoted at least an average of 45 minutes a 
day to literacy instruction, which does surpass the target. Daily sessions reported by teachers for literacy 
instruction were often 45 minutes or more; however, it was the number of sessions per week that usually 
led the value to being less than 225 minutes.33 It was not asked in the baseline how many sessions a week 
the teachers performed literacy instructions and thus the baseline values may have been greatly 
overestimated. 
Regarding the target of 75% average teacher attendance rate for each school and aggregated by district 
(target appears to have been adjusted to 95%), from examining data collected by CRS it appears that 
average teacher attendance had decreased over the project, starting at over 90% attendance and 
dropping to around 85% at endline. However, 94.3% attendance was reported by principals during the 
endline evaluation, meaning that the target of 95% was very close to being met. The McGovern‐Dole 
project had planned to reduce teacher absenteeism by constructing housing, but this has not been yet 
accomplished. Thus, it is unclear how the project would have had any impact on this indicator. It should 
be noted that CRS has requested for a no‐cost extension to the project to allow them to build the teacher 
accommodation and other infrastructure work as per the project agreement. 
The original target of providing 4,230 textbooks and other teaching and learning materials (numbers in 
the ToRs) was reached in the third semester of the project. Subsequently it appears that the target was 
increased to 1,013,746 textbooks and other teaching and learning materials, and impressively this was 
also surpassed at 1,034,518. It is thus interesting to note that numerous treatment school teachers and 
principals still noted the lack of French textbooks and activity books, even in the CI and CP classes. 
Originally, the target was for 144 teachers to use the national literacy curriculum and its related 
instructional materials—this was changed to the percentage (target of 100% in one document; target of 
95% in previous biannual report) of teachers using the national literacy curriculum and the related 
instructional materials. Although CRS itself reports that 100% of teachers are using the national literacy 
curriculum and the related instructional materials, the endline found that 96.1% of CI and CP treatment 
school teachers (92.7% of all teachers) reported using them. Similar percentages were also found in the 
principal survey—principals were asked if their school followed the national curriculum (reading and 
writing) and used related teaching materials and 91.5% of treatment survey respondents stated yes (8.5% 
no). Thus, the target of 95% has been met, but it is unlikely that 100% of teachers are using the national 
curriculum as reported by CRS internally. 
It appears that targets for training and teachers using the skills and knowledge learned have been met 
and will be greatly surpassed if CRS continues to implement training this semester. For example: 
 WEI implemented a training of trainers, following which the training of teachers of CI (first 
grade) and CP (second grade) classes and their directors was organized in November 2017. The 
general objective of this training was to equip teachers of CI/CP classes and their principals with 
                                                            
33 Consultants were informed during the pre‐test of tools that a question should be added regarding the number of 
sessions teachers teach each week, as some taught sessions each day and others twice a week. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   47
knowledge of the pedagogical principles of teaching, learning, and assessment of reading and 
writing. 
 WEI also implemented other types of teacher capacity building. These were the community of 
practice, which was a forum of exchange of experiences and pedagogical practices between 
teachers who taught the same class, under the guidance of a pedagogical advisor or head of 
pedagogical unit; and strengthening classes with school children in CP classes on Wednesday 
evenings. 
Endline results showed 97.4% CI and CP teachers in target schools demonstrated the use of new and 
quality teaching techniques or tools. Similarly, the targets for increased skills and knowledge of school 
administrators have been met and will be greatly surpassed if CRS implements any other training this 
semester. Endline results showed that 91.5% (94.5% extrapolated by strata) of school administrators in 
target schools demonstrated the use of new and quality teaching techniques or tools 
Improved Attentiveness  
The target was 75% of students in target schools being identified as attentive during class/instruction. The 
value at baseline was 69.5% and rose to 80.9% at mid‐term; however, the MTE consultants used a 
different method of measuring attentiveness.34 The endline value of 65.9% (using the same method as 
the baseline), shows that there has not been much change in attentiveness and, in fact, there was a slight 
decrease. This could be, in part, because of the large sizes of those classes observed, a factor which 
would decrease attentiveness even if the children were well fed. Detailed evidence from classroom 
observation analyses is supplied in Box 1, below. 
Box 1 
Evidence from Classroom Observation Analyses 
The evaluation conducted a specific analysis of performance (attendance, attentiveness, and result) of school 
children who received THRs compared to other children who did not. To this end, it used the same methodology 
as the baseline to assess attentiveness, with the exception that the evaluators always attempted to sample equal 
ratios of boys and girls in each class. Several factors—including class size, whether the student eats school meals, 
and whether the student receives THRs—were evaluated using binary logistic regressions to assess their impact 
on student attentiveness. These factors were measured on a “by student” basis. 
Based on classroom observations, and looking at all schools combined, the variable class size has a significantly 
negative coefficient (z=‐3.231, p=0.001), indicating that the probability of a student being attentive decreases 
with the size of the class. In other words, the greater the overcrowding in a class, the less likely students are to 
be attentive. Interestingly, if treatment schools are examined alone the effect is not significant (z=‐1.503, 
p=0.133) and if control schools are examined alone the effect is significant (z=‐2.030, p=0.042). Considering that 
the average size of observed treatment school classrooms is 43.1 students (minimum=13; maximum=153) [see 
note] compared to 27.6 students (minimum=3; maximum=64) in observed control schools, it appears that the 
threshold where classroom overcrowding affects attentiveness has already been surpassed in the many of the 
treatment schools, but not yet in control schools. 
                                                            
34 Unfortunately, the tool was not provided in the MTE report, thus details on how the tool differed cannot be 
discussed. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
48 
Box 1 
Evidence from Classroom Observation Analyses 
In treatment schools, the vast majority of students were reported to have eaten school meals (860 “yes” and 40 
“no”). However, eating appeared to have a significantly positive coefficient (z=3.012, p=0.002), indicating that 
students in treatment schools were more likely to be attentive when they ate. In contrast, in control schools, the 
vast majority of students were reported not to have eaten school meals (27 “yes” and 836 “no”) and eating did 
not appear to have a significant effect on student attentiveness (z=‐1.76, p=0.078). Examining all schools 
together, there is a relatively even distribution between the population of students who eat and do not eat 
(50.3% and 49.7%). In particular, this equal distribution is noticeable on the one hand between the attentive 
pupils who eat and who do not eat (34.8% and 35.6%) and on the other hand between the non‐attentive pupils 
who eat and who do not eat (15.5% and 14.1%), ultimately leading to a non‐significant effect of eating school 
meals on student attentiveness (z=‐1.135, p=0.256). It should be noted that control school students, although 
not directly receiving “school meals,” may have eaten on their own throughout the day. 
In treatment schools, a total of 482 students (53.6%) observed for attentiveness reported receiving a THR and 
418 students (46.4%) reported not receiving a THR. In treatment schools, receiving the THR had a significant 
effect on attentiveness, with those receiving the THR being more likely to be attentive (z=2.335, p=0.019) than 
those who did not receive such a ration. Only three students (total of 863 students) reported receiving a THR in 
control schools, thus it is not appropriate to make comparison between those who did and those who did not. 
In general, observed students from treatment schools were somewhat less attentive (65.9%), than students from 
control schools (69.1%). Boys were slightly less attentive in treatment classrooms than girls, but the opposite was 
true in control schools. Also, in treatment schools, students in urban schools were less attentive than those in 
rural schools, but the opposite was true in control schools (see Figure 4). 
Figure 4
Percentages of Students Observed for Attentiveness (Weighted) who were ‘Considered to be Attentive,’ from 
Treatment and Control Schools and Disaggregated by Gender and Rural/Urban School Designation 
Note: This is for the schools sampled. The sample method (probability proportional to size) used to select schools in the 
baseline means the probability of selecting a unit is positively proportional to its size. Thus, larger schools were likely 
sampled. Many smaller McGovern‐Dole schools may not have this issue of overcrowding. 
68.2
59.5
63.9 68.0 65.9 68.1
72.6
70.2 70.0 69.1
0
10
20
30
40
50
60
70
80
90
100
Rural Urban Boy Girl Total Rural Urban Boy Girl Total
Treatment Control
%
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   49
One of the goals of the McGovern‐Dole project was for less than 20% of parents in target schools to 
indicate that their children were “hungry” during the school day. At the baseline, 47% (baseline 
calculation: 55%) of parents indicated that their child was hungry; this dropped to 12.1% at the mid‐term, 
and again at the endline to 7.1%. This is a significant change since the baseline, indicating that the 
presence of canteens in the schools has significantly affected this variable. 
CRS estimates that 98.3% of students in target schools are consuming daily meals at school, thus 
surpassing the target of 95%. The endline also confirms that on average more than 95% of children are 
consuming daily meals at school. The strike did not affect this indicator because CRS considers the 
maximum number of school children who have eaten at least once in the semester. However, the actual 
number of meals provided per semester to both boys and girls is somewhat below the target, and the 
overall number of meals to be provided over the life of the project is significantly below the original 
target (69% of the life‐of‐project target).  
Overall, a total of 98,212 THRs have been provided because of USDA assistance, which surpasses the 
target of 80,703. Progress equals the total number of THRs provided to students of grades 4 to 6 who met 
the attendance rate requirement (>95%), plus the number of monthly food for work (FFW) rations 
provided to cooks. During this past semester, the FFW rations were distributed every month, but the 
THRs were distributed only once because of the strike. In contrast, the number of boys and girls receiving 
THRs because of USDA assistance each semester was consistently below the target. 
Improved Student Attendance 
The target of 80% for the number of students (male and female) regularly attending USDA‐supported 
classrooms/schools was met in 2017 and will likely be reached in 2018, if school populations did not 
decrease too significantly. 
CRS developed a target of 2% of students who report a decrease in health‐related absences. Health‐
related absences were 6.8% (reported in the baseline report as 5%), 7.4% at mid‐term and 6.6% at 
endline.35 This target would be difficult to meet because of the already low baseline value and the issue 
that 60% of reported child illnesses in May 2018 were due to malaria. Because good hygiene and 
nutrition will not decrease the incidence of malaria, this indicator is not effectively measuring the results 
that are due to the McGovern‐Dole project. 
In terms of improved school infrastructure, some targets have not yet been met. However, plans are in 
place to meet the targets for kitchens and latrines (at the moment less than 50% of latrines have been 
constructed) before September 30, 2018. The strike did cause some delays in these activities not having 
been completed earlier in the year. 
Targets have been met for storerooms (138 rehabilitated/constructed, 6 schools did not need this done) 
and wells and water stations/systems rehabilitated/constructed (refers to hand‐washing stations or tippy 
taps). However, it should be noted that the endline study found that one school now does not have a 
tippy tap apparatus present at the school. Also, enumerators found 95.7% of schools had storerooms; 
however, there is the possibility that in the two instances where the enumerator reported no storeroom 
they did not make the tour of the school with the principal or a teacher and storerooms were therefore 
                                                            
35 All of these values utilized the equation for calculation used at the baseline, which was not the correct calculation, 
but is being used for comparability. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
50 
difficult for them to find. The consultants believe this was an oversight on the part of the enumerators 
and that CRS records showing all schools having a storeroom are correct. 
Using semester values before the strike, targets for increased school enrollment of boys had been met 
and almost met for girls. The collection of data by CRS on school children enrolled took place from 
December 2017 to January 2018. This past semester, due to the strike, the number of children enrolled in 
the school year 2017‐2018 decreased, which is not usual for the project. This could be explained by the 
fact that some parents who would have registered their children in January after the end of the harvest 
would not have done so because of the strike that began in mid‐January. Also, one school reported a 
decrease in enrollment due to flooding in the Malanville area. The last semester figures represent 93.1% 
of the life‐of‐project target (female) and 95.8% of the life‐of‐project target (male). However, endline 
school population values reported by principals indicate that enrollment may have been less than even 
those values collected in December 2017‐January 2018. 
At the baseline 40.4% of parents in target communities could name at least three benefits of primary 
education;36 however, this decreased at the mid‐term to 5.4%—a huge decrease most likely resulting 
from the way that the question was asked in the survey. During the endline, 55.3% of parents in target 
communities could name at least three benefits of primary education, which was very close to the target 
of 60% and a significant improvement. There is a chance this target will be reached by the end of the 
project if radio messages and other interventions are still occurring. Detailed evidence from the parent 
survey is supplied in Box 2, below. 
Box 2 
Evidence from the Parent Survey 
Parents were first asked “Do you consider that the time your children spend in school is worthwhile?” This was
slightly different from the question asked in the baseline, “Do you consider your children’s education 
important?” In the baseline, the parent needed to answer “yes” to then be allowed to provide reasons as to why; 
however, in the endline all parents were asked “Why do you consider your children’s education important?” [see 
note]. Overall, surveyed parents from treatment schools considered the time their children spend in school 
worthwhile (88.4%), slightly more so than parents from control schools (82.7%). Those individuals in urban areas 
considered the time their children spend in school worthwhile slightly more often than those from rural areas 
(see Figure 5). 
Taken as a whole, surveyed parents from treatment schools could name three or more benefits of primary 
education (55.3%), more often than parents from control schools (38.4%) (X2
=28.6, df=1, p<0.01). Generally, 
fathers could name three or more benefits of primary education slightly more often than mothers (see Figure 6). 
Most commonly, surveyed parents reported that primary education would help their children find a job, give 
them a better life and allow their children to care for their parents. The only category that was reported more 
often by surveyed control school parents was that primary education would allow their children to care for them 
(see Table 26). 
                                                            
36 Based on baseline calculations, 38.8% of parents (male: 46.6%, female: 26.0%) could name at least three benefits 
of education in the intervention areas and 46.7% (male: 48.3%, female: 41.4%) in the control areas. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   51
Box 2 
Evidence from the Parent Survey 
Figure 5
Percentages of Surveyed Parents who Consider the Time their Children Spend in School is Worthwhile in 
Treatment and Control Schools and Disaggregated by Gender and Rural/Urban School Designation 
Figure 6
Percentages of Surveyed Parents who Could Name Three or More Benefits of Primary Education from Treatment 
and Control Schools and Disaggregated by Gender and Rural/Urban School Designation 
As part of the McGovern‐Dole project, CRS and WEI promoted a series of radio messages in favor of education. 
Parents were asked “Have you heard the radio messages promoted by McGovern‐Dole (CRS, WEI) in favor of 
education?” Overall, a total of 75.0% of surveyed parents from treatment schools (rural: 76.0%, urban: 68.1%) 
stated that they had heard these radio messages, compared to only 30.7% from control schools (rural: 29.9%, 
urban: 35.0%). 
86.7
100.0
86.5 90.0 88.4
81.8
87.5 85.9
75.5
82.7
0
10
20
30
40
50
60
70
80
90
100
Rural Urban Male Female Total Rural Urban Male Female Total
Treatment Control
%
53.6
66.7
57.0 54.0 55.3
41.7
21.3
39.2 36.8 38.4
0
10
20
30
40
50
60
70
Rural Urban Male Female Total Rural Urban Male Female Total
Treatment Control
%
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
52 
Box 2 
Evidence from the Parent Survey 
However, the fact that 30.7% of parents from control schools indicated that they had heard these radio 
messages means that there could be some contamination for the above indictor naming the benefits of 
education. Also, when asked if these radio messages had changed their opinion about education, a striking 98.3% 
(354 of 360) of surveyed treatment school parents and 92.7% (127 of 137) of control school parents said “yes.” 
Of those treatment parents who heard a radio message, 57.3% could name three benefits of education 
compared to 46.9% who had not heard a radio message. Similarly, of those control parents who heard a radio 
message, 51.3% could name three benefits of education compared to 33.7% who had not heard a radio message 
(X2
=40.9, df=3, p<0.01). 
Table 26
Number and Percentage of Responses Regarding the Benefits of Primary Education by Surveyed Parents in 
Treatment and Control Schools 
Benefit  Treatment  Control 
Number  %  Number  % 
Job  362 68.6 307 61.2 
Care parents  261 49.4 262 52.2 
Better job  162 30.7 92 18.3 
Better life  326 61.7 286 57.0 
Better society  90 17.0 49 9.8 
Social status  59 11.2 52 10.4 
Do not know  5 0.9 1 0.2 
Other  37 7.0 35 7.0 
Total  528 100.0 502 100.0 
Note: Even if a parent did not consider the time their children spend in school to be worthwhile, they could still often provide 
reasons (although typically less than three) why their child’s education was important. 
Foundational Results 
The target of students in 144 schools assessed using the EGRA tool was achieved, as CRS reports having 
assessed students annually in all McGovern‐Dole schools. Additional classrooms were assessed during the 
endline evaluation and the evaluation team believes that WEI was also doing EGRA testing in May‐
June 2018. 
CRS met the target of 97 PTAs37 or similar “school” governance structures supported as a result of USDA 
assistance early on in project implementation. Activity in targeted communities38 included such things as 
listening sessions, support to organizing village assemblies, support to conducting a census of school‐age 
children, the provision of subsidies for the construction of various infrastructures, and follow‐up sessions 
with parents who had not enrolled their children. 
                                                            
37 These 97 PTAs covered the 144 schools under the project, as per WEI information.  38 It is not clear if several schools can be in a “targeted community.” If “targeted community” refers to schools, the 
evaluation team is not sure why the number 97 was selected and not all 144 schools. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   53
Subsequently, the next target is 100% of schools in target communities with active PTAs or similar 
“school” governance structures. According to the McGovern‐Dole project performance monitoring 
plan (PMP), this indicator is collected once a year. It will be done this current semester through a rapid 
organizational assessment (ROA) of the AMEs. The last ROA of the AMEs that WEI organized revealed that 
74 out of the 97 AMEs, or 76%, could be considered as active during the reporting period, as per the 
definition in the PMP (organize meetings and have meeting minutes). With the strike and its implications, 
it may be difficult for this target to be met. However, principals surveyed in the endline also indicated 
that all schools have PTAs and that they would consider 98.6%39 active or moderately active and 93.6% of 
AMEs active or moderately active.40
4.2.3  Results Increasing the Use of Health and Dietary Practices 
McGovern‐Dole developed a target of 85% of school‐age children receiving a minimum acceptable diet. 
The baseline found 67% of children receiving a minimum acceptable diet and the mid‐term 89.1%, and 
CRS reported during the March‐October 2017 semester that this rose to 98% (although it is not indicated 
how this was assessed). The endline found that 73.7% of school‐age children were receiving a minimum 
acceptable diet. Thus, the values found at the endline have increased since the baseline, but nowhere 
near as much as what CRS reported during 2017. It should be noted that 10.7% of treatment school 
children had only eaten one or two times in the day and that the canteen meal is only one meal per day. 
Improved Knowledge of Health and Hygiene Practices 
The percentage of parents in target schools who achieved a passing score on a test of good health and 
hygiene practices was 10.1% at baseline and 8.4% in June 2017, whereas the percentage of students in 
target schools who achieved a passing score on a test of good health and hygiene practices was 3.0% at 
baseline and 6.7% in June 2017. Targets are 40% and 50%.  
According to the PMP, this indicator is to be collected annually and this data will be collected during this 
semester. A refresher training of the WASH‐friendly clubs (including students’ parents) was done in 
November 2017, during which each club developed its action plan. However, the strike situation did not 
allow for any implementation and monitoring of these plans. When schools reopened, the project 
planned to intensify its support and accompaniment of the clubs. It was also planned to train the clubs on 
animation/sensitization techniques. 
Based on the above, it is unlikely that CRS will reach these targets, and it is also uncertain if this intensive 
learning will be sustainable and long‐term. 
Increased Knowledge of Safe Food Preparation and Storage Practices 
In April 2018 it was reported that 80.1% of food preparers at target schools achieved a passing score on a 
test of safe food preparation and storage, falling somewhat short of the 100% target. 
This information was collected through February and March 2018 during the refresher training post‐test. 
The number of participants in this training was 494 cooks, but only 472 participated in the test. The 
percentage who passed decreased compared to the previous semester (92.9%). Some new cooks had 
joined the group, but this only partially explains the decrease. CRS decided that a more thorough analysis 
                                                            
39 Extrapolated values by strata: 141 APEs active and 134 AMEs active.  40 It is not clear if CRS is now solely focusing on AMEs or if APEs are still relevant. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
54 
was necessary, and this was to be undertaken in May 2018 to understand the reduced achievement of a 
passing score, which will then be addressed through specific on‐site support activities. 
Increased Knowledge of Nutrition 
Twenty females and 105 males were trained in child health and nutrition as a result of USDA assistance, 
greatly surpassing the targets of 4 and 21, respectively. This indicator included both the creation of school 
gardens and the training of teachers in nutrition and child health, along with gardening techniques. 
Increased Access to Clean Water and Sanitation Services 
The target of 87 schools using an improved water source is close to being met, as 80 schools now have 
such water sources and plans are in place for the final seven. During last semester, the installation of 
tanks for the 47 schools of Kandi and Gogounou was completed. Final acceptance took place in 
April 2018. In addition, preparations were finalized for the hydrogeological study to form the basis for site 
selection for the seven remaining boreholes. The project also informed local government and contracted 
technical services for these upcoming activities to be carried out by CRS as part of the improved water 
source supply in the communities. 
Overall, the goal is for 144 schools to have improved sanitation facilities. Currently, CRS reports that 
121 schools have adequate latrines at this stage of the project. During the last semester, preparatory 
activities, such as signing contracts, paying first installments and recruiting the two engineers, were done. 
The CRS Global Technical Adviser for Shelter and Settlement was present to support the project team for 
the construction (reviewing the designs) of latrines in November 2017 and again in January‐
February 2018. The rest of the latrines are launched and will be completed during this semester. 
Confirming the above results, the environmental observation tool used in the endline found 102 schools 
(extrapolated value; however, 73.8% of survey schools) to have a water system, and 130 schools 
(extrapolated value; however, 95.7% of survey schools) to have latrines. However, it should be noted that 
both principals and teachers reported slightly lower values of “functional” latrines. Many girl students 
also mentioned that the toilets were very dirty and unpleasant to use. 
Increased Access to Requisite Food Preparation and Storage Tools and Equipment 
All project schools received their equipment at the beginning of fiscal year 2016, meeting the target for 
schools receiving improved food preparation and storage equipment. 
Foundational Results 
According to the document review, six government staff in relevant ministries/offices have been involved 
in canteen/commodity management training, surpassing the target of four. The evaluation team could 
not confirm this information with government officials. 
Overall, 28 nutrition or health initiatives or activities have been pursued in partnership between 
government and local community groups, which is still a long way from reaching the target of 71. During 
the last semester, 12 initiatives related to health, hygiene and nutrition were supported by CRS: World 
Hand‐Washing Day, International Cooking Day, and African School Feeding Day in the four intervention 
areas of the project. CRS supported these community initiatives by presenting their experience on the 
topic or by reflecting together with the participants to assess potential challenges. These days were also 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   55
an opportunity to communicate on good hygiene and sanitation practices and to advocate for more 
parent and community support. The McGovern‐Dole project advocated for town councils to support 
parents by contributing to school canteen activities. The Prefect of Alibiori Department, the highest 
authority in the department, proposed that a budget line be included in town council budgets to support 
school canteens.  
McGovern‐Dole Illustrative 
The endline study assessed the percentage of schools with soap and water at hand‐washing stations 
commonly used by students, and it was found 64.6% of schools had soap and water at most hand‐
washing stations and that students were commonly using them. 
4.2.4  Evidence from FGDs 
Overall, the qualitative evidence from participants in FGDs confirms that inputs of the McGovern‐Dole 
project are perceived as mutually reinforcing and constructive, although the FGDs did not provide much 
by way of concrete examples. The teacher training and regular Saturday meetings, the student reading 
groups meeting on Wednesday afternoons, and the distribution of reading materials were viewed very 
positively by the majority of FGD respondents, as were the school meals, infrastructure improvements 
and awareness‐raising of the parents. Capacity building, and especially the school canteen elements, have 
been a clear driving force for the mobilization of the parent associations. 
Broadly, and more so than indicated by the statistics, there was a feeling that the objectives regarding 
access and learning are being achieved by the project although, again, there was a certain vagueness as 
to the specifics. Comments such as the following were indicative of most: 
“The impact has been positive because our children study well and enjoy coming to school 
on their own will, without the parents running after them.” [Parent, Kidakperou] 
“There is a high level of success of children. Children perform better and are more 
intelligent. They speak better and more regularly French, even among themselves. They 
attend school more regularly.” [Mother, Soukarou] 
“The project has affected positively the participation of communities in the activities as the 
parents are now better interested in the education of the children, in case of a child 
dropping out the parents will be able to re‐integrate the child in the school. And the children 
themselves enjoy coming to the school because of the school canteen” [Member of male 
network, Soukarou] 
“Since the project has started the children are regulars in the school, there is stability in their 
health conditions, also due to the canteen. The number of children in the school has grown.” 
[Female teacher, Kidakperou]  
Most qualitative evidence from parents, including those who did not send students to school and whose 
illiteracy made it difficult to support their academic performance, recognized and valued the importance 
of education. FGDs with best preforming students indicated the key role of family in academic 
performance—parents and older siblings provide strong oversight and structure for children in terms of 
setting time to study and for chores and having breakfast before going to school. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
56 
4.3  Efficiency 
This section of the report discusses whether the McGovern‐Dole project used the right process to achieve 
and measure its expected results. It provides evidence that speaks to the timely achievement of 
objectives, implementation efforts and coordination between project partners, monitoring of project 
activities, and actioning of recommendations made in the MTE report. 
4.3.1  Timely Achievement of Project Objectives 
Evidence supplied in Section 4.2 suggests that approximately 70% of objectives were achieved or close to 
being achieved at the time of the endline evaluation. Timeliness is achieving objectives was not always 
consistent, but sometimes CRS continued providing assistance even after targets were met, as suggested 
by the following observations: 
 Overall, once the McGovern‐Dole project got up and running the targets were consistently met 
each biannual period (semester) for the number of individuals benefiting directly (male and 
female) and indirectly from the USDA‐funded McGovern‐Dole intervention. 
 WEI has continued to provide new textbooks and teaching and learning materials through the 
life of the project, even after reaching the target. Similarly, WEI continued to provide training to 
both teachers and administrators even after targets were met in the middle of program 
implementation. 
 Biannual results tended to be consistently below the targets for the number of meals provided 
and THRs provided per semester to both boys and girls, except for one semester for boys’ meals 
in the middle of program implementation. 
 Regular attendance targets were slightly below targets at the beginning, but by the middle of 
the project attendance targets were consistently being met. Also, generally the values for 
enrollment of girls and boys increased over the life of the project, although the strike caused a 
decrease towards the end. 
 In the 97 communities being measured for active PTAs or similar “school governance” 
structures, it appears that each semester WEI promoted activities. However, whether these are 
considered active seems to have fluctuated greatly over the course of the project. 
 Even after reaching the targets for females and males trained in child health and nutrition as a 
result of USDA assistance, CRS still implemented additional trainings. 
 Targets at 40% and 50% for the percentage of parents and students in target schools who 
achieve a passing score on a test of good health and hygiene practices appear to have been very 
far from being met throughout the course of program implementation. This is also the case for 
students achieving passing EGRA scores; however, when the MTE really brought the lack of 
progress to light there was increased effort in this area and improvements have been made. 
These types of indicators likely require the constant implementation of activities that can 
improve them, which is something that the McGovern‐Dole project did for other indicators 
(sometimes even when the indicator was achieved). For example, only 28 nutrition or health 
initiatives or activities have been pursued in partnership with government and local community 
groups, which is still quite a long way from reaching the target of 71. During the last semester 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   57
alone, 12 initiatives related to health, hygiene and nutrition were supported by CRS; if these 
activities had been done yearly or had been promoted more thoroughly, this could have helped 
improve the ability of students and parents to achieve a passing score on the test of good health 
and hygiene practices. 
 Infrastructure improvements directly related to canteen services, such as storerooms and 
kitchens, were finished early in program implementation. This makes sense as these really 
needed to be in place for the implementation of the canteen, which is the backbone of the 
McGovern‐Dole project. However, subsequent infrastructure developments more related to 
sanitation, such as latrines and water sources, were started in the beginning or middle of the 
project, but then results slowed down and currently the McGovern‐Dole seems to be playing a 
bit of catch up to complete these improvements. Furthermore, the construction of housing for 
teachers was never done. 
4.3.2 Implementation Efforts and Coordination Between Project Partners 
Evidence from the endline evaluation suggests that implementation of project activities and coordination 
between project partners were challenging at first, but greatly improved over time. For instance, 
difficulties were experienced initially with infrastructure programs, especially those implemented by WEI. 
For instance, kitchens that had been built had to be demolished as they did not meet mandatory 
structural criteria, and some contractors involved in the early stages of infrastructure development 
proved to be incapable of carrying out work, or outright fraudulent in how they implemented it. The 
project’s internal control and monitoring systems were successful at identifying and stopping malpractice 
on the part of contractors. Wrinkles in the hiring of technical staff to oversee infrastructure projects have 
now been worked out with CRS taking over this responsibility. Measures have also been taken to avoid 
duplication with agencies like UNICEF in the construction of latrines. 
Literacy training is another area that experienced problems initially. Right from the start, there was an 
eight‐month delay that prevented WEI from implementing training elements as per the prescribed 
design. Major improvements have since been made in delivering literacy training to teachers, and several 
innovative approaches have been identified to support student learning, such as Wednesday afternoon 
reading clubs. WEI has subcontracted two NGOs to assist in the implementation of capacity‐building 
initiatives within community‐based structures, such as APEs/AMEs. 
With regards to food distribution, evidence from the FGDs indicates that monitoring of storage facilities 
and quantities of food available have been adequate. Food items would be sent from Cotonou every 
three months and distributed to schools. Where shortages occurred in a given school, food items would 
be passed from another school that held a surplus, and “repaid” when the missing food items arrived. 
When asked about the fact that food items are all stationed in Cotonou, stakeholders made it clear that 
lack of infrastructure (warehouses) and weather (Kandi is warmer) were serious constraints to keeping 
food items locally. 
Overall, early challenges associated with implementation and partner coordination were successfully 
addressed. To quote one partner interviewed by the evaluation team, progress has been made, and CRS 
is working on making things even better: 
“They used to have quarterly meetings to do planning and avoid duplication or scheduling 
conflicts in things like school monitoring. They are now planning to meet more frequently 
with a more strategic focus on using the resources, as they move into the next phase. There 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
58 
were some problems with delays earlier, due in part to slow USDA action with the post‐
election change in administration, and this affected things like scheduling of training, but 
bumps have been ironed out.” 
Evidence from the evaluation suggests that some local government officials would expect further 
progress in coordinating and communicating project activities, increasing the number of joint monitoring 
visits, and preventing any form of duplication. For instance: 
 Government partners at the district level (MEMP and regional circle officials) and the national 
level (DAS) would wish for more coordination with the project teams. One partner in Kandi 
mentioned that he would appreciate greater collaboration when teams are visiting particular 
schools, to ensure these visits do not interfere with other planned activities. Other partners did 
not necessarily share this opinion, but still said they would like information to be shared more 
readily.  
 Stakeholders from DAS also said they would like greater access to program information, as they 
have not been kept apprised of project activities. DAS would like to have their teams, whose 
priorities are not necessarily well represented by district‐level actors, participate in monitoring 
visits. DAS feels they could also assist in addressing challenges faced by schools or staff who 
experience difficulties, to ensure timely and successful implementation of activities. As they are 
a part of the Ministry, DAS officials argue they can leverage internal mechanisms to ensure 
compliance with schools and staff members, which could be an added efficiency measure for 
project implementation. 
4.3.3  Monitoring of Project Activities 
As discussed elsewhere in this report, some of the indicators developed for the McGovern‐Dole project 
may not be entirely appropriate, such as those for child absences due to sickness, attentiveness, and 
minimal acceptable diet (a child complaining of hunger during the school day would be more appropriate 
to measure). Similarly, certain teacher‐level indicators should focus more on the CI and CP classes that 
were most involved in project activities. 
Some of the progress markers established by the project may also have been flawed. For instance, the 
EGRA survey suggests that the EGRA target was greatly overestimated at the baseline stage. Echoing this 
view, one official from WEI indicated that the targets set at the beginning of the project were probably 
“too ambitious given the low quality of the schools, teaching and materials; it would have been more 
sensible to have benchmarks set as gains from the baseline.” Setting different progress markers would 
probably have been more constructive as a way for guiding teachers in their pedagogy and use of 
materials available to them.  
4.3.4  Actioning of Recommendations Made in the MTE Report 
The MTE report identified several shortcomings in the teacher training program management, 
implementation, and follow‐up. Evidence gathered by the endline evaluation suggests that positive 
actions have been taken since the MTE recommended that more focus needed to be put onto teacher 
education and improved practice. For instance: 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   59
 WEI, along with international and national consulting teams, in 2017 developed and 
subsequently validated for application in Benin a “training module on early childhood reading 
and writing” to reinforce the capacity of CI/CP teachers.41
 Teachers and their principals have been trained on new techniques for teaching French 
language and reading on a generally twice‐monthly basis. 
 An observation tool has been developed to assess the effectiveness of this training at the level 
of teachers and students. 
4.4  Sustainability 
This section of the report discusses whether the McGovern‐Dole project is going to last after program 
completion. It provides evidence that speaks to activities or outcomes most likely to be maintained, major 
contributing factors to sustainability, and government’s contribution to future sustainability. 
4.4.1  Activities/Outcomes Most Likely to be Maintained 
McGovern‐Dole will undergo a second phase projected to be implemented in the same schools as the 
first phase, so many activities, including the canteens, will continue. This makes sustainability42 less of an 
issue in the short‐term. Nevertheless, the evaluation considered the degree to which results associated 
with the various components of the project will remain sustainable after completion of the McGovern‐
Dole intervention. 
Feeding Programs 
Principals and teachers generally indicated positive attitudes about the sustainability of the canteen once 
the McGovern‐Dole project is complete. However, 19.1% of principals and 29.6% of teachers reported 
thinking it was little sustainable or not sustainable at all (see Figure 7). Principals reported (58 answers; 
45 respondents) that parents must/will/are prepared to take over (35.6% of respondents), that there is a 
need to set up farmer field schools/gardens (13.3%) and that there is a need/desire to carry on (11.1%) to 
sustain the canteen. However, in a few instances it was noted that parents lack resources/interest (8.9%) 
and that there are insufficient resources to carry on (4.4%). Teachers also reported (202 answers; 
163 respondents) that parents must/will/are prepared to take over (28.8% of respondents), that there is a 
need to set up farmer field schools/gardens (9.2%) and that there is a need/desire to carry on (4.9%) to 
sustain the canteen.  
Teachers’ responses were slightly less positive than principals. Many felt parents lack the 
resources/interest (12.3%) and that there are insufficient resources to carry on (4.9%). They also 
commonly mentioned that support from McGovern‐Dole /CRS will remain critical (9.2%) and investments 
and other contributions will be needed to move forward (6.1%). Finally, the need for help from all levels, 
                                                            
41 FY18 Semi‐Annual Narrative Report CRS BENIN McGovern Dole Award Reporting Period: October 2017‐
March 2018. p. 4.  42 It is important to note that no clear or common understanding of what sustainability means emerged from 
evidence reviewed by the evaluation team. To communities, it tended to mean maintaining the inputs: a 
continuation of the school canteens and other substantive interventions. Sustainability for CRS is a continuation of 
the literacy elements of the program, and other elements that can be supported through community structures, 
such as the APE/AME the hygiene and sanitation elements. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
60 
including parents, APEs, NGOs (through food donations, for instance), leaders/managers, the community, 
and government, to maintain the canteen is also clearly expressed by the principals and teachers. 
Figure 7 
Perception of Surveyed Principals and Teachers on how Sustainable the Canteen will be once the 
McGovern‐Dole Project is Completed 
To CRS staff, it is clear that the school feeding program is not sustainable in its current format. CRS is 
quite realistic; continuing the canteen without imported food items will be difficult, but continuation of 
these supplies is not a sustainable endeavor. CRS is implementing a program that allows it to procure 
food in local markets for the project. While this is a welcome development, as it procures products 
locally, the financial resources to buy the food would still have to be assured. Some parents believe that 
school gardens and/or production fields could be established to supply food to school canteens, but this 
would require testing. CRS’ assessment is that providing food to substitute CRS’ inputs for the school 
feeding activities through food grown in the school gardens is not feasible. Some schools have set up 
gardens, but it does not seem that these are even close to being productive enough. The evaluation team 
was not able to conduct a thorough assessment of the production capacity of schools, but in our opinion 
schools do not generally have sufficient land to grow food in the required quantities to adequately 
substitute CRS inputs for the school feeding activities, and some individuals have mentioned that the 
relations with the community with regards to these school gardens have been counterproductive, with 
produce being stolen from them, and animals left to graze in the gardens. 
Learning 
Principals and teachers were also asked to comment on the extent to which they believed their school’s 
improvements will be sustainable in improving the quality of learning. Overall, responses pointed to 
individuals being somewhat more positive for sustaining the improvement in learning compared to the 
canteen, but there were still many individuals who indicated the results would be little to not sustainable. 
It should be noted that the most common response was “sustainable” (see Figures 7 and 8), pointing 
towards the belief that schools can maintain these results to some degree.  
19.1
17.9
36.2
30.7
10.6
6.7
10.6
17.3
8.5
12.3
14.9
15.1
0 204060 80 100
Principal
Teacher
%
Very sustainable Sustainable Somewhat sustainable
Little sustainable Not sustainable Do not know
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   61
Regarding learning, principals indicated (53 answers; 41 respondents) that there is a desire to put into 
practice methods/systems inherited from the program (19.5% of respondents); however, there is a need 
to continue to deliver training (14.6%), which is somewhat contingent on government support (12.2%) 
and support from/involvement of parents (12.2%). Other ways to help sustain learning results include: 
motivated/knowledgeable teachers sustaining program achievements (9.8% of respondents), 
personnel/managers remaining at the school (7.5%), monitoring (7.5%), and ensuring teaching 
material/booklets are available (7.3%). Teachers also indicated (187 answers; 153 respondents) that there 
is a desire to put into practice methods/systems inherited from the program (15.7% of respondents) but 
indicated maintaining learning quality is somewhat contingent on parent support (15.0%). Other common 
answers provided by teachers included the need for motivated/knowledgeable teachers to sustain 
program achievements (9.2%), ensure monitoring (8.5%), and maintain/build awareness (7.8%). 
Figure 8 
Perception of Surveyed Principals and Teachers on how Sustainable the McGovern‐Dole Project 
Improvements in Improving the Quality of Learning will be Once the McGovern‐Dole Project is Completed 
Health and Hygiene 
Principals and teachers were also asked to comment on the extent to which they believe health and 
hygiene improvements43 will be sustainable once the McGovern‐Dole project is completed. Principals 
commonly indicated (49 answers; 41 respondents) that there is a desire to put into practice 
methods/systems inherited from the program (24.4% of respondents), there is the will/capacity to ensure 
results are sustained (14.6%), a desire to support the health/well‐being of children (7.3%), and that 
children will not stop/have assimilated good hygiene habits (9.8%). Similarly, teachers mentioned (188 
answers; 158 respondents) that there is a desire to put into practice methods/systems inherited from the 
program (20.3% of respondents) and that children will not stop/have assimilated good hygiene habits 
(12.7%); however, they also commonly mentioned the need to maintain/tap on awareness‐building 
efforts (19.6%). In a few cases it was noted by principals that supplies can be bought/are not costly 
                                                            
43 Other infrastructure improvements, such as kitchens, latrines, and storerooms, should be relatively sustainable at 
least for the short‐term. 
19.1
17.9
59.6
41.3
10.6
13.4
6.4
15.6
2.1
2.8
2.1
8.9
0 20 406080 100
Principal
Teacher
%
Very sustainable Sustainable Somewhat sustainable
Little sustainable Not sustainable Do not know
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
62 
(7.3%), yet a few teachers commented on the lack of resources to maintain results. Once again it was 
relayed that the sustainability of health and hygiene improvements depends on the support of all actors 
(such as parents, teachers, principals, community, and government). Unfortunately, in one instance in 
both the principal and teacher survey it was noted that hand‐washing devices have been stolen, which 
could explain why many stakeholders have suggested the need for fences around the schools. 
Improvements such as hand‐washing stations and usage could be sustainable in many cases because they 
are relatively simple and inexpensive, although, to be viable, a school will need to have sources of both 
water and ash, as buying soap may be too expensive for some schools. The hand‐washing surveys show 
good usage of school hand‐washing stations. Overall, 97.9% (47 of 48) of treatment schools visited had 
hand‐washing stations. Only one urban treatment school visited did not have a hand‐washing station. In 
contrast, only 26.5% (12 of 49) of control schools had hand‐washing stations. These results indicate that 
CRS has played a significant role increasing the number of hand‐washing stations at their project’s schools 
and consequently improving the hygiene of the individuals at these schools. Treatment schools with 
hand‐washing stations were also more likely than control schools with hand‐washing stations to have 
both water and soap/ash present at these hand‐washing stations. Enumerators also found indications of 
hand‐washing stations being much more commonly used (for instance, saw students using the stations, 
the presence of used soap or ashes, used water in a bucket or on the floor) at treatment schools (very 
much: 70.2%; somewhat: 29.8%) compared to control schools (very much: 38.5%; somewhat: 46.2%). 
This was further supported by enumerators reporting to directly seeing hand‐washing stations being used 
“a lot” at treatment schools 63.8% of the time, while only 38.5% of the time in those control schools that 
had hand‐washing stations. In a few cases, enumerators indicated that some repairs or upkeep could be 
done to improve the hand‐washing stations. 
Ultimately, 64.6% of treatment schools were found to have soap and water at hand‐washing stations and 
were commonly used by students (see Figure 9), which was significantly greater than only 8.2% of control 
schools (X2
(1) = 31.06, p<0.01). Thus, CRS’ McGovern‐Dole project has greatly improved the application of 
one of the three key hygiene practices, namely hand‐washing. 
Capacity 
The durability of capacity development is a core matter, of course. With respect to teachers, this is 
difficult to assess without stronger indications that the pedagogical learning has been consolidated at the 
individual teacher level and institutionalized in the relevant Ministry units. AMEs/APEs have gained 
motivation, but the canteen has been a major factor here. It would have been instructive to see what role 
they played during the strike to fill in the gap of missing classes, but no data on this are available.  
With respect to literacy, EGRA regressions indicated that surveyed children who indicated that they read 
out loud in class every day were significantly more likely to obtain a passing indicator score than those 
who read once per week (z=‐2.275, p=0.022), once per month (z=2.288, p=0.022), or never (z=2.112, 
p=0.034), with odds ratios indicating that students who read once per week, once per month, or never 
are 1.5 times, 5.3 times or 4.7 times less likely to pass than those who read every day. However, no 
significant effect was found for the question regarding how often they are reading a single reading book 
in class. 
Figure 9 
Percentage of Schools that Were Found to Have Soap and Water at Hand‐Washing Stations that were 
Commonly Used by Students 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   63
As described in Section 4.1 (Table 22), the component EGRA scores show a reasonably good improvement 
from the baseline and as compared to control schools. They are, nevertheless, still very low in terms of 
the targets for each of the components and far from allowing a determination that the children are 
emerging as independent readers. Of 11 components, students in treatment schools fell below the target 
on 9 of them, sometimes considerably. 
A positive aspect going forward is that a second amendment to the McGovern‐Dole project, issued in 
March 2018, is expected to allow “a stronger education component.”44 In this vein, according to WEI, 
participation in reading education by technical MEMP officers has been becoming stronger, and 
treatment/analysis of EGRA data is beginning to be adopted for wider use in the Ministry generally. 
One of the fundamental causes of children’s poor reading levels stems from Benin’s national policy 
requirement that they learn to read in French when most do not speak or understand it. It is promising, 
according to WEI, that the project team is aware of the problem this presents for children and that 
recognized good practice suggests initial reading be done in their mother tongue, with the second 
language introduced orally and gradually through text. For this reason, the second phase of the project 
will give greater attention to the early childhood education level, where the curriculum and pedagogical 
policy are less strict. Introduction to reading in these classes will be in the mother tongue, with French 
introduced orally towards the end of the semester.  
This rethinking is also positive evidence of the project’s flexibility in responding to conditions on the 
ground. Many small additions have been made as the project works with teachers who are “bringing up 
innovations they’ve seen or developed; sharing them with peers, putting them on their phone and 
exchanging. They have been trying different approaches [...] and there has been no problem with that” 
within the project. 
The improvement in scores, even if low in overall terms, is important in showing a positive trajectory and, 
presumably, serving as a motivator for teachers to continue to participate in training and sharing with 
colleagues. It is notable in this respect that WEI has been conducting EGRA tests on a somewhat rolling 
                                                            
44 FY18 Semi‐Annual Narrative Report CRS BENIN McGovern Dole Award Reporting Period: October 2017‐
March 2018. p. 1. 
0
10
20
30
40
50
60
70
Treatment Control
%
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
64 
basis as a means of strengthening MEMP capacity with respect to assessment processes and providing 
feedback to schools and families based on which they can, in theory anyway, take remedial action.  
It is possible that repeated exposure to the testing may have made those children more comfortable with 
it; this may have proven to be a learning tool for them in how to attend to directions, listen for meaning 
in read text, or answer comprehension questions. While this could be seen as a risk to the validity of the 
EGRA in a formal evaluation, its potential pedagogical value for them and teachers could be viewed as 
outweighing this risk.
4.4.2  Contributing Factors to Sustainability 
Regression analysis of the EGRA surveys indicate that parental involvement could contribute positively to 
sustainability. Surveyed children who indicated that they either read or study French at home (z=0.924, 
p=0.002) or have a French manual or book at home (z=0.740, p=0.011) were significantly more likely to 
obtain a passing indicator score than those children who did not (no significant interaction between 
treatment and read/studying at home or between treatment and French manual/book at home). The 
corresponding odd‐ratios indicate that students who read or study French at home have a 1.878 times 
greater chance to pass than those who do not and students who have a French manual or book at home 
have a 1.520 times greater chance to pass than those who do not. However, children indicating that they 
speak French at home did not have a significant impact on whether they obtained a passing score for the 
indicator. 
4.4.3  Government Contribution 
In the endline surveys, principals/teachers often noted that for the results of the project to be more 
sustainable, government would need to help more. There are a series of sustainability reflections to be 
had at a more strategic level. There has been an important political commitment to school canteens by 
the Government of Benin (GoB), and this is important to note. For the 2016‐17 Government Budget, the 
budget for school feeding programs has been increased from 1 billion CFAF to 7 billion CFAF, with the 
goal of reaching 51% of schools by 2021. It is also important to note that due to resource constraints, the 
GoB is deliberately not investing in areas where other partners (notably international NGOs) are 
implementing school feeding programs. This means that unless planned, it will be challenging to ensure 
that the government will take over the school feeding activities as the second phase of McGovern‐Dole 
ends.  
DAS recognizes that there is resistance from the donors to engage in a direct support model to the school 
feeding programs due to the sustainability challenges, such as the continuous acquisition of food items 
for the programs. Having said that, some school directors have stated that they have received notification 
to prepare a request to implement a school feeding program in their schools, with support from the 
government and conversations with the DAS/MEMP noted that the government was open to 
experimenting with innovative solutions. 
There is a window of opportunity here, particularly with the second phase McGovern‐Dole at the 
beginning, to test a few partnership models with the government and other donors. Where possible the 
assumptions regarding the role of school gardens should be tested, with regards to the extent the school 
gardens can produce enough food to sustain or contribute to future school feeding programs; testing 
financing mechanisms to use local procurement to supply food items for the school feeding program, and 
its financing. If the project partners with DAS or other government bodies to test approaches for 
sustainability, it will give a greater chance of mobilizing resources from other sources to continue to 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   65
enhance sustainability for the schools in project areas, and potentially allow for replication of models in 
other areas too. 
4.5  Impact 
This section of the report discusses what are the most important changes for beneficiaries and key 
partners of the McGovern‐Dole project. It provides quantitative evidence that speaks to the degree of 
achievement of the project’s desired effects in terms of overall, strategic and intermediate goals. The 
section also provides elements to inform a judgment discussion on what all of this means. 
4.5.1  Quantitative Evidence on Desired Effects 
The evaluation team carried out DID analyses on endline indicators that had appropriate baseline and 
endline data, including teacher attendance rate, national curriculum usage, active APEs (added), active 
AMEs (added), time spent on literacy instruction, reported child hunger, minimum acceptable diet, 
benefits of education, children’s absences due to illness, children’s ability to read and understand grade‐
level text, and classroom attentiveness. DID essentially uses observational study data and studies the 
differential effect of a treatment on a “treatment group” versus a “control group.” It calculates the effect 
of a treatment on an outcome by comparing the average change over time in the outcome variable for 
the treatment group, compared to the average change over time for the control group or in other words 
the difference between the treatment and control at endline, compared to between the treatment and 
control at baseline. It is intended to mitigate for the effects of extraneous factors that could have caused 
changes in the outcome. 
Table 27 provides the results for all of the DID analyses. The coefficient of “interaction” is the treatment 
effect of the CRS project. There is no need to look at the coefficients of the two dummies variables on 
treatment status and post program. 
Several indicators, such as average teacher attendance rate, national curriculum usage, active APEs, and 
time spent on literacy instruction had high values at the time of the baseline and control schools also 
displayed results similar to the treatment schools. In contrast, the McGovern‐Dole project does seem to 
have had a significant impact on the creation and activity level of AMEs (increased the probability by 
0.266 of there being an active AME at the school).  
The DID was significant for parents reporting if their children were hungry during the school day. Overall, 
surveyed treatment school parents are much less likely than control school parents to report their child 
attending the respective schools being hungry during the school day (however, DID analysis shows an 
increased probability by 0.051 of a yes answer45). Only 7.1% of children attending treatment schools were 
reported by their parents as being hungry, compared to 38.6% of children attending control schools. This 
                                                            
45 If you look at the coefficient of "postCRS" you will see that both control and treatment groups experience the 
dramatic drop in the probability of reporting "yes". When you look at the coefficient of "treated", the negativeness 
means that being in treatment group itself already implies the lower probability of reporting "yes". The NET effect of 
the program is not positive. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
66 
is a drop from the baseline,46 which found 47.0% of treatment parents and 51.6% of control parents 
stated that at least one of their children had reported being hungry during the school day. Thus, it 
appears that the McGovern‐Dole project may be helping to decrease levels of child hunger, but 
statistically results are not as convincing.  
                                                            
46 As mentioned previously in this report, the baseline was calculated differently, if one child in a family was hungry 
the result was positive, and data for all children in the family has not been made available. Also, the values for the 
baseline weighted calculations were 54.7% of treatment parents and 58.1% of control parents. 
Endline Evaluation of CRS MCGOVERN‐DOLE in Benin – Final Report  September 2018 
  67 
Table 27 
Results of Difference‐in‐Difference Analyses for Various Indicators Associated with CRS’ McGovern‐Dole Project in Benin 
Variable 
[1] 
Dependent variable [2] 
Teacher 
attendance 
National 
curriculum [3]
Active APE  Active AME  Literacy 
instruction 
Absences 
sickness 
(girl) [4] 
Absences 
sickness 
(boy) [4] 
Child hunger  Minimum 
acceptable 
diet 
EGRA indicator Attentiveness 
(weighted) 
Benefits of 
education 
(0)  (1)  (2)  (3) (4) (5) (6) (7)  (8) (9) (10) (11) (12)
Post CRS  ‐0.002   ‐0.854 *** 0.042 0.160 0.119 *** ‐0.009 ‐0.020   ‐0.524 *** 0.105 ** 0.057 *** 1.020 *** ‐0.047
  (0.023)    (0.044)  (0.039) (0.082) (0.025) (0.011) (0.011)    (0.016) (0.037) (0.013) (0.309) (0.031)
Treated  0.002   0.079   0.020 0.386 *** 0.029 ‐0.003 0.001   ‐0.051 ** 0.041 ‐0.018 ‐0.096 ‐0.005
  (0.024)    (0.045)  (0.039) (0.082) (0.023) (0.010) (0.011)    (0.019) (0.030) (0.013) (0.304) (0.031)
Interaction  0.012  ‐0.080  ‐0.021 0.266 * ‐0.029 0.012 0.018  0.051 * ‐0.093 0.069 *** ‐0.298 0.175 ***
(0.034)  (0.064)  (0.055) (0.116) (0.034) (0.015) (0.016)  (0.024) (0.054) (0.018) (0.411) (0.044)
Urban/rural  ‐0.014   ‐0.021   ‐0.002   0.020   ‐0.005   ‐0.020 *  ‐0.012   ‐0.007   0.118 ***  0.019   ‐0.353   ‐0.007  
  (0.023)    (0.044)  (0.038) (0.081) (0.021) (0.009) (0.010)    (0.016) (0.035) (0.012) (0.288) (0.031)
Gender child     0.017   ‐0.021   0.006   ‐0.281   ‐0.032  
    (0.011) (0.025) (0.009) (0.204) (0.022)
Class size    0.170 ***
    (0.004)
Active AME  ‐0.002 ‐0.001  
(0.009) (0.009)   
Constant  0.932 *** 0.857 *** 0.937 *** 0.164 ** 0.882 *** 0.062 *** 0.067 *** 0.517 *** 0.600 *** 0.018 ‐1.358 *** 0.449 ***
  (0.017)    (0.032)  (0.028) (0.059) (0.018) (0.007) (0.008)    (0.014) (0.024) (0.011) (0.300) (0.024)
Observations  174   189   191   191   721   704   709   2,804   1,462   2,719   3,293   2,024  
Adjusted R2 ‐0.019   0.808   ‐0.012 0.358 0.046    0.394 0.001 0.043 0.429 0.015
Legend: *** p<0.001; ** p<0.01; * p<0.05;  
Notes: 
1. For each variable in column (0) its coefficient and standard deviation are given. Standard deviation is in parentheses. 
2. Results are based on 5 surveys (principal, teacher, parents, EGRA, attentiveness), also the parents survey has two very different values depending on if the question addressed the actual parent or the children of parents 
(parents commented on all children and each were to count as an independent observation based on indicator definitions). Different values within each survey are because of missing values for a question. 
3. DID can only be done using the principal survey data, as the baseline only asked this question in the principal survey and not in the teacher survey. 
4. Unit of measure: number of students absent due to illness/total number of absent students. Baseline only asked for the number of students absent due to illness in the last month and not the total number of absences, thus 
results here are based on the baseline calculation. Values for girls and boys were asked separately. 
Note: all indicators are binary except for teacher attendance and absences due to sickness. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
68 
The effect of the McGovern‐Dole project on minimum acceptable diet was not significant. As noted 
above, certain factors such as children only eating once or twice per day cannot be controlled by the 
project. 
The DID analysis did find that the McGovern‐Dole intervention had a significant effect on whether parents 
could name three or more benefits of primary education (increased the probability by 0.175). Most 
commonly surveyed parents reported that primary education would help their children find a job, give 
their children a better life and allow their children to care for them. 
The McGovern‐Dole intervention also had a significant effect on whether students, by the end of two 
grades, can read and understand the meaning of grade‐level text (calculated using sections 8, 9 and 10 of 
the EGRA test) (increased the probability by 0.069). This illustrates that the McGovern‐Dole intervention 
is in fact improving the ability of students to read and write French. 
4.5.2  Discussion 
Across the several project implementation areas, the school feeding program has probably had the most 
obvious impact. It has changed, at least for the period of the project, the nutritional and food security of 
the students in participating schools and, in consequence, appears to have had an influence on health 
and attendance.  
With regards to literacy levels, although the project has not achieved the target results it had set out to, 
the students have outperformed control school students, and demonstrated improvement from the 
baseline. Having said that, the various elements that the project does not control (such as teacher 
retention and number of students per class) may well have had negative impacts on the results of the 
project.  
There is some indication of ownership exhibited by the AMEs/APEs, as these groups have played a key 
role in McGovern‐Dole implementation. As noted in Section 4.4, however, the durability of this as an 
impact will be judged a year or so down the line. 
The McGovern‐Dole intervention does not appear to be having a statistically significant influence on 
attentiveness. However, as discussed below, because there are so many additional factors that can affect 
attentiveness outside of being better fed, such as class size, these results are suggested as cautionary. 
Beyond the expected finding that increases in class size correlated negatively with attentiveness, the 
observation‐based attentiveness data overall were not especially revealing of patterns. The differences 
between variables are either not major or ambiguous. Even the class‐size variable should be approached 
with caution; for instance, there seemed to be a threshold before which the numbers did not make a 
difference.47
The link between not being hungry, paying attention and learning should not be drawn too finely. While 
children having meals at school correlated positively with attentiveness—which is logical, assuming 
hunger leads to distraction—there were clearly other factors contingent on being a McGovern‐Dole 
school: better trained teachers, materials, classroom and WASH infrastructure that could be making a 
                                                            
47 There is a methodological issue here too: the larger the class size, the more difficult it probably was for the data 
collectors to focus on any single child’s behavior, especially the more subtle ones used to grade attentiveness. Also, 
please note that there were class size differences at baseline, as control schools had an average size of 183 students 
and treatment schools 282 students. 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   69
difference to their engagement with the lesson. The canteen may have brought children to school as the 
data show, but there were also indications it might not have kept them there all day; children reportedly 
going home after lunch and not returning would cut into lesson time. Finally, the control schools did not 
show a correction between eating and attentiveness, although, without canteens, it was not possible to 
know if children had eaten or not.  
The issue of gender is also important here. While boys were found to be “slightly less attentive than girls” 
in the McGovern‐Dole schools, the opposite was true for the control schools. Factors such as age, seating 
arrangements, teachers’ (and observers’) attitudes and expectations about how boys and girls “should” 
behave could be intervening variables to the extent girls and boys in Benin differ in how they engage with 
learning irrespective of the meals, numbers, content of the lesson. These are issues warranting further 
follow‐up with respect to continued program delivery.  
The positive correction between THRs and attentiveness was not explained in the observation data; 
presumably, the effect would depend on how the rations were used, for whom and when. Here, too, 
interactive effects may matter: parents who received rations may have been more likely to support their 
children’s learning, make sure they attended, perhaps even helped with homework. There was, in fact, 
some data suggesting parents were sending children to school even when ill to collect the rations.  
A small non‐random sample of students answered a grid of questions on their class behavior, and while 
these numbers are not statistically valid, they are indicative perhaps of a trend that is positive. Both boys 
and girls appear to answer and listen to teachers. Neither are prone to falling asleep in class and both, 
especially girls, are rarely punished, which might have been a sign that teachers found them disruptive in 
being inattentive or losing track of the class activity. Somewhat less positive, perhaps, were the responses 
indicating children did not ask questions in the class, a sign of engagement and more child‐centred 
teaching. Girls especially appeared to be reluctant. 
On this last point of gender, among other things, impact concerns longer‐term changes in the well‐being 
of beneficiaries and in this respect, a crucial concern is the effect the McGovern‐Dole project is having on 
girls and women. Overall, it was difficult to draw strong conclusions about how gender is being handled in 
the project. The quantitative data provide some indication, but in general answers from the survey and 
the FGDs did not manage to deeply interrogate the quality or impact of actions with respect to if and in 
what ways girls and women might be benefiting differently from boys and men. It may be the case that 
stakeholders themselves are not clear. 
For instance, teachers are reported as not knowing the reasons for students’ absences. Thus, while fewer 
girls reportedly miss school because of “work,” it is not clear what this includes—only paid work, or also 
perhaps sibling care and household duties. For those schools that have toilets, but do not separate them 
by gender (some 50%), it will be important to track how girls are being affected; what they think about it 
and how they are being “counted” if they miss school during menstruation (that is, as sick, just missing, or 
perhaps not at all). Similarly, it will be important to track whether girls are making equal use of the 
canteens, and whether they are engaging equally in the reading classes. These will be data important for 
CRS and WEI to track in monitoring the equity as well as the effectiveness of the project and planning the 
second phase.  
Again, based on a non‐random sample, grid answers did show a perhaps troubling trend towards girls 
being appreciably less happy than boys at being in school. They indicated being less comfortable asking 
questions and felt much less than boys that they were helped by the teacher. Also, they were strongly 
ambivalent about their enjoyment of reading and writing, and especially doubtful about writing; boys, on 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
70 
the other hand, were sure they enjoyed both. In general, the girls also felt their latrines were dirty and 
unpleasant to use more often than boys, although there is also a significant number of boys with similar 
observation. The data were not clear as to why these observations might have been the case; it is an 
implied pattern warranting further attention by the McGovern‐Dole project perhaps.48
On a more positive note, the principal survey indicates that the project has encouraged the formation 
and activities of AMEs in a significant way, moving from a baseline of 54% in treatment schools to 97%. 
This is an important pro‐gender equality finding insofar as women are being given a larger space for 
expressing, at least potentially, their ideas, priorities and concerns, as exercising decision‐making 
authority. It is also important to the extent women are more likely than men to notice if and how girls 
might be missing out of benefits and take action to address that. 
Finally, as an overarching point of analysis, while schools are the ultimate focus of educational 
improvement, they are also part of a wider system that both supports and limits effectiveness for 
children. In this sense, efforts to reform educational practice are more relevant when they focus at a 
higher level. Focusing on a subset of schools based on areas of higher food deficit or insecurity rather 
than the full district, as is the case with the McGovern‐Dole project, is counterproductive to a certain 
extent, due to issues like the extremes of overcrowded and unfilled classrooms; the equitable distribution 
of materials; and managing teacher deployment so as to retain trained personnel in the local system are 
difficult to handle at the level of the individual school.  
At the same time, opportunities may be missed through a school‐based focus. For example, while 
teachers are beginning to share ideas and experiences about their teaching and “what works” in their 
literacy lessons, mechanisms for sustaining such exchanges are most useful if facilitated systematically 
across several schools. Related to this, too, district supervisors who, according to one respondent, may 
appreciate and take up the new teaching methods, materials, etc., are hindered under the current 
arrangement in being able to act on the insights, knowledge, skills and resources for only some of their 
schools, leaving out others perhaps in equal need. In a meeting held in May 2018 between WEI and 
INFRE, INFRE Director expressed desire to work with Ministry of Education to adopt mechanisms, namely, 
Communities of Practice, introduced by the project, integrating them into systematic teacher exchanges 
across schools—beyond the life of the project.  
5.  CONCLUSION AND LESSONS LEARNED 
5.1  Conclusion 
Overall, the McGovern‐Dole project was relatively successful in meeting or almost meeting its targets. 
Since the mid‐term, and particularly as a result of adjustments made after the MTE, there appears to 
have been an improvement in children’s literacy, as evidenced by EGRA scores. In fact, this improvement 
may even have been suppressed by the effects of the strike. Parents in the treatment area also seem to 
be much more aware of the benefits of education, likely as a result of McGovern‐Dole activities. 
                                                            
48 It would be important for CRS to dig deeper into the above findings related to gender, through focus group 
discussions with children. Findings will help CRS devise some gender‐sensitive programming aspects to help increase 
the quality of education for both boy and especially girl students. This will likely include specific gender‐sensitive 
training for teachers. Also, incorporating one or two indicators that relate to the happiness, comfort and enjoyment 
levels of students at school should be considered for the McGovern‐Dole Phase II.  
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   71
Continued focus on raising parents’ awareness and increasing their involvement in their children’s 
learning could further enhance results. The AMEs/APEs are also energized and engaged in the project. 
They are a critical element in the implementation of the project and in bridging the relationship between 
the school and communities. 
Infrastructure projects have also been better supported by technical staff since mid‐term. Infrastructure 
improvements and the implementation of hand‐washing stations were well received by parents and 
community stakeholders and have also likely had a moderate effect on the well‐being of children. The 
canteens also appear to have been successful in improving children’s nutrition and reducing their hunger, 
and future efforts should focus on increasing the sustainability of canteen services. 
Feedback from all stakeholders was very positive and, overall, project participants seemed very satisfied 
with the assistance they had received. There were some suggestions for improvement and better 
coordination with partners, but the main comments were about expanding the project to include all the 
other classes for literacy training and the wider project. This is itself is a testimony of stakeholders’ 
satisfaction with the project. The SILCs are very valued, although their contribution to children’s literacy 
and school attendance is indirect and complex to assess.  
However, in some cases, particularly children’s literacy, results fell short of expectations. As discussed in 
Section 4, some of these less impressive results could be due to the fact that several of the indicators may 
not have provided an adequate assessment of program achievement. There are also important elements 
that have affected the project, but that are not under its direct influence. Key among these are the high 
number of students enrolled in treatment schools, and the rotation of teachers.49 Both these factors have 
had a negative effect on the efforts of the project. For example, literacy training given to a teacher who 
subsequently leaves a treatment school is a lost investment for the project, and high numbers of students 
in a classroom correlate positively with decline in students’ attention. 
5.2  Lessons Learned 
Complex socio‐economic issues in target communities can have a direct influence on the outcomes of a 
food‐for‐education project. Such issues include food security, capacity of teachers, school infrastructure, 
and water and sanitation in schools. While the McGovern‐Dole project has addressed these issues, the 
endline evaluation has identified others as gaps—the impact of malaria is a key one, as well as matters 
related to gender broadly framed and, more specifically, menstrual hygiene as an important cause of 
student absence. The incidence of malaria is a larger communal issue, and it remains to be seen if there 
are opportunities for the next phase of the project to leverage partners’ capacity and resources to 
subsidize and distribute bed nets in program areas. Issues related to girls and women are clearly within 
the mandate of the project. 
Physical punishment in school, such as the use of the whip, and student bullying, are also factors that may 
influence the outcomes of a food‐for‐education project. In the qualitative exercises, particularly with girls, 
there was consistent mention of the pervasive use of whips in physical disciplining techniques. Out of the 
four schools where the qualitative exercises were conducted, in three schools there was a specific 
mention of being hit by a whip, and in one school there was a mention of being hit. The evaluators 
believe these references warrant more investigation. A first recommended step would be for CRS, from a 
                                                            
49 Teacher rotation includes both teacher turnover (e.g., a teacher leaves a school after a short time period because 
they were not paid or have found different employment) and allocation of teachers to different schools by the 
Ministry of Education. Principals provided general reasons for teacher rotation at their school. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
72 
management perspective and WEI from the pedagogical, to confirm in detail the scope, severity and 
regularity of the “punishment”. Both are duty bearers in this case and from the perspective of the 
Committee on the Rights of Children, it will be important to distinguish on a continuum between actions 
that constitute abuse, especially if there are indicators of sexual abuse since girls appear to be particularly 
at risk, and those that are closer to over‐harsh corporal punishment. In all cases, intervention is 
warranted, but the nature of the punishment will guide follow‐up action and who should be involved. To 
the extent the behavior is mild and broadly acceptable in the community, then the intervention should be 
set within a participatory learning framework: parents, principals, teachers and students discussing the 
causes and consequences of using violence as a means of managing the classroom, options for use of 
positive discipline, and agreement on a plan for implementing and monitoring change. There could be 
also be a specific training conducted with teachers focusing on disciplining alternatives and classroom 
management; a training that would train teachers to manage student’s behavior through positive 
engagement and limit the use of physical disciplining. Altogether interventions should strive to support a 
change in the school culture towards the concept of child‐friendly schools. Working with other 
stakeholders, such as UNICEF, maybe an option. Where serious abuse is happening, of course, there 
should be legal action. 
Several factors that have a strong influence on a food‐for‐education project can be outside its scope of 
action. One is the issue of teacher rotation and specifically that when teachers are reallocated to other 
schools, the training the project has given them does not have the planned impact on McGovern‐Dole ‐
targeted schools. Another is the insufficient number of teachers overall, leaving it to communities to hire 
and pay locally based teachers—costs that can be prohibitive for families with several school‐aged 
children. 
Financial constraints can be a major obstacle to school enrollment in a food‐for‐education project, as 
evidenced by comments recorded in the majority of FGDs with families of children who had not enrolled 
in school. The endline evaluation found indications that these children were important labor resources, 
and that families could not afford to have all their children go to school when they were required to do 
work at home and in the fields. This was not observed in FGDs conducted in Soukarou, a community 
where children who were not in school were members of migrant families from Niger. These were 
farming (rice) and animal herding (cows) communities where parents also had no schooling and did not 
speak the local languages. They did not understand the radio programs, for example, and therefore had 
not understood the broadcast messages about the importance of education. However, they did 
demonstrate a very coherent and well‐articulated understanding of the value of education and were 
being held back by economic activities that were human labor‐intensive. These families preferred their 
children working at home, in the fields, or herding. There is also a culture of early marriage among them, 
with some expressing views such as “for the girls who will marry soon, why should we send them to 
school now? It is not important to do so.” 
In some cases, the level of enrollment and classroom numbers can influence the quality of the 
educational experience in communes targeted by a food‐for‐education project. Through the McGovern‐
Dole school feeding program, CRS creates a powerful incentive for families to enroll their children in 
project schools, even if these are not the closest schools to their residence, creating at times an irregular 
and PTR. 
In a food‐for‐education project, food preparation in the school canteen can prove to be difficult at times, 
due the small supply of AME members available daily to cook meals for the students. Food preparation is 
an ongoing activity for which women participating in the McGovern‐Dole project received cooking oil as 
an incentive. While the value of volunteer contributions has been recognized by the project, further 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   73
thought should be given to the fairness of this approach if implementation is to be maintained. Perhaps 
there would be grounds to provide incentives that are more commensurate with the demands made on 
AME members who agree to spend part of their days preparing meals for children in schools. 
6.  RECOMMENDATIONS 
1.  The EGRA data (or other data) should not be used for the baseline undertaken for the second 
phase of McGovern‐Dole (McGovern‐Dole FY17) if DID is going to be used, as doing so could limit 
McGovern‐Dole FY17 to sampling the same 49 to 50 schools as were sampled for the first phase 
(McGovern‐Dole FY17). Including an appropriate proportion of urban and rural schools in the 
sample would also increase the reliability of results; although it was not found that either rural or 
urban schools consistently performed better. Furthermore, examining the size of schools (school 
population) when choosing control schools for McGovern‐Dole FY17could help to increase the 
reliability of future results. 
2.  Continuing to sample the same control schools in McGovern‐Dole FY17as in McGovern‐Dole FY14 
could be unethical. Sampling takes at least a day of a school’s time, placing a burden on schools 
that receive no benefits from the project. One principal of a control school noted, “It’s been four 
years since a CRS team has visited us, and there has been no follow‐up,” suggesting that control 
school principals may believe they are being surveyed to see if they can be included in the project. 
Similarly, many control school principals also expressed the desire to have a canteen at their 
school. Therefore, the program should consider the possibility of identifying a new set of control 
schools, or at least including a significantly higher number of urban control schools. If the same 
control schools are sampled, CRS should consider providing some sort of incentive to these schools 
for their time and involvement and making the reason for the surveys very clear.  
3.  The project should continue to use EGRA as a tool for formative evaluation of the McGovern‐Dole 
pedagogical interventions, and to refine it as a way of supporting teachers’ professional 
development to better gauge the link between what they do in teaching reading and the reading 
outcomes children realize. 
4.  Reported literacy values were higher from surveyed control school principals and teachers. 
Therefore, although already high, government officials could be encouraged to put effort into 
further increasing the use of national curriculum and literacy instruction in treatment schools. The 
percentage of students with French manuals and activity books could also easily be increased, 
which would promote greater learning. 
5.  Although McGovern‐Dole has little control over the efforts of individual teachers, it could be 
beneficial to think of ways to incentivize hard work. Several teachers mentioned that they would 
like to receive canteen services and, considering that the number of meals the project provided 
was lower than the target, this could be an easy and inexpensive and way to improve teachers’ 
motivation. It could also help to increase teacher retention—an important objective, considering 
that almost 20% of treatment school principals reported that, on average, teachers remain at their 
school for fewer than six months. Building housing for teachers was expected to increase teacher 
attendance, but it has not yet been undertaken. Currently, an amendment is under review by USDA 
that would provide the project an additional year to complete housing construction (and other 
infrastructure work). Housing construction or a housing subsidy (although the latter would be less 
sustainable) could be considered for McGovern‐Dole FY17 and indicators tracked to determine its 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
74 
effect on attendance and retention. The project could also survey teachers to ask them what would 
increase their retention and attendance.  
6.  Extending meals to teachers and offering housing incentives should also reduce teacher rotation in 
project schools. The evaluation team could not identify the mechanisms or reasons for teacher 
rotation, although the clear lack of teachers in the education system could be the root cause. If 
teachers have a say in the rotation process (for instance, if they apply for rotation themselves) 
project incentives for teachers could reduce such rotation. If there are other factors influencing 
teacher rotation—such as those related to DDEMP planning and allocation of resources—CRS could 
lobby the government to reduce teacher rotation in project areas, attempting, for example, to 
keep teachers in a given location for at least two school years.  
7.  A further way to mitigate the effects of teacher rotation would be to develop an intensive short 
course on the key aspects of literacy training used by the program for new teachers. This training 
could take place at the beginning of the school year or at whenever in the school calendar there is 
a higher rotation of teachers. This could bring new teachers up to speed with techniques and 
materials and enable the teacher to start implementing them in treatment schools as early as 
possible. 
8.  Because McGovern‐Dole FY17is planned for the same schools that were involved in McGovern‐
Dole FY14, sustainability is not currently a key issue. However, the findings from the principal and 
teacher surveys indicate that promoting and perhaps intensifying activities such as school gardens 
could help to promote canteen sustainability. Also, CRS could work with stakeholders at all levels 
devise plans for canteen sustainability. Sustaining learning outcomes could be promoted by 
providing a space for motivated and knowledgeable teachers to sustain program achievements or 
by promoting a peer system of training. A first step in achieving this is through the identification 
and recognition of “Master Teachers,” which the project is already doing. Health and hygiene 
sustainability could be promoted through awareness activities and setting up awareness and 
monitoring communities. 
9.  Developing a pathway for sustainability could involve testing various options, such as school 
gardens, local food procurement, or financial contributions to the schools. This would need to be 
tested locally to develop a “proof of concept” that could be replicated at a significantly larger 
number of schools. The KIIs with DAS identified a great degree of openness to this type of an 
approach. This enthusiasm could be leveraged by negotiating with the government to gradually 
overtake McGovern‐Dole FY17 activities, or other such opportunities. 
10.  Two of McGovern‐Dole program goals—increasing student enrollment and increasing student 
attentiveness—are, on the surface, contradictory. Treatment schools and classes sampled for 
attentiveness had an average class size of 43.1 students and ranged from 13 to 153 students. 
Among the 48 schools sampled for the evaluation, the average class for CI had 68.5 students (46.7 
for control classes), and for CP, 53.9 students (36.3 for control classes). In treatment schools, there 
were 28 classes with more than 100 students (3 in control), including one class with 166 students, 
and one class with 231 students. The concept and measure of attentiveness are complicated. Also, 
while very large classes may negatively affect attentiveness (whether students are better fed), PTR 
is relative and international data do not necessarily link numbers to reduced learning outcomes 
(attentiveness is not generally used as a variable). Having more students attend school is clearly to 
be desired—their distribution across classrooms and among schools is a larger matter of 
government policy, as is training teachers to group children more effectively, the use of teacher 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   75
aids, involving parents, and self‐directed materials (along the lines of distance learning). The 
consequences of increased class size, therefore, need to be examined and any indicators 
associated with attentiveness re‐thought. The consequences of using PPS sample selection and 
consequently selecting larger schools for evaluation should also be revisited. 
11.  Recognize, and find ways to promote, those McGovern‐Dole “seed” initiatives that have 
implications for long‐term sustainability of pedagogical capacities. As the data indicates, there will 
be teachers “somewhere in the system with better skills and confidence” and the project has “tried 
to make them feel they matter” by taking up their ideas and helping share them; creating stable 
mechanisms for continuing to do so—such as school clusters, professional development centres for 
teachers, and principals’ associations—is essential. 
12.  Consider using those most competent teachers from the 48% who have been serving two to five 
years and more as mentors or trainers to newer colleagues. 
13.  The survey data suggest teachers and principals have at least a general concept of what effective 
reading instruction is, as well as what constitutes “hard‐working” teachers. It was not clear what 
indicators were being used, however. Planning a series of “reflection workshops” around these and 
similar instances of professional judgment specifically to explore what they mean, to share and 
interrogate criteria, and to begin to collect and test data to confirm and plan remedial strategies 
around them could be an effective way of sustaining the progress made by McGovern‐Dole by 
building capacities within schools.  
14.  Pursue the plan referenced in the 2016 Progress Report to conduct post‐training assessment of 
teachers as they work to apply their learning in the classroom. This is essential both to ensure the 
relevance of the training design and the effectiveness of its outcomes, as well as to enable teachers 
to consolidate their learning and moving them from “knowing what” to “knowing how.” Developing 
such assessment in collaboration with teachers themselves through peer support groups, would 
strengthen the process even further.  
15.  Students often mentioned teachers hitting them as what they most disliked about going to school. 
It will be important to investigate physical punishment of students and to introduce components 
on disciplining students into the teachers’ literacy training curriculum to eliminate these practices.  
16.  A sustainability model for school meals should be tested in select schools, in partnership with the 
DAS/Government of Benin, and CRS. Various models—including food production via school gardens 
and local procurement could be tested. If a model is tested and proven to work, it will be 
considerably easier for the government, in partnership with CRS, to seek assistance from partners 
to support the sustainability of the school feeding project, rather than simply supplying food items. 
17.  Constructing or improving water points/wells/water systems still seems to be a very important 
concern of treatment school principals and teachers, and McGovern‐Dole FY17should continue 
improving these systems. 
18.  Improving the functionality and cleanliness of latrines and increasing the number of female‐only 
latrines would be beneficial. This could be done by creating a development and maintenance plan, 
with assistance from the schools and parents’ associations. 
September 2018  Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report 
76 
19.  CRS should consider including the construction of classroom modules as part of McGovern‐Dole 
FY17’s infrastructure investments, as requested by some schools. 
20.  Indicators for McGovern‐Dole FY17should be modified to focus on achievements that can be more 
directly attributed to the program. For example, if training is essential for CI and CP teachers, 
indicators for using training methods and national curricula, and for the amount of time spent on 
literacy training, should focus on these grades. Currently, only the use of the national curriculum is 
focused on these grades. However, if CRS has a reason for examining all teachers—such as the 
rapid movement of teachers between grades—then indicators could be left as they are. 
21.  The theory of change and logic model for McGovern‐Dole FY17should more clearly separate inputs 
and, especially, expected outcomes related to learning from those related to nutrition and health. 
While the correlation between a child’s well‐being and learning is unquestionable at the ultimate 
outcome level, at the immediate and intermediate outcome level the integration of the streams 
obscures the learning pathway and diminishes the likelihood of successful change. 
22.  Unfortunately, McGovern‐Dole did not have an impact on decreasing the percentage of student 
absences related to sickness; however, the values at baseline were already low. Even more 
important, according to the parent survey, around 60% of children were sick due to malaria, a 
mosquito‐borne illness that neither better health and hygiene nor better nutrition can effectively 
mitigate. Measures to reduce malaria include such things as reducing mosquito breeding areas, 
insecticidal sprays, and bed nets, which were not part of the McGovern‐Dole project. Thus, the 
indicator for illness‐related absences is not effectively measuring results that can be attributed to 
McGovern‐Dole. Another approach for future programming would be to measure child absences 
due to work. It would be important to see if these absences are consistent across the school year 
or related to the agricultural season. Work‐related absences appear to be more common than 
health‐related ones and improving parent’s knowledge of the benefits of primary education could 
help decrease them. 
23.  With respect to radio messages and potential contamination of the indicator on the benefits of 
primary education, overall, 75.0% of surveyed parents from treatment schools (rural, 76.0%; urban, 
68.1%) stated that they had heard these radio messages, but 30.7% of parents from control schools 
(rural, 29.9%; urban, 35.0%) also heard them. This should be taken into consideration when 
developing indicators, especially if radio messaging is one of the main influencing methods. 
However, influencing parents from control schools to send their children to school is a positive 
thing. 
24.  CRS should clarify the type of indicator and report final values appropriately. Some indicators are 
cumulative, some are measured by semester, and some are percentages—they should not be 
added together for a final value. For example, for the indicators that had “double‐counted” in 
Table 25 this refers to the fact that the total indicator value would better refer to the value 
obtained during the last reporting semester. For example, for the indicator “22,621 males 
benefiting directly from USDA‐funded interventions” the total to date would better read “24,344” 
rather than “131,474,” which is derived from adding all semesters together. There is double‐
counting because each semester many of the same students would be counted as enrolled and 
receiving feeding services from the canteen. However, some beneficiaries may only have been 
counted once, such as those receiving nutrition training. A future recommendation would be to 
ensure that each student receiving feeding services is only counted once. In addition, the unit of 
measurement (number/percentage) was sometimes changed so that some indicators were 
Endline Evaluation of CRS McGovern‐Dole in Benin – Final Report  September 2018 
   77
reported in different units than specified—for example, the indicator for “students in 144 schools 
assessed using Early Grade Reading Assessment tool” was reported as the number of classrooms 
assessed.  
25.  Meeting targets for indicators such as the percentage of parents and students in target schools 
who achieve a passing score on a test of good health and hygiene practices, or for students who 
achieve passing EGRA scores, will likely require the constant and consistent implementation of 
activities that can improve them.  
26.  Future programming should consider reducing the use of semolina (couscous) in school meal 
programs. If using this food item has already been agreed upon with USDA, perhaps a larger 
amount of it could be monetized and more of the other food staples kept for the project. Most 
stakeholders have spoken negatively about semolina, which is not appreciated by the students, and 
is more vulnerable to insects and other contaminants.  
27.  The most commonly heard request or recommendation from all stakeholders was to expand the 
scope of the program. For example, many teachers requested that more teachers and students 
from other classes receive training; many teachers also asked to be included in the meal program. 
CRS should conduct a study to determine the costs of additional teacher training and meals, 
specifically because the most expensive element of the project (the feeding program) is apparently 
being extended to students in higher classes. WEI also believes that there would be a generational 
impact if other classes in the school were included in the project. 
28.  Greater collaboration with government is recommended, particularly at local, with the CRPs and 
central level, with the DAS. Information sharing, greater coordination in implementation, and 
implementing more joint monitoring field trips would be beneficial and develop a greater sense of 
collaboration.  
29.  If the project chooses to invest more in the school gardens, particularly to grow enough food to 
partially cover the deficit left in some schools at the end of McGovern‐Dole FY17, it should include 
an education component that would allow students to contribute to the school gardens and link 
this contribution to pedagogical outcomes, such as learning about farming or the natural 
environment. 
30.   A limited group of mothers takes responsibilities for all the cooking for the children during the 
school year. They receive cooking oil as an incentive, but this is not an adequate or equitable 
incentive commensurate with the work done. More parents could be brought in to assist with the 
cooking, perhaps by rotation, or more incentives could be given to the mothers who cook for the 
children on a regular basis.  
31.  Some of the members of the AMEs/APEs mentioned that they would benefit greatly from exchange 
visits with other schools. This could be a powerful mechanism for replicating best practices and, if 
logistically feasible, the project could use it as a capacity‐building tool. There were exchange visits 
organized among AMEs across school communities in May 2018, and this would be a relevant 
practice to carry forward.  
 
Endline Evaluation of Catholic Relief Services’ Food for Education (FFE) 
Benin Project
APPENDICES TO THE FINAL REPORT 
Presented to Catholic Relief Services
Submitted by Advisem Services Inc.
Ottawa, Canada
September 2018

Endline Evaluation of CRS FFE in Benin – Final Report September 2018
i
CONTENTS 
Abbreviations, Acronyms and Symbols ...........................................................................................................iii
Appendix 1 – Terms of Reference.................................................................................................................... 1
Appendix 2 – Evaluation Matrix..................................................................................................................... 20
Appendix 3 – List of Documents Reviewed ................................................................................................... 25
Appendix 4 – EGRA Survey............................................................................................................................. 26
Appendix 5 – Parent Survey........................................................................................................................... 33
Appendix 6 – Principal Survey........................................................................................................................ 43
Appendix 7 – Teacher Survey......................................................................................................................... 55
Appendix 8 – List of Participants in Key Informant Interviews...................................................................... 64
Appendix 9 – Guide for Interviews with School Directors............................................................................. 66
Appendix 10 – Guide for Interviews with Heads of School Districts............................................................. 68
Appendix 11 – Guide for Interviews with Senior Policy-Makers, Program Managers and 
implementers of FFE ...................................................................................................................................... 70
Appendix 12 – Guide for Focus Group Discussions with Teachers............................................................... 73
Appendix 13 – Guide for Focus Group Discussions with Students................................................................ 75
Appendix 14 – Guide for Focus Group Discussions with Best Performing Students.................................... 78
Appendix 15 – Guide for Focus Group Discussions with Parents.................................................................. 80
Appendix 16 – Guide for Focus Group Discussions with Parents of Children not attending School............ 82
Appendix 17 – Guide for Focus Group Discussions with Members of Parent-Teacher Associations, 
Savings and Internal Lending Communities and Canteen Management Committees................................. 84
Appendix 18 – Hand-Washing Station Observation Tool .............................................................................. 86
Appendix 19 – Physical Environment Observation Tool................................................................................ 88
Appendix 20 – Attentiveness Observation Tool ............................................................................................ 90
Appendix 21 – Complete List of Intervention Schools and Control Schools................................................. 92
Appendix 22 – Details on Surveys done by the Evaluation Team ............................................................... 100
Appendix 23 – Checking for Bias in Data Collection.................................................................................... 108
Appendix 24 – Indicator Calculations.......................................................................................................... 129
Appendix 25 – Field Mission Schedule ........................................................................................................ 136
Appendix 26 – Performance Measurement Framework and Information Available at the Inception 
Report Stage and before Field Work ........................................................................................................... 138
Appendix 27 – Terminology and Definitions ............................................................................................... 144
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
ii
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
iii
ABBREVIATIONS, ACRONYMS AND SYMBOLS 
ACE Agent contractuel de l’État [government contractor]
AME Association des mères d’élèves [school mothers’ association]
APE Association des parents d’élèves [parent-teacher association]
CCS Head of school district [chef de circonscription scolaire]
CFAF African Financial Community franc
CI Cours d’initiation [introductory class]
CP Cours préparatoire [preparatory class]
CRS Catholic Relief Services
DAC Development Assistance Committee
DANA Direction de l’alimentation et de la nutrition appliquée [Food and Applied Nutrition
Directorate]
DAS Direction de l’alimentation scolaire [School Feeding Directorate]
DEP Direction de l’enseignement primaire [Primary Education Directorate]
DID Difference-in-difference
DIP Direction de l’inspection pédagogique [Pedagogical Inspection Directorate]
EGRA Early Grade Reading Assessment
FFE Food For Education
FFW Food for Work
FGD Focus group discussion
GoB Government of Benin
INFRE Institut national pour la formation et la recherche en éducation [National Institute for 
Training and Research in Education]
INGO International non-governmental organization
IPTW Inverse probability weighting
KII Key informant interview
M&E Monitoring and evaluation
MEAL Monitoring and evaluation, accountability and learning
MEMP Ministère de l’enseignement maternel et primaire [Ministry of Preschool and Primary 
Education]
MGD McGovern-Dole
MTE Mid-term evaluation
NGO Non-governmental organization
ODK Online Data Kit
OECD Organization for Economic Cooperation and Development
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
ABBREVIATIONS, ACRONYMS AND SYMBOLS (CONTINUED) 
iv
PMF Performance measurement framework
PMP Performance monitoring plan
PTA Parent-teacher association
PTR Pupil-teacher ratio
RNG Random number generator
ROA Rapid organizational assessment
SILC Savings and internal lending community
SO Strategic objective
THR Take-home ration
ToC Theory of change
ToRs Terms of Reference
UN United Nations
UNDP United Nations Development Programme
UNICEF United Nations Children’s Fund
US United States of America
USAID United States Agency for International Development
USDA United States Department of Agriculture
WASH Water sanitation and hygiene
WEI World Education, Inc.
WFP World Food Programme
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
1
APPENDIX 1 – TERMS OF REFERENCE 
The following draft Terms of Reference were issued by USDA in September 2017.
I. Purpose of the Terms of Reference 
The purpose of these Terms of Reference is to describe the tasks and responsibilities of an external 
consultant or firm to conduct an endline evaluation of the McGovern-Dole (MGD) International Food for 
Education (FFE), implemented by Catholic Relief Services (CRS) in Benin in partnership with World 
Education, Inc. (WEI).
II. Project Background 
Catholic Relief Services in Benin (CRS Benin) is implementing a four-year, $19,080,649 United States 
Department of Agriculture (USDA) funded McGovern-Dole International Food for Education and Child 
Nutrition Program (FFE) to improve literacy and alleviate hunger for approximately 43,804 primary 
school-age children attending 144 primary schools in four communes in the Alibori and Borgou 
Departments of Northeast Benin. 
The project, which started on September 30, 2014, and continues through September 30, 2018,
coordinates activities with several education related ministries and departments, works with 
212 teachers, 97 parent-teacher associations and 97 mother associations to implement the program, 
including the distribution of 4,420 million tons of commodities (soy-fortified cornmeal, vegetable oil, 
lentils and milled rice). 
The main aims of the project include the following: (i) to improve the literacy of school-age children; (ii) to 
improve the quality of literacy instruction by training teachers in early grade reading methodology; (iii) to 
improve student attendance by providing take-home rations for students; (iv) to increase government 
capacity and ownership by working with the Ministry of Education; (v) to improve the physical learning 
environment by building or rehabilitating kitchens, storerooms, latrines and water stations; and (vi) to 
improve health and dietary practices and raise awareness of the importance of education through radio 
broadcasts.
The official launch of the FFE Project was at the end of March 2015 and the go-ahead to start activities 
was given after the first semester of Fiscal Year 2015. Canteen activities began October 26, 2015 in some 
schools in the intervention area. CRS and WEI have involved several governmental and non-governmental 
stakeholders, including the Ministry of Preschool and Primary Education (MEMP); the School Feeding 
Directorate (DAS); the Food and Applied Nutrition Directorate (DANA); the National Institute for Training 
and Research in Education (INFRE); the Primary Education Directorate (DEP); the Pedagogical Inspection 
Directorate (DIP); and the Ministry of Health.
Despite some challenges including strikes and impacts of political elections, overall, the project has 
operated in a favourable social and economic context, with the support and involvement of local 
authorities and the community. In order to materialize the theory of change described in Appendix A, the 
project carried out the following activities, led by either CRS or the sub-recipient World 
Education, Inc. (WEI):
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
2
CRS WEI
• Building/rehabilitation: latrines
• Building/rehabilitation: water stations and systems
• Create WASH-friendly schools
• Distribution: school furniture and equipment
• Establish school gardens
• Form savings and lending groups
• Provide school meals
• Take-home rations
• Training: food preparation and storage practices
• Training: good health and nutrition practices
• Building/rehabilitation: kitchens
• Building/rehabilitation: storerooms
• Capacity building: local, regional, national level
• Distribution: school supplies and materials
• Enrollment campaigns
• Increase girls’ attendance at school
• Promote teacher attendance
• Raising awareness of the importance of education
• Training: parent-teacher associations
• Training: teachers
• EGRA
In addition, the FFE project’s strategic objectives and expected results are shown in Appendix B. In order 
to measure progress towards the achievement of the expected results, further sections under endline 
evaluations state the specific indicators to be assessed. 
III. Purpose of Endline Evaluation 
The purpose of the endline evaluation is to assess the state of completion as compared to expected 
results and to measure the impact of project interventions. It will also serve to identify the strengths, 
challenges, and lessons learned and capture the knowledge of the various stakeholders in order to 
leverage these for future projects on the same theme. It will also look at whether the project has exposed 
any new dynamics in learning, the quality of teaching, attendance and attention of students, the school 
environment and the importance of education. The lessons learned and recommendations of this study 
will promote the improvement of the quality and impact of future interventions.
Specific indicators that need to be assessed at the endline evaluation are the following: 
 Percent of students who, by the end of two grades of primary schooling, demonstrate that they 
can read and understand the meaning of grade-level text (male);
 Percent of students who, by the end of two grades of primary schooling, demonstrate that they 
can read and understand the meaning of grade-level text (female);
 Percent of students in target schools who are identified as attentive during class/instruction;
 Percent of parents in target schools who indicate that their children were “hungry” during the 
school day;
 Percent of students who report a decrease in health-related absence;
 Percent of parents in target communities who can name at least three benefits of education;
 Number of classrooms assessed using the Early Grade Literacy Assessment instrument;
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
3
 Percent of schools with soap1
and water at a hand-washing station commonly used by students;
 Number of government staff in relevant ministries/offices implicated in canteen/commodity 
management training;
 Number of teachers who devote at least 45 minutes a day to literacy instruction; 
 Number of teachers using the national literacy curriculum and the related instructional 
materials; 
 Number of school administrators and officials in target schools who demonstrate use of new 
techniques or tools as a result of USDA assistance; 
 Number of teachers/educators/teaching assistants in target schools who demonstrate use of 
new and quality teaching techniques or tools as a result of USDA assistance;
 Percent of school-age children receiving a minimum acceptable diet.
Again, the level of achievement of the monitoring indicators, as evidenced by CRS’ monitoring data, 
should also be assessed2
and the consultant is required to review where the project is with regards to the 
achievement of targets for all of the indicators of the project, and to identify the reasons behind any 
underachievement or overachievement. 
In addition, the endline evaluation will measure the completion of the CRS management responses to 
recommendations from the mid-term evaluation. 
The endline evaluation will use quantitative and qualitative methods to assess the relevance of 
interventions, efficiency, effectiveness, impact and sustainability of the project. 
IV. Key Evaluation Questions 
The endline evaluation should measure the impact on reading levels and attentiveness of children, on 
parents’ perceptions of the importance of education, and on the level of adoption of good hygiene and 
sanitation practices by the beneficiaries and impact of other interventions on creating a better 
environment for learning (such as SILCs, gardens, latrines and WASH, kitchens).
The questions for the endline evaluation study will be built on the same criteria used in the previous 
studies (baseline and mid-term), while placing a particular emphasis on the project’s impact. This impact 
may encompass either positive or negative changes that are the direct or indirect result of the project, 
whether intentional or not attributable to project interventions. These questions are as follows: 
 
1 The project also uses a locally accepted substitute (ash) for soap in schools. Please take them into account when 
assessing this indicator.
2 The complete list of monitoring indicators is available in the performance monitoring plan (PMP).
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
4
• Relevance: 
 To what extent were the implementation strategies relevant enough to improve (1) children’s 
literacy; (2) enrollment and attendance among students, particularly girls; and (3) community 
participation and engagement? 
 Are stakeholders (students, PTAs, mothers’ associations, teachers, and local authorities) 
satisfied with their participation in the project? Why or why not?
• Effectiveness: 
 To what extent has the theory of change of the project been verified and which evidence shows 
the contribution of each stakeholder to that?
 To what extent were the objectives of the project and the yearly benchmark indicators 
achieved? 
 Did the project meet its intended targets? 
 To what extent have targeted schools witnessed improved quality of education/learning? Were 
the strategies used appropriate to achieve the intended results?
 To what extent have members of PTAs/mothers’ associations increased their knowledge and 
understanding of literacy, nutrition, health and hygiene? Was the approached used to support 
and strengthen PTAs and AMEs effective? 
 Have targeted beneficiaries (parents) demonstrated improved understanding of the importance 
of education? The endline evaluation should triangulate the sharp drop from baseline to mid￾term in the number of parents that can identify three benefits of education.
 Have targeted beneficiaries (students and parents: APE/AME) demonstrated improved 
understanding of health and hygiene practices? Was the material and methodology used 
appropriate and effective? 
 To what extent did the Saving and Lending scheme complement the other project activities and 
support parents in meeting costs related to schooling and education? 
 Were the implementation strategies effective enough to improve (1) children’s literacy;
(2) enrollment and attendance among students, particularly girls; and (3) community 
participation and engagement? 
 Was the management system including oversight of WEI activities effective? 
 Has program implementation been effectively monitored? How well did the M&E mechanism in 
place help the implementation of the project? 
 To what extent did the project collaborate with other stakeholders, and to what extent did it 
take their experiences into account? 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
5
 To what extent/how effectively was the original skill level of teachers and other literacy partners 
taken into account before designing and while implementing the literacy intervention?
• Efficiency:
 Were activities cost efficient? 
 Were objectives achieved on time? 
 How well did CRS and WEI coordinate to ensure effective implementation, and what were the 
impediments to maximizing effective implementation?
 How effective and efficient were the WEI sub-recipients in furthering the objective of the 
program?
• Sustainability:
 What activities and/or outcomes (both expected and unexpected) of the project are likely to be 
sustained? What evidence is there to suggest this? 
 What are the major factors that can influence the achievement or non-achievement of project 
sustainability? 
 How do government’s capacities, policies, procedures, and priorities contribute to 
sustainability? What strategies were used to obtain long-lasting support from communities and 
local/central administration that will extend beyond the life of the project?
 How has the student success rate changed since the start of USDA support to FFE schools?
 To what extent has the country taken ownership of the school meals program?
• Impact: 
 To what extent have the different components of the projects led to an improvement in 
students’ literacy levels and in their health and dietary practices? The endline evaluation should 
triangulate the regression from baseline to mid-term in the percentage of students who missed 
class then report a health-related absence.
 Are there other strategies which would have worked better and had more impact? 
 What are the general contributions of the project towards the educational development in the 
four target communes? 
 Have the target beneficiaries (in particular PTAs and local government officials) taken ownership 
of the project concept and approach? 
 Are project impacts likely to be long-lasting? 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
6
 To what extent has the project achieved the desired impact (in terms of overall, strategic and 
intermediate goals)? 
 Are the observed changes attributable to the project? Did all the intended beneficiaries benefit 
from changes made by the project in the same way? If not, who benefited the most, who has 
benefited least, why? 
 What are the key success factors of the project? What are the main limiting factors? 
 What are the unintended effects of the project (positive or negative)? 
 What are the key lessons that can be learned from the project? 
 What are the key recommendations for improving the effectiveness of the program?
V. Methodology 
The endline evaluation should measure the impact of the intervention on reading levels and attentiveness 
of children, on parents’ perceptions of the importance of education, and on use of good hygiene and 
sanitation practices by the beneficiaries. In the previous studies (baseline and mid-term), qualitative 
(focus groups, interviews) and quantitative data were collected to answer the different evaluation 
questions in order to assign an impact to the implementation of the project. Also to attribute impact to 
project interventions, the “difference in difference” methodology was used. This same methodology can 
be used in this evaluation with the addition of qualitative data collection from students.
For the data analysis, the consultant may use his or her preferred software, but the “difference in 
difference” method should be applied to estimate the change attributable to the project. 
Concerning application of the “difference in difference” method, in previous studies within the 
framework of FFE project implementation, statistically significant differences between treatment and 
comparison groups
3 were noted on both observable characteristics and characteristics that the program 
will not impact—such as gender, age, size of schools, experience and education level. This evaluation 
should use as much as possible the same comparison group, but pay careful attention to indicators that 
need to be disaggregated by gender, making attempts to gather responses from equal portions of males 
and females at each level of data collection. Finally, respondents in the comparison area should be asked 
if there is any similar intervention that has been implemented during the program implementation. Any 
school which has experienced a similar intervention can no longer provide an adequate counterfactual. 
This fact is of high importance as the government and other INGOs have implemented school canteen 
programs in the same area, and CRS is out of the decision process regarding schools that benefit from 
these programs in the country.
Strong attention should be paid to the issues in this endline evaluation, and we recommend addressing 
the above concerns in econometric models, which can account for differences in school size, gender 
proportions, and potential contamination on issues addressed by the intervention. As in previous studies 
(baseline, mid-term), all stakeholders will be involved in the validation of data collection tools to ensure 
 
3
In total, 50 schools of control (no FFE intervention) against 50 schools of intervention had been randomly sampled.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
7
that the results of the endline evaluation will not be contested. For that purpose, a validation workshop 
involving stakeholders will be held to agree on the conclusions, lessons learned and recommendations 
from the evaluation.
In any event, this evaluation should be as rigorous as the previous studies. Annex C provides an outline of 
the methodology used for the mid-term evaluation.
Other Important Aspects of the Methodology
V.1. Study Design and Sampling
The endline evaluation will use a quasi-experimental design4
to measure the impact of the project and to 
attribute the findings of the evaluation to the intervention. Sampling will be determined based on 
expected impact to be detected at endline. The goal is to achieve sufficient statistical power to detect the 
impact we expect to find. School sampling will be stratified by urban and rural schools, and will be 
calculated proportionally to the number of students in each school in a way that each student has an 
equal probability of selection into the sample, despite the different size in schools.
Stratification will be completed by classifying the schools as rural or urban, as it is assumed that urban 
and rural populations differ on relevant indicators, and an equal number from each category in both 
intervention and comparison groups should be selected. Schools then have to be selected randomly 
within these strata and an equal number of treatment and control groups will be considered, following 
the baseline’s methodology. 
The sampling strategy has to focus also on outcome indicators for parents. The two main outcomes of 
interest for parents are the percentage of parents who can name at least three benefits of primary 
education, and the percentage of parents who receive a passing score on a test of good health and 
hygiene. In past studies, the children database was used as the sampling frame to develop the sample of 
parents. 
V.2. Data Collection Methods and Instruments 
The data collection methodology used for the endline evaluation will be as comparable as possible to the 
baseline and mid-term evaluation data. The evaluation will also retain the same sampling methodology 
utilized for previous studies (baseline and mid-term). The methods should include Surveys, focus group 
discussions (FGDs), key informant interviews (KIIs) and observations. As an indication, FGDs should 
include APE/AME and parents who did not send their child to school; KIIs should include regional 
pedagogy advisors of the intervention zones, and directors of the Department of Maternal and primary 
Education of Borgou and Alibori regions; INFRE; DAS; FFE staff. As for observations, they should be related 
to student attentiveness during class sequences; quality of latrines, tippy taps, boreholes, etc. However, 
the evaluator will be responsible for adapting as necessary, given any proposed changes to the 
methodology and evaluation questions, all the while ensuring comparability to the baseline and MTE.
 
4 https://www.urbanreproductivehealth.org/toolkits/measuring-success/types-evaluation-designs.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
8
In addition, to the methodology used at baseline and MTE, the endline evaluation will include students 
interviewed to have some qualitative data in terms of how they have perceived the project, and how it 
has changed their school.
List stakeholders Level
United States Department of Agriculture (USDA) Donor
Students Community
Parents: APE/AME Community
Teachers Community
Principals Community
World Education as sub-recipient, WEI local partners Departmental
Local and international partner NGO/agencies (UNICEF, WFP, etc.) Departmental and national
Head of school district Departmental
Ministry of Preschool and Primary Education (MEMP) National 
Direction départementale de la santé Departmental
School Feeding Directorate (DAS) National
Direction de l’alimentation et de la nutrition appliquée (DANA) National
National Institute for Training and Research in Education (INFRE) National
Department of Public Education (DEP) National
Pedagogical Inspection Directorate (DIP) National
Surveys will use quantitative data collection, open-ended questionnaires to collect quantitative 
information while FGDs and KIIs will use interview guides. Enumerators will be trained on how to collect 
data using observation.
Please notice that before the field survey, USDA expects to be interviewed first. Discussions/interviews 
with communities will be done in local language and in French for teachers and project staff. Translation 
can be made by a local consultant hired by the international consultant, for any independence purpose. 
According to ethical standards, the consultant must get a written authorization from the Ministry of 
Education before starting the field survey. All respondents will also be assured of a willing participation in 
the study or a willing withdrawal and the confidential treatment of all their answers.
VI. Deliverables 
The consultant will produce several key deliverables: 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
9
 An inception report including the methodology, sampling, data analysis plan, data collection 
tools (focus groups, direct observation, interview guide, individual survey guides, quantitative 
data collection, etc.5
), quality assurance plan, and a detailed work plan;
 The database (qualitative and quantitative) and the dictionary of variables (including both raw 
and clean datasets, as well as KII transcripts and notes from FGDs);
 Program (command) used to analyze the collected data;
 Draft report with a precise summary of key findings;
 Final report with an executive summary, context, methodology, evaluation questions, 
recommendations, lessons learned and an annex of all standard and custom performance 
indicators with columns for baseline, mid-term, and final values compared to the life of project 
targets (some of this data can even be taken from CRS’ own monitoring data);
 A standalone 2-3-page summary of the evaluation which includes the methodology, key 
findings, and key recommendations;
 Final report revised to integrate comments from USDA;
 Presentation of final report to stakeholders;
 Webinar to present a summary of methodology, key findings, and recommendations (for CRS 
staff).
The final version of the evaluation report should be free from personal identifying information; and the 
consultant should, with the support of CRS, ensure coordination with and involvement of all stakeholders. 
VII. Evaluation Resources 
This service takes into account data collection and analysis for endline evaluation (final data to be 
compared with the baseline and mid-term) and collection, analysis and reporting for the endline 
evaluation.
VIII. Timeline of Evaluation Activities 
The endline evaluation will take place between January and March 2018, the same time of the year as the 
baseline study.6
The anticipated timeline of activities is presented in the table below.
 
5 As part of the inception report, the consultant will highlight for each stakeholder (i) what data/information we 
want to get from them (which indicators or key evaluation questions); and (ii) which tools will be used. It will also 
outline how specific ethic and protection requirements linked to interviewing students will be met.
6 The mid-term evaluation took place in November 2016.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
10
Calendar Activities
September 2017 Preparatory workshop to discuss ToRs and partners’ roles and responsibilities
September 2017 Submit draft ToRs to USDA for review
October 2017 Submit final ToRs to USDA
October 2017 USDA approval
December 2017 Contracting of the consultant
January 2018 Conduct evaluation (data collection, analysis, draft report)
February 2018 Submission of the draft report to CRS
February 2018 – at least 
15 days before the end of the 
evaluation process
Organize a workshop to validate the draft endline evaluation report with 
stakeholders
March 2018 Submission of the final report – Validation by CRS
March 2018 Submit final endline evaluation report to USDA within 60 days from start of data 
collection
IX. Audience and Key Stakeholders 
Key stakeholders should be involved in the evaluation process and participate in discussions to respond to 
the findings and recommendations of the evaluation. The list of stakeholders and their possible interests 
are described in the table below.
Stakeholder Interest
Government of Benin – Ministry of 
Preschool and Primary School Education
Contribution of the project to the national strategy on education
USDA Measurable progress towards results; major accomplishments and 
challenges; accountability to US government
CRS, WEI and partners Major accomplishments, challenges, and lessons learned; improve 
linkages with GoB; USDA reporting and accountability requirements
Other UN and NGO agencies General progress data; lessons learned, collaboration and synergy
Schools Seeing progress on enrollment/attendance/learning, and school 
development
Communities Seeing progress in children’s education and school development as 
compared to others
X. Dissemination 
A workshop to disseminate the results of the endline evaluation will be conducted. During this workshop 
the various stakeholders will be invited. Recommendations will be made for the improvement of primary 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
11
education. After this workshop, the results of the evaluation will be published on the CRS SharePoint and 
made public. 
XI. Consultant for the Endline Evaluation 
CRS will competitively contract an external consultant or consulting firm to conduct the endline 
evaluation. CRS will name a team and point person to work and interact with the selected 
consultant/consultancy firm in advance of the assessment to ensure the validity of the methodology and 
in order to facilitate logistical arrangements.
The consultant will be led by a Team Leader who will bring significant experience in the design of 
evaluations for development projects in Africa (data collection, analysis, reporting, etc.). A good 
knowledge of school feeding projects, education and other school-related interventions will also be 
essential, as will the ability to respect legal and cultural requirements.
The third-party evaluator should avoid conflicts of interest or the appearance of conflicts of interest.
Team Leader Profile 
• At least a Master’s degree in statistics or related field (PhD preferred). At least 15 years of 
experience in the evidence-based evaluation of international development projects, preferably in 
Francophone West Africa.
• Prior experience implementing quasi-experimental methodologies.
• Prior experience evaluating literacy or school feeding interventions.
• Good knowledge of the requirements and working methods of USDA/USAID.
• Fluency in written and oral English and French.
• Availability during the evaluation period.
• Past experience with an EGRA tool is an asset.
• Ability to lead multicultural teams.
Other Members of the Consultancy Team (if a consultancy team is selected)
• At least a Master’s degree in statistics or a related field.
• Fluent English and at least passable oral French.
• Experience/knowledge of school feeding, education, health/nutrition sector trends and best 
practices.
• Experience with survey design and data collection in African countries.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
12
• Past experience with an EGRA tool is an asset.
• Gender balance.
XII. Coordination of the Endline Evaluation 
The contracted consultant will report technically and administratively to CRS. The technical focal point for 
CRS will be the MEAL Coordinator and the administrative focal points will be the Director of the FFE 
Project and the FFE Project MEAL Officer. The entire team will work together to ensure the quality of data 
and assessment results. All logistics aspects linked to the preparation and the evaluation itself will be 
organized and coordinated by the administrative focal point.
The independent evaluator will be free to draw their own conclusions free from organizational or political 
pressure.
At the beginning of the mission, a scoping session will be held with the MEAL Team of the project, the 
Project Manager, the MEAL Coordinator and the consultants team to clarify the objectives and 
expectations of the mission.
After submission of the interim report, the key team of the FFE project (CRS and WEI) will review the 
findings and provide their feedback. A second level of inputs will then be sought from CRS Regional MEAL 
Technical Adviser and the CRS Senior Technical Adviser for Education in the headquarters.
USDA will participate in a stakeholder call with the evaluator, will also provide comments on the draft 
evaluation report, and approve the final report.
XIII. Structure of Proposal and Submission Guidelines 
Consultants or consultant firms wishing to apply to conduct this evaluation should send their CVs, along 
with a technical proposal that includes at least the following specifications: 
 A description of the firm’s expertise (maximum 5 pages);
 The different tasks they are planning to undertake in order to fulfill the mission (including the 
methodology they will use);
 A detailed budget according to the illustrated models in Appendix C;
 A sample of similar work undertaken as the lead consultant.
NB: It should be noted that these ToRs are subject to review and changes by USDA prior to contracting 
the consultant or consultant firm, for which negotiations with the selected consultant or firm will be 
required to finalize the terms of the contract. 
The proposal should be sent either in a sealed envelope to CRS at the following address: Catholic Relief 
Services-USCCB, Benin-Togo Program, Lot 11 Les Cocotiers, 01 BP 518 RP, Cotonou, Benin, or 
electronically in PDF, no later than October 22, 2017, at 17:00 PM (Benin local time) to 
benin@global.crs.org, with copy to the CRS Head of Programming (marie.rongeard@crs.org). The subject 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
13
line should read: “Consultant for the endline evaluation of the CRS Benin FFE Project.” Any additional 
information sent beyond what has been requested will not be considered.
APPENDIX A: Theory of change 
Problematic 
Benin is one of the world’s poorest countries with an annual per capita income of $365. This figure is 
below the sub-Saharan Africa average of $470 and is 159th out of 174 countries ranked according to the 
Human Development Index of the United Nations Development Programme (UNDP). 
Due to colonization heritage, the development of the northern Benin is far behind the development of 
the south where the colonies had their interests in the littoral. This lack of balance is seen in terms of 
economic poverty. All of the rural northern departments (Borgou, Alibori, Atacora, Donga) have levels of 
economic poverty significantly above average (32-36%). The insufficient number of primary schools is a 
clear example of the lack of social services in the north: less than the half (48%) of the primary schools in 
the country are located in these four departments whereas they represent the 73% of Benin’s superficies. 
This means that schools are very remote for many school-aged children to attend. In addition, the Borgou 
and Alibori departments are the less assisted by both the government and NGOs as far as school canteens 
are concerned (Ministère de l’éducation maternelle et primaire, MEMP). The inherited poverty causes a 
severe yearly lean season whereas an acute lack of development in the north generates poorly educated 
parents who are not involved in the education of their children. The population does not have basic 
nutrition concepts, nor does it practice the hygiene minimum standard. The children do not have enough 
to eat and suffer from stunting growth. Therefore, they are not learning well at school or they do not 
attend school at all.
There are local constraints such as social and cultural beliefs which causes parents not to know the 
importance of education. In addition, at national level, one notices the dysfunctional education system 
which causes teacher absenteeism. 
Possibilities in Terms of Solutions 
In order to solve the above problematic, the following actions can make a difference: 
 Government’s actions: Government should actually apply the educational policies, including: 
 True decentralization of the education system;
 Training and capacity building of parents and teachers;
 Zero tolerance for corruption/misuse;
 Appropriate technology for learning;
 More incentives for teachers.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
14
 Parents’ actions: 
 Change perception of education value;
 From both sides: 
 Reinforce positive behaviour;
 Improve school environment;
 Negotiated school schedule.
“We wish to have productive Beninese citizens through making all girls and boys healthy and with a 
quality and complete primary education.” 
IF we...
 Promote importance of education to parents;
 Improve teacher incentives and capacity;
 Advocate for policy application, decentralization, parent involvement;
 Provide inputs (school infrastructure, meals)
THEN...
All children, boys and girls will have a positive, healthy, quality learning experience and complete their 
education to be productive citizens. 
Adopted Solution: the FFE project 
Targets: The 38,000 primary school children in the communes of Malanville, Kandi, Gogounou and Kalale; 
144 schools; 212 teachers/school administrators; 92 AMEs. 
Outcomes: 
 Healthy boys and girls;
 All children learn in a good quality, safe environment;
 All children improve their reading capacity;
 All children complete the learning cycle—they are literate;
 Parents involved in education;
 Reading materials available;
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
15
 Improved education system;
 Good curriculum;
 Productive, educated citizens;
 Gender equality.
APPENDIX B: Strategic Objectives and Expected Results 
Result stream Intermediate result Expected result
MGD SO 1: Improved literacy of school-age children
• 65% of boy students who, by the end of two grades of school can read and understand the meaning of grade-level text
• 22,621 males benefiting directly from USDA-funded interventions.
• 60% of girl students who, by the end of two grades, can read and understand the meaning of grade-level text.
• 22,879 females benefiting directly from USDA-funded interventions.
• 220,783 individuals benefiting indirectly from USDA funded interventions.
1.1 IMPROVED QUALITY OF 
LITERACY INSTRUCTION 
70% of teachers who devote 
at least an average of 45 
minutes a day to literacy 
instruction
1.1.1 More consistent teacher 
attendance
75% average teacher attendance rate for each school and aggregated by 
district 
1.1.2 Better access to school 
supplies and materials
4,230 textbooks and other teaching and learning materials provided as a 
result of USDA assistance
1.1.3 Improved literacy 
instructional materials
144 teachers using the national literacy curriculum and the related 
instructional materials
1.1.4 Increased skills and 
knowledge of teachers
144 teachers/educators/teaching assistants trained or certified as a result 
of USDA assistance
144 teachers in target schools who demonstrate use of new and quality 
teaching techniques or tools
1.1.5 Increased skills and 
knowledge of school 
administrators
144 school administrators or officials trained or certified as a result of 
USDA assistance
106 school administrators in targeted schools who demonstrate use of 
new techniques or tools
1.2 IMPROVED 
ATTENTIVENESS 
75% of students in target 
schools who are identified 
as attentive during 
class/instruction
1.2.1 Reduced short-term hunger Less than 20% of parents in target schools indicate that their children were 
“hungry” during the school day
1.2.1.1 Increased access to food 
(school feeding)
95% of students in target schools consuming daily meals at school
22,268 boy students receiving daily school meals (breakfast, snack, lunch) 
as a result of USDA assistance
21,536 girl students receiving daily school meals (breakfast, snack, lunch) 
as a result of USDA assistance
18,756,998 daily school meals (breakfast, snack, lunch) provided to school￾age children as a result of USDA assistance
80,703 take-home rations provided as a result of USDA assistance
13,884 girls receiving take-home rations as a result of USDA assistance
13,017 boys receiving take-home rations as a result of USDA assistance
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
16
Result stream Intermediate result Expected result
1.3 IMPROVED STUDENT 
ATTENDANCE 
19,607 boy students 
regularly (80%) attending 
USDA -supported 
classrooms/schools 
19,069 female students 
regularly (80%) attending 
USDA supported 
classrooms/schools
1.3.2 Reduced health-related 
absences
2% of students who report a decrease in health-related absences
1.3.3 Improved school 
infrastructure
144 school kitchens rehabilitated/constructed as a result of USDA 
assistance
252 cabins rehabilitated/constructed as a result of USDA assistance
144 storerooms rehabilitated/constructed as a result of USDA assistance
1.3.4 Increased student 
enrollment
21,536 girls enrolled in schools with USDA assistance
22,268 boys enrolled in schools with USDA assistance
1.3.5 Increased community 
understanding of benefits of 
education
60% of parents in target communities who can name at least three 
benefits of primary education
1.4 FOUNDATIONAL 
RESULTS
1.4.1 Increased capacity of 
government institutions
Students in 144 schools assessed using Early Grade Reading Assessment 
tool
1.4.4 Increased engagement of 
local organizations and community 
groups
22,268 male social assistance beneficiaries participating in productive 
safety nets as a result of USDA assistance
21,536 female social assistance beneficiaries participating in productive 
safety nets as a result of USDA assistance
100% of schools in target communities with active PTAs or similar “school” 
governance structures
92% of parent-teacher associations or similar “school” governance 
structures supported as a result of USDA assistance
MGD SO 2: Increased use of health and dietary practices
• 95% of students in schools receiving USDA assistance receiving a minimum acceptable diet
2.1 IMPROVED 
KNOWLEDGE OF HEALTH 
AND HYGIENE PRACTICES
40% of parents in target schools who achieve a passing score on a test of 
good health and hygiene practices
50% of students in target schools who achieve a passing score on a test of 
good health and hygiene practices
2.2 INCREASED 
KNOWLEDGE OF SAFE FOOD 
PREP AND STORAGE 
PRACTICES
100% of food preparers at target schools who achieve a passing score on a 
test of safe food preparation and storage
2.3 INCREASED 
KNOWLEDGE OF NUTRITION
4 female teachers trained in child health and nutrition as a result of USDA 
assistance (school gardens)
21 male teachers trained in child health and nutrition as a result of USDA 
assistance (school gardens)
2.4 INCREASED ACCESS TO 
CLEAN WATER AND 
SANITATION SERVICES
100% of schools using an improved water source
144 schools with improved sanitation facilities (latrines)
2.6 INCREASED ACCESS TO 
REQUISITE FOOD PREP AND 
STORAGE TOOLS AND 
EQUIPMENT
144 target schools with improved food preparation and storage 
equipment
2.7 FOUNDATIONAL 
RESULTS
2.7.1 Increased capacity of 
government institutions
4 government staff in relevant ministries/offices implicated in 
canteen/commodity management training
2.7.4 Increased engagement of 
local organizations and community 
groups
71 nutrition or health initiatives or activities pursued in partnership 
between government and local community groups
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
17
APPENDIX C: Methodology of the Mid-Term Evaluation 
The design must take into account how data collection will take place, and must ensure that the 
evaluation has the necessary rigour and scientific validity. To this end, both quantitative and qualitative 
methods will be used. The methodology of data collection will be based on random methods where 
appropriate. Data collection methods may include interviews with stakeholders, some of which may be 
chosen at random, observation, and focus groups among others. Project staff may also be interviewed. 
Like the baseline, the mid-term evaluation study will target school children, parents, teachers, PTAs, AME, 
and CCSs and some members of the community who have not sent their children to school. It will also 
collect the necessary data on the various performance indicators to help measure the progress achieved 
thus far through the implementation of the project. Also, as a key stakeholder, the donor, USDA, will be 
interviewed prior to data collection and, ideally, prior to the finalization of any data collection tools. All 
data collected during the mid-term evaluation should be modelled on the methodology used for the 
baseline study and data collection in order to enable triangulation and comparison of results across the 
different studies. 
Within the framework of the baseline study, the quantitative portion has had two main components. The 
first component has consisted of surveys of principals, teachers and parents, as well as a classroom 
observation. The second component has consisted of an Early Grade Reading Assessment (EGRA). The 
qualitative component has consisted of interviews with key stakeholders such as the head of the school 
districts (CCS) in each intervention commune, and parents’ and mothers’ associations which were 
identified by the quantitative component as particularly active. 
Sampling Design for Schools, Principals, Teachers, Students and Parents
• Schools – Sampling was determined based on expected impact to be detected at mid-term or 
online. The goal is to achieve sufficient statistical power to detect the impact we expect to find. 
School sampling should be stratified by urban and rural schools, and calculated proportional to the 
number of students in each school. In this way each student has an equal probability of selection 
into the sample, despite the different size in schools. Stratification should be completed by 
classifying the schools as rural or urban, and selecting equal numbers from each category in both 
intervention and comparison groups. Schools should then be selected randomly within these 
strata. In the baseline, treatment and control groups each consisted of seven urban schools, and 
43 rural schools. The complete list of schools which were sampled at baseline included 
142 intervention schools and 98 comparison schools.
• Teachers and principals – All teachers and principals were planned to be surveyed at each school so 
no selection methodology was implemented. 
• Students – Classrooms for observation were also selected randomly. At each school, two 
classrooms were randomly selected for observation. Within the classroom, 10 students were again 
randomly selected to be observed and classified as attentive or inattentive at the time of 
observation. For this component of the study, only classrooms from 3rd to 6th grade were 
selected, as the students in 1st and 2nd grade were being pulled out for EGRA testing. Thus it was 
assumed that those classrooms would have sufficient disturbance to negatively bias the results, 
since students would be paying attention to the movement of their peers instead of the teacher. 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
18
• Parents – The sampling strategy focused on outcome indicators for parents. The two main 
outcomes of interest for parents are the percentage of parents who can name at least three 
benefits of primary education, and the percentage of parents who receive a passing score on a test 
of good health and hygiene. For both indicators, the goal is 60%. Using a significance of 5%, a 
statistical power of 80%, it was determined that with 50 treatment and 50 control schools, and 
10 parents randomly selected using the school rosters surveyed per school, the study would have 
power to detect a minimum change from 46% to 60%. For selection of the schools based on these 
calculations, the gsample7
command in STATA was used. Within each stratum, schools were 
randomly selected with a probability proportional to the size of the school. Since for some schools 
the probability was greater than the available selection probability, these probabilities were limited 
to 1/50.
Sampling Design for the Qualitative Component
The qualitative component focused on key stakeholders and key informants on project outcomes of 
interest. First, interviews were conducted with the head of school districts (CCS), in each intervention 
commune. Second, focus groups were conducted among parents’ and women’s associations (APEs and 
AMEs) that were identified by the principals as most active. These associations were asked to identify 
parents who have not enrolled their children in school and interviews were conducted with those parents 
as well. In this way, the qualitative interview employed a snowball method, in which key informants 
identify other key informants.
The whole approach and methodology used for the baseline evaluation will be made available to the 
selected consultant
Indicatively, the consultant will be responsible among others: 
 To define criteria for selecting interviewers;
 To train support staff (interviewers) – CRS and the consultant team will work together to ensure 
quality of data collection and supervise data collectors: without encroaching on the consultant 
team’s independence, the Country Program MEAL Coordinator will help in the selection and 
training of the interviewers, especially in their ability to conduct the interviews in respect of the 
local context such as the local languages. Whereas the CRS MEAL Coordinator will assist in the 
process of hiring and orienting/training the interviewers, the consultant will be fully responsible 
for overall supervision of the interviewers as well as the data collection efforts in the field. As far 
as the project team is concerned, examples of how it will support the effort include: ensuring 
that by the end of the training of the interviewers, the Ministry of Education gives authorization 
to undertake the data collection at schools; facilitating the logistics (vehicles, accommodations, 
supplies, etc.); and supporting the consultant(s) with any problem-solving related to data 
collection;
 To conduct field tests;
 
7
Jann, B. (2006). gsample: Stata Module to draw a Random Sample. Available from 
http://ideas.repec.org/c/boc/bocode/s456716.html.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 1 – TERMS OF REFERENCE (CONTINUED) 
19
 To supervise data collection:
 To clean the database;
 To ensure survey coordination;
 To communicate regularly with CRS to guarantee appropriate support for the whole process.
Note: For the EGRA test, priority will be given to the enumerators who had been used for the same 
exercise during the baseline study.
In any case, the different collection tools and methods to be definitively chosen for the mid-term and 
endline evaluations should be submitted to the coordination team before all field trips. A validation of 
these tools will be necessary, in agreement with all stakeholders. After data collection, there should be 
verification of the quality of the data followed by analysis using a rigorous methodology to measure the 
performance of the project intervention. The consultant should submit the relevant sampling method to 
generalize the results and make comparisons between different categories of students. A specific analysis 
will be made on performance (attendance, attentiveness and result) of school children who received 
take-home rations (THRs) compared to other children who did not.
The consultant will provide a well-detailed report followed by a lessons-learned document and 
recommendations. CRS will hold a validation workshop to discuss preliminary results.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
20
APPENDIX 2 – EVALUATION MATRIX 
Questions and sub-questions Indicators Data Source Data Collection 
Method
1. Relevance (are we doing the right thing with the right people at the right place?)
1.1 To what extent were the 
implementation strategies 
relevant enough to improve 
(1) children’s literacy?
• Improved Literacy of School-Age 
Children
• % of boy students who, by the end of 
two grades of can read and 
understand the meaning of grade level 
text
• % of girl students who, by the end of 
two grades, can read and understand 
the meaning of grade level text
• Students
• Parents
• Teachers, principals
• CRS staff/WEI staff
• Survey
• FGDs
• KIIs
• EGRA 
1.2 To what extent were the 
implementation strategies 
relevant enough to improve 
(2) enrollment and attendance 
among students, particularly girls?
• Reduced Health-Related Absences 
Increased Student Enrollment
• Students
• Parents
• Teachers, principals
• CRS staff/WEI staff
• Survey 
• School 
records
• FGDs
• KIIs
1.3 To what extent were the 
implementation strategies 
relevant enough to improve 
(3) community participation and 
engagement?
• Increased community understanding 
of benefits of education
• Increased engagement of local 
organizations and community groups
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• Survey 
• FGDs
• KIIs
• Desk review
1.4 Are stakeholders (Students, PTAs, 
mothers’ associations, teachers, 
and local authorities) satisfied with 
their participation in the project? 
Why or why not?
• Level of satisfaction of stakeholders 
with their participation in the project
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• Survey 
• FGDs
• KIIs
2. Effectiveness (are we achieving our expected results?)
2.1 To what extent has the theory of 
change of the project been 
verified and which evidence shows 
the contribution of each 
stakeholder to that?
• Improved Literacy of School-Age 
Children
• Increased Use of Health and Dietary 
Practices
• Increased Capacity of Government 
Institutions
• Increased engagement of Local 
organizations and community groups
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• Survey 
• School 
records
• FGDs
• EGRA 
• KIIs
2.2 To what extent were the 
objectives of the project and the 
yearly benchmark indicators 
achieved?
• Expected results and yearly planned 
progress vs actual progress
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
2.3 Did the project meet its intended 
targets?
• Comparison of survey/desk review 
data with initial program targets 
(stated in the PMP)
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 2 – EVALUATION MATRIX (CONTINUED) 
21
Questions and sub-questions Indicators Data Source Data Collection 
Method
2.4 To what extent have targeted 
schools witnessed improved 
quality of education/learning? 
Were the strategies used 
appropriate to achieve the 
intended results?
• Perception of various stakeholders of 
the quality of education/learning 
• Perception of contribution/attribution 
of CRS/WEI interventions to the 
improved quality of learning
• Students
• Parents: APE/AME
• Teachers, principals
• Survey
• FGDs
• KIIs
• Desk review
2.5 To what extent have members of 
PTAs/mothers’ associations 
increased their knowledge and 
understanding of literacy, 
nutrition, health and hygiene? 
Was the approached used to 
support and strengthen PTA and 
AME effective?
• Extent to which community groups 
have increased their knowledge and 
understanding of literacy, nutrition, 
health and hygiene
• Parents: APE/AME • Survey
• FGDs
• Desk review
2.6 Have targeted beneficiaries 
(parents) demonstrated improved 
understanding of the importance 
of education? 
• Extent to which parents demonstrate 
improved understanding of the 
importance of education 
• Parents: APE/AME
• CRS staff/WEI staff
• Survey
• FGDs
• Desk review
• KIIs
2.7 Have targeted beneficiaries 
(students and parents: APE/AME) 
demonstrated improved 
understanding of health and 
hygiene practices? Was the 
material and methodology used 
appropriate and effective?
• Increased understanding of health and 
hygiene practices by students and 
parents (APE/AME)
• Perception of stakeholders regarding 
the materials and methodology used
• Students
• Parents: APE/AME
• Teachers, principals
• CRS staff/WEI staff
• Survey
• FGDs
• KIIs
• Desk review
2.8 To what extent did the Saving and 
Lending scheme complement the 
other project activities and 
support parents in meeting costs 
related to schooling and 
education?
• Perception of the impact of the saving 
and lending scheme to the other 
project activities and supporting 
parents in meeting costs related to 
schooling and education
• Parents: APE/AME
• Principals
• CRS staff/WEI staff
• Survey
• FGDs
• KIIs
2.9 Were the implementation 
strategies effective enough to 
improve (1) children’s literacy, 
(2) enrollment and attendance 
among students, particularly girls, 
and (3) community participation 
and engagement?
• Comparison of survey data with initial 
baseline values related to literacy, 
enrollment/attendance, and 
community participation
• Perception of stakeholders regarding 
the implementation strategies
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
2.10 Was the management system 
including oversight of WEI 
activities effective?
• Perception of various stakeholders 
(CRS, WEI, USDA, Government 
Departments, school management) on 
the management system including 
oversight of WEI activities
• Local government authorities
• CRS staff
• WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
2.11 Was program implementation 
effectively monitored? How well 
did the M&E mechanism in place 
help the implementation of the 
project?
• Perception of various stakeholders 
(CRS, WEI, USDA,) of the M&E 
mechanism.
• Perception of the quality and 
timeliness of reporting processes by 
the relevant stakeholders.
• CRS staff
• WEI staff
• USDA staff (UNICEF, WFP, 
etc.)
• Local government authorities
• FGDs
• Desk review
• KIIs
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 2 – EVALUATION MATRIX (CONTINUED) 
22
Questions and sub-questions Indicators Data Source Data Collection 
Method
2.12 To what extent did the project 
collaborate with other 
stakeholders, and to what extent 
did it take their experiences into 
account?
• Perception of various stakeholders 
(CRS, WEI, USDA, Government 
Departments, school management, 
PTAs, AMEs) on project collaboration 
and capacity to engage with their 
experiences. 
• Parents: APE/AME
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
3. Efficiency (are we using the right process to achieve and measure our expected results?)
3.1 Were objectives achieved on 
time?
• Planned achievement of objectives vs 
actual achievement. 
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
3.2 How well did CRS and WEI 
coordinate to ensure effective 
implementation, and what were 
the impediments to maximizing 
effective implementation?
• Perception of relevant stakeholders 
(CRS, WEI, USDA, Government 
Departments, school management) on 
coordination and implementation of 
activities by CRS and WEI
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff
• WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
3.3 How effective and efficient were 
the WEI sub recipients in 
furthering the objective of the 
program?
• Perception of various stakeholders 
(CRS, WEI, USDA, Government 
Departments, school management) on 
the contribution of WEI activities to 
the overall results of the program.
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff
• WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
3.4 To what extent were the 
recommendations from the mid￾term evaluation implemented?
• Number and type of recommendations 
from the mid-term review 
implemented
• CRS staff
• USDA representatives
• Mid-term evaluation report
• KIIs
• Desk review
4. Sustainability (is it going to last after program completion)
4.1 What activities and/or outcomes 
(both expected and unexpected) 
of the project are likely to be 
sustained? What evidence is there 
to suggest this?
• Degree to which results/outcomes 
associated to the various components 
of the program are to remain after 
program completion
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Survey
4.2 What are the major factors that 
can influence the achievement or 
non-achievement of project 
sustainability?
• Perception of relevant stakeholders on 
the sustainability of activities and 
outcomes
• Number and type of factors that could 
limit or enhance program sustainability
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
4.3 How do government’s capacities, 
policies, procedures, and priorities 
contribute to sustainability?
• Increased capacity of
• Government Institutions
• Number and type of factors that could 
limit or enhance program sustainability
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
4.4 What strategies were used to 
obtain long-lasting support from 
communities and local/central 
administration that will extend 
beyond the life of the project?
• Review of best practices and 
stakeholder perception on strategies 
that were used to obtain long lasting 
support from communities and 
local/central administration that will 
extend beyond the life of the project. 
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 2 – EVALUATION MATRIX (CONTINUED) 
23
Questions and sub-questions Indicators Data Source Data Collection 
Method
5. Impact (what are the most important changes for our beneficiaries and key partners?)
5.1 To what extent have the different 
components of the projects led to 
an improvement in student’s 
literacy levels and in their health 
and dietary practices? 
• Measure progress in sample student 
beneficiaries of the project and control 
group (literacy and health)
• Student, parent and teacher 
perception on the relevance of various 
components and their impact in 
improved literacy and health
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
5.2 Are there other strategies which 
would have worked better and 
had more impact?
• Perception of stakeholders on other 
strategies
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
5.3 What are the general 
contributions of the project 
towards the educational 
development in the four target 
communes?
• Types of impacts reported by 
stakeholders regarding the quality of 
education
• Perception of relevant stakeholders of 
the general contributions of the 
project towards the educational 
development in the four target 
communes
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• Local and international 
partner NGO/agencies
• CRS staff/WEI staff
• Survey
• FGDs
• KIIs
• Desk review
5.4 Have the target beneficiaries (in 
particular PTAs and local 
government officials) taken 
ownership of the project concept 
and approach?
• Perception of the target beneficiaries 
of their ownership and leadership in 
the project
• Number of examples of program 
ownership displayed
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
•
• FGDs
• KIIs
5.5 To what extent has the project 
achieved the desired impact (in 
terms of overall, strategic and 
intermediate goals)?
• Review of planned results and 
achieved results
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• Survey
• FGDs
• KIIs
• Desk review
5.6 Are the observed changes 
attributable to the project? Did all 
the intended beneficiaries benefit 
from changes made by the project 
in the same way? If not, who 
benefited the most, who has 
benefited least, why?
• Types of impacts reported by 
stakeholders regarding their quality of 
life
• Perception of stakeholders on the 
attribution and contribution of the 
project to the results. 
• Comparison with control group
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• Survey 
• FGDs
• KIIs
• Desk review
5.7 What are the key success factors 
of the project? What are the main 
limiting factors?
• Number of best practices reported by 
stakeholder
• Perception of stakeholders on the key 
success and limiting factors for the 
project
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
5.8 What are the unintended effects 
of the project (positive or 
negative)?
• Number and type of positive and/or 
negative unexpected results stemming 
from the implementation of FFE 
programming
• Perception of the relevant 
stakeholders regarding the unintended 
effects of the project
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 2 – EVALUATION MATRIX (CONTINUED) 
24
Questions and sub-questions Indicators Data Source Data Collection 
Method
6. Lessons learned and best practices (what should we remember for a future program?)
6.1 What are the key lessons that can 
be learned from the project? What 
implications for future Food for 
Education activities can be 
extracted from those lessons 
learned?
• Number and type of operational 
lessons learned from the program
• Number and type of developmental 
lessons learned from the program
• Perception of stakeholders on the key 
lessons learnt from the project. 
• Students
• Parents: APE/AME
• Teachers, principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
6.2 What are the key 
recommendations for improving 
the effectiveness of the program?
• Number and type of best practices or 
recommendations developed by the 
program
• Students
• Parents: APE/AME
• Teachers, Principals
• Local government authorities
• CRS staff/WEI staff
• USDA representatives
• FGDs
• KIIs
• Desk review
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
25
APPENDIX 3 – LIST OF DOCUMENTS REVIEWED 
The following is a list of documents reviewed by the consulting team for this evaluation.
• Catholic Relief Services Food For Education Theory of Change.
• Evaluation Plan Food For Education 2014.
• Project Agreement between the Government of the United States of America and the Catholic 
Relief Services – United States Conference of Catholic Bishops for the Provision of Agricultural 
Commodities through the McGovern-Dole International Food for Education and Child Nutrition 
Program Act. 
• Food For Education PMP (Performance Management Protocol).
• CRS Food For Education: Baseline Study. Report prepared by Notre Dame Initiative for Global 
Development (NDIGD) (Danice Brown and Juan Carlos Guzmán). March 2015, and available data 
and tools from this study. 
• Mid-Term Evaluation of FFE: Intervening in a Demanding Context report prepared by World 
Education (Maurice Garnier, Pierre GBENOU and Serge Marcel Loukpe). April 2017 and available 
data and tools from this study. 
• CRS Benin –Management Response to recommendations made in the Midterm Report.
• Development Assistance Committee of the Organisation for Economic Co-operation and 
Development. Principles for Evaluation of Development Assistance. Paris, OECD, 1991.
• Strategie WASH – Catholic Relief Services
• The Sustainability Through SILC Group – Catholic Relief Services
• APPRENDRE A LIRE - Manuel de formation préparé pour les Enseignants de CI et CP – CRS, WEI
• MODULES DE FORMATION DES STRUCTURES APE ET AME
• Curriculum de formation sur la nutrition par: Bio Ozias DOMAGUI, Garden & Kitchen Officer, Projet 
FFE
• RECUEIL DE METS LOCAUX AMELIORES - PROJET FOOD FOR EDUCATION (FFE).
• GUIDE - ENSEIGNEMENT DE L’HYGIENE ET DE LA NUTRITION A L’ECOLE PRIMAIRE BASE SUR LES 
COMPETENCES DE VIE.
• Progress Reports CRS/Progress Reports WEI.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
26
APPENDIX 4 – EGRA SURVEY 
Please note this is exactly the same survey that has been implemented in the past.
INSTRUMENT EGRA BENIN : FRANÇAIS - ECOLIER CI/CP 
LIVRET DE MARQUAGE DE L’ENUMERATEUR
Circonscription scolaire : …………………………………. Ecole 
……………………………………………………………
Ecolier N° …………………………………………… Classe CI CP
Date d’administration du test / / 201 ____
Nom de l’énumérateur : ………………………………………. Signature …… 
Visa du superviseur ou contrôleur …………………………………………………… 
Numéro du questionnaire …………….. 
Heure du début du test …………………. Heure de la fin du test ………………
Consentement verbal obtenu: OUI
On va commencer, je vais prendre quelques questions sur toi même et ta famille.
1 Zone de vie de l’écolier Urbaine Rurale
2 Quel est ton âge? ou ta date de naissance ? ……………………………………………
3 Genre (regardezl’enfant et cochez) Masculin Féminin
5 Quelle était ta classe l’année passée ? CI CP pas à l’école (maison)
6 As-tu fréquenté une école maternelle ? oui non pas de réponse
7 Quelle(s) langue(s) parles-tu à la maison?
Francias
Langues nationals (preciser________________)
oui non pas de réponse
oui non pas de réponse
8 Lis-tu ou étudies-tu en français à la maison? oui non pas de réponse
9 As-tu un manuel ou livre de lecture de français à la 
maison?
oui non pas de réponse
11
Qui t’aide avec le travail d’école à la maison ?
oui non Pas de
réponse
Papa ?
Maman ?
ton frère ou ta sœur ?
Autre personne (préciser)
…………………………..
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
27
12 Y-a-t-il des personnes dans ta famille qui
savent lire ?
Papa ?
Maman ?
ton frère ou ta sœur ?
Autre personne (préciser)
…………………………..
………………………….
14 Que fais-tu en dehors des cours Je vais aux champs
Je fais les travaux
Domestiques
Je vends avec mes parents
Je vais á l’école coranique
Autre (préciser)
15 Manges-tu? Avant de venir a l’école?
Pendant la recréation?
À la cantine?
16 Si oui à la cantine, nombre de fois par semaine une fois 2 fois 3 fois 4 fois 5 fois
17 Avec quelle fréquence le maître te désigne-t-il pour lire 
au tableau ou dans le livre à l’école?
Tous les 
jours
Une fois 
par semaine
Une fois 
par mois
Jamais
18 Combien etes-vous à lire un seul de lecture en classe? Je 
lis le livre de lecture en classe?
Je ne lis 
pas en 
classe
Seul à 2 à 3 à 4 plus 
de 4
Section 1 : VOCABULAIRE
Code Lisez les mots du vocabulaire en français
Montre-moi :
Nombre de
réponses 
correctes
A : Partie du corps Ta tête – ta main – ton ventre – ton pied … /4
B : Objets scolaires Un chiffon – une ardoise - un cahier – un crayon … /4
C : Nature un caillou – le sol –– un bol – une table … /4
D: Termes d’orentation 
dans l’espace
Ton côté droit –le dessous de la table ––– le coin de la table. …/3
Total correct …/15
Section 2: COMPREHENSION ORALE
Démarrez (lisez ce qui suit en français) Réponse correcte
(et auto correction)
A Prends le crayon et dépose-le (á repeater pour chaque option):
• sur la feuille
• derrière toi
• par terre
• dans la boite
• à ta gauche
…/1
…/1
…/1
…/1
…/1
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
28
B Tu prends le crayon avec ta main
et tu le déposes sur la table à ta droite
…/2
…/3
C Avec ta main gauche, prends la craie
et dépose-la sous la table.
…/2
…/2
D Prends le chiffon avec la main droite
et pose-le sous ta chaise.
…/2
…/2
Total des réponses correctes …/18
Section 3 : IDENTIFICATION DU NOM DES LETTRES 
Démarrezle chronomètre dès que l’écolier commence à lire la première lettre. Suivez avec votre
crayon sur votre feuille.
Ligne 1 o e a s i è b A o U …/10
Ligne 2 e c l p t m ê v B R …/10
Ligne 3 o s i r t n d p u Q …/10
Ligne 4 a o L u e é è a n i …/10
Ligne 5 e ê t i f n M r v d …/10
Ligne 6 b q g e s f j k é u …/10
Ligne 7 h x è n M é m l g o …/10
Ligne 8 i o r f s t p n d v …/10
Ligne 9 t G u v a p d o a i …/10
1 Mettez une croix (X) si l’exercice a été arrêté par manque de réponses correctes à la première
Ligne
2 Nombre total de lettres lues :
3 Nombre total de lettres non lues (pas de réponse) :
5 Temps total si toutes les lettres sont lues en moins de 60 secondes :
6 Nombre de lettres pour lesquelles le son ou le nom a été correctement donné en 1 minute
Section 4 : IDENTIFICATION DU SON DES LETTRES 
Démarrezle chronomètre dès que l’écolier commence à lire la première lettre. Suivez avec votre crayon sur
votre feuille.
Ligne 1 i an o è ou eu in r ê è …/10
Ligne 2 ei p l t em am d ei v ai …/10
Ligne 3 y in oi ai b om er s c ez …/10
Ligne 4 er k on w f au x h j ed …/10
Ligne 5 ou r q ein p eau f ez er ei …/10
Ligne 6 o in on au q m br ei cr ez …/10
1 Mettez une croix (X) si l’exercice a été arrêté par manque de réponses correctes à la
première ligne
2 Nombre total du son des lettres lues :
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
29
3 Nombre total du son des lettres non lues (pas de réponse) :
5 Temps total si toutes les lettres sont lues en moins de 60 secondes :
6 Nombre de lettres pour lesquelles le son a été correctement donné en 1 minute
Section 5 : ANALYSE PHONEMIQUE - Identification du son initial
Quel est le tout premier son dans le mot « » ? « » ? (Lire chaque 
mot deux fois)
dur /d’/ o Correct o Incorrect o Pas de réponse
lait /l l l l l/ o Correct o Incorrect o Pas de réponse
car /k’/ o Correct o Incorrect o Pas de réponse
sac /ssssss/ o Correct o Incorrect o Pas de réponse
jour /jjjjj/ o Correct o Incorrect o Pas de réponse
fête /ffffff/ o Correct o Incorrect o Pas de réponse
tour /t’/ o Correct o Incorrect o Pas de réponse
balle /b’/ o Correct o Incorrect o Pas de réponse
pâte /p’/ o Correct o Incorrect o Pas de réponse
voleur /vvvv/ o Correct o Incorrect o Pas de réponse
1 Mettez une croix (X) si l’exercice a été arrêté par manque de réponses correctes ou incapacité
de segmentation aux cinq premiers mots (auto-stop)
2 Nombre total de sons donnés :
3 Nombre total de sons non donnés (pas de réponse) :
4 Nombre total de sons incorrectement donnés :
5 Score = Nombre total de sons correctement produits:
Section 6 : LECTURE DE MOTS FAMILIERS ISOLES
Démarrez avec le chronomètre 
Ligne 1 Tu ton la oui fête …/5
Ligne 2 Ou vélo tata table kaki …/5
Ligne 3 Maison peau balai pipe képi …/5
Ligne 4 École beau chat grand épi …/5
Ligne 5 Maman enfant voilà bras elle …/5
Ligne 6 Lune bras matin non pied …/5
Ligne 7 Vélo bon craie pipe papa …/5
Ligne 8 Route canari jolie café coton …/5
Ligne 9 Tout moi robe école papa …/5
Ligne 10 Kepi canari mardi jeu livre …/5
1 Mettez une croix (X) si l’exercice a été arrêté par manque de réponses correctes ou par incapacité
de lecture à la première ligne.
2 Nombre total de mots familiers lus :
3 Nombre total de mots familiers non lus (pas de réponse) :
5 Temps total si tous les mots sont lus en moins de 60 secondes :
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
30
6 Score = Nombre de mots familiers correctement lus en 1 minute
Section 7: LECTURE DE MOTS INVENTES/PSEUDOMOTS 
 
Démarrez avec le chronomètre
Ligne 1 Ga pité po mu ki …/5
Ligne 2 Sar piti dandi nebou sèko …/5
Ligne 3 Topa dio doa neau kal …/5
Ligne 4 Tasé ita pa vo tobin …/5
Ligne 5 Kor pata pik tapau dron …/5
Ligne 6 Fumi pamu rou sula samou …/5
Ligne 7 Davo ko fa pouta pémi …/5
Ligne 8 Gato tupo lolo topé dogo …/5
Ligne 9 Kipé kolo dola pi gomi …/5
Ligne 10 Boda banso lipa roga faba …/5
1 Auto-stop. Mettez une croix X si l’exercice a été arrêté par manque de réponses correctes ou
par incapacité de lecture à la première ligne.
2 Nombre total de mots inventés lus :
3 Nombre total de mots inventés non lus (pas de réponse) :
5 Temps total si tous les mots sont lus en moins de 60 secondes :
6 Nombre de mots inventés correctement lus en 1 minute
Section 8 : LECTURE DU TEXTE ET COMPREHENSION 
 
Maintenant tu vas répondre à
quelques questions sur l’histoire
Réponses de l’écolier
Texte de lecture Mots Questions Correct Incorrect Pas de
réponse
Mon école
1. Mon école est jolie. Elle 
est à côté de ma maison.
2. Elle a quatre classes. Sa 
cour est grande et
propre.
3. Dans la cour de mon 
école, on trouve des 
arbres.
4. Je joue dans la cour avec 
mes camarades.
5. J’aime mon école.
…/ 2
…/11
… / 10
… / 10
… / 8
… / 3
1. Où est mon école ? (à côté de ma 
maison)
2. Mon école a combien de 
classes ? (quatre)
3. Comment est sa cour ? (grande, 
propre ou grande et propre)
4. Qu’est-ce qu’on trouve dans la 
cour de mon école ? arbres)
5. Avec qui je joue ? (mes 
camarades)
Arrêt précoce: 
Nombre total de mots correctement lus en un minute:
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
31
Nombre total de secondesindiquées sur le chronomètre si l’ecolier a tout lu: ___ 
Nombre total de bonne(s) reponse(s) donnee(s): _______
Section 9 : COMPREHENSION DE L’AUDITION 
Maintenant je vais te poser quelques questions sur le texte que tu as écouté.
Questions Réponse correcte (ne pas
lire)
Réponses de l’écolier
Correct Incorrect Pas de réponse
1. De quelle fête s’agit- il dans le texte
?
(tabaski)
2. Qui a invité ses amis à la maison ? (Le papa de Moussa)
3. Depuis quand maman est à la 
cuisine ?
(le matin)
4. Que prépare maman ? (de bon repas)
5. A qui moussa donne t-il à manger ? (à tout le monde)
Réponsestotales correctes ……/5
Section 10 : DICTEE OU ECRITURE DES MOTS: Baké va à l’école le matin
Feuille de marquage de l’énumérateur [à remplir à la fin de la journée après
que toutes les évaluations sont terminées]
Mot de la dictée Correct
= 2
Partiellement
correct = 1
Incorrect
= 0
Pas
d’écriture
Note
Baké 2 1 0
va 2 1 0
à 2 1 0
l’école 2 1 0
le 2 1 0
matin. 2 1 0
Nombre d’espace
entre les mots
2
5 espaces
1
Moins de 5
espaces
0
Majuscule au début
de la phrase (B)
2 1 0
Point à la fin de la
phrase
2 0
NOTE TOTALE ………./18
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 4 – EGRA SURVEY (CONTINUED) 
32
Feuille de l’écolier pour le texte de la dictée
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
33
APPENDIX 5 – PARENT SURVEY 
QUESTIONNAIRE – PARENTS
Cette section doit être remplie par l’intervieweur (le questionnaire en ligne aura un format déroulant):
Date: (JJ-MM-AAAA)
Nom de l’intervieweur: (spécifier)
Numéro d’enumerateur: (spécifier)
Département: (spécifier)
Commune: (spécifier)
Numéro d’identification 
unique de l’école:
(spécifier)
Nom de l’école: (spécifier)
École de Contrôle ou de 
Traitement:
(spécifier)
Voulez-vous participer à ce sondage? 1. Oui 2. Non 
NB : Commencer à administrer le questionnaire si le consentement est donné
Bonjour. Je m’appelle __________________ et j’aide à mener l’évaluation finale du programme 
FFE mis en œuvre par CRS. Je travaille avec Advisem et CRS. Nous menons un sondage et nous 
aimerions que vous y participiez. Je voudrais vous poser des questions sur les enfants de cette 
école. Cette information nous aidera à évaluer les activités mises en œuvre par CRS pour améliorer 
les services d’éducation, fournir plus de nourriture et de médicaments, et améliorer la santé et 
l’hygiène. Le sondage prend normalement 20 minutes. L’information que vous nous donnerez sera 
strictement confidentielle et ne sera pas montrée à d’autres personnes. Votre identité ne sera pas 
liée à vos réponses. Votre participation est volontaire et vous pouvez choisir de ne pas répondre 
à certaines questions, ou même à toutes les questions. Votre participation à de futurs programmes 
CRS ne dépend pas de vos réponses à cette enquête. Cependant, nous espérons que vous 
participerez à cette enquête puisque votre opinion est importante.
NB : Proposer au répondant de lui fournir les contacts de CRS s’il le désire
Maintenant, avez-vous des questions à poser sur le sondage ? 
Puis-je commencer l’interview maintenant ?
Présentation
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
34
Si le parent ne peut pas participer, s’il vous plaît donnez-en les raisons.
S’il vous plaît donnez-en les raisons pour lesquelles vous ne voulez pas participer.
:……………………………………………………………………………………………………………………………………………………………………..
…………………………………………………………………………………………………………………………………………………………………………………….
N° Questions Réponses Saut 
1. Caractéristiques du parent 
101
Le répondant est-il de sexe masculin ou féminin?
(NB : Observez, ne demandez pas)
1. Féminin 
2. Masculin
102
Quel âge avez-vous?
(NB : mettre 99 si la personne ne veut pas répondre)
/_____/ ans
103
Combien de personnes vivent dans votre ménage, y 
compris vous-même?
(NB : mettre 99 si la personne ne veut pas répondre)
/______/ personnes
104
Quel est votre niveau d’éducation? 1. Aucun
2. Primaire
3. Cycle 1 secondaire
4. Cycle 2 secondaire
5. Cycle 1 supérieur.
6. Cycle 2 supérieur
7. Alphabétisé
8. Autre (à spécifier)
9. Pas de réponse
105
Quelle est votre occupation principale? 1. Sans emploi
2. Ménagère
3. Travail agricole
4. Propriétaire foncier 
5. Journalier(ière)
6. Marchand(e)
7. Travailleur(euse) de 
bureau
8. Artisan(e) 
9. Retraité(e)
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
35
N° Questions Réponses Saut 
10. Pas de réponse 
11. Autre (spécifier)
106
Combien d’enfants avez-vous?
(NB : mettre 99 si la personne ne veut pas répondre)
/______/ enfants
107
Combien de vos enfants sont inscrits dans cette 
l’école?
1. 1
2. 2
3. 3
4. 4
5. 5
6. 6
7. 7
Le nombre de ‘108’ dépendra 
de cette réponse. Par 
exemple, si le répondant a 
deux enfants, 108 (1-18) 
s’ouvrira deux fois.
108
Je voudrais vous poser des questions sur chacun de vos enfants inscrits à l’école. Nous commencerons par votre plus 
jeune enfant inscrit à l’école (par exemple, CI) et terminer avec votre enfant le plus âgé inscrit à l’école (par exemple, 
CM2).
N° Questions Réponses Saut 
1 Nom et prénoms
(NB : 
Mettre ‘pas de répondre’ si la personne ne veut pas 
répondre)
/______/
2 Sexe 1. Féminin 
2. Masculin
3 Age 
(NB : Laissez le champ vide si aucune réponse.)
/______/
4 Classe 1. Cours d’Initiation (CI)
2. Cours Préparatoire (CP)
3. Cours élémentaire 1
4. CE2
5. CM1
6. CM2
5 {nom} était-(il) (elle) malade en décembre 2017?8 1. Oui 
2. Non
Si 1 6
6 De quelle maladie a-t-il (elle) souffert? 1. Infections respiratoires 
aiguës
2. Diarrhée
3. Eruptions/rougeurs, 
maladies 
dermatologiques
Choisissez la 
maladie 
principale.
 
8 This reference to Decembre 2017 may need to be changed once the final dates of the data gathering exercise are 
determined. 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
36
4. Anémie 
5. Paludisme
6. Autre (spécifier)
7 Est-ce que {nom} a dit qu’il (elle) avait faim à l’école la 
semaine dernière?
NB : Le jour de l’école est de 8h00 à 17h00.
1. Oui 
2. Non
3. Ne sait pas / Pas de réponse
8 Est-ce que {nom} a mangé pendant la journée d’école 
hier?
NB : Le jour de l’école est de 8h00 à 17h00. 
Si le jour précédent n’était pas un jour d’école, 
demandez spécifiquement le dernier jour d’école.
1. Oui 
2. Non
3. Ne sait pas / Pas de réponse
9 Combien de fois a il/elle a mangé hier; pendant la 
journée à l’école / la maison ou la nuit, à la maison ou 
à l’extérieur de la maison?
1. 0
2. 1
3. 2
4. 3
5. 4
6. 5
7. 6
8. 7 ou plus
9. Ne sait pas / Pas de 
réponse
Si ‘0’, mettez 
‘Non’ pour les 
questions 11-17.
10 Maintenant, j’aimerais vous poser des questions sur les liquides ou les aliments que {nom} a pris hier 
pendant la journée à l’école / la maison ou la nuit, à la maison ou à l’extérieur de la maison.
11 Grains, racines et tubercules 
Pain, Riz, Pâtes, maïs, Masa, Paté, autre nourriture à 
base de grains, Pomme de terre, tarot, patates 
blanches, ignames blanches, manioc, ou autres 
tubercules
1. Oui 
2. Non
12 Légumineuses et noix
Des aliments à base de haricots, pois, lentilles, ou de 
noix (Y compris Alélé/Toubani)
1. Oui 
2. Non
13 Produits laitiers (lait, yogourt, fromage, beurre) 1. Oui 
2. Non
14 Aliments à chair (viande, poisson, volaille et foie / 
abats)
Du foie, rognon, cœur, autres abats ou viandes 
provenant d’organes animaux ; viande, comme du 
bœuf, du porc, de l’agneau, de la chèvre, du poulet ou 
du canard ; Du poisson ou des crustacés frais ou 
séchés ; Des escargots, des insectes, ou autres petits 
aliments contenant des protéines
1. Oui 
2. Non
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
37
15 Des œufs 1. Oui 
2. Non
16 Aliments enrichis en vitamine A, y compris l’huile 
végétale, les fruits et légumes.
Carottes ou patates douces qui sont jaunes ou oranges 
à l’intérieur, mangues, papayes mûres, pastèques
Des légumes feuilles vert foncé (n’importe lesquels : La 
salade ou laitue, choux, haricot vert, épinards, crin crin, 
feuille de baobab, gombo, afonu. Des aliments 
préparés avec de l’huile de palme rouge, de noix de 
palme, ou de la pulpe de noix de palme.
1. Oui 
2. Non
17 Autres fruits et légumes
Autres fruits (Goyave, Banane, Citron, Orange, Ananas) 
y autres légumes ; Une sauce (tomate ou oignon)
1. Oui 
2. Non
109
Y-a-t-il une association de parents d’élèves à ${nom de 
l’école}?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 110
Si 2 111
Si 3 112
110
Si oui, Faites-vous partie de cette association? 1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
111
Si non, Voudriez-vous qu’il y ait une association de 
parents d’élèves (APE) dans l’école?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
112
Y-a-t-il une association de mères d’élèves (AME) à 
${nom de l’école}?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 113
Si 2 114
Si 3 115
Seulement pour les femmes.
113
Faites-vous partie de cette association? 1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Seulement pour les femmes.
114
Voudriez-vous qu’il y ait une association de mères 
d’élèves (AME) dans l’école?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Seulement pour les femmes.
115
Pensez-vous qu’il y a des salles de classe ayant besoin 
de réparations à l’école?
1. Oui 
2. Plus ou moins
3. Non
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
38
4. Ne sait pas / Pas de 
réponse
116
Pensez-vous qu’il y a des latrines ayant besoin de 
réparations à l’école?
1. Oui 
2. Plus ou moins
3. Non
4. Ne sait pas / Pas de 
réponse
117
Considérez-vous que le temps que vos enfants passent 
à l’école en vaut la peine?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
118
Pourquoi considérez-vous l’éducation de vos enfants 
comme importante? (Réponse multiple) (NB : Ne pas 
lire les réponses) (Seulement choisi exactement ce qui a 
été dit.)
1. Peut obtenir un emploi
2. Peut prendre soin des 
parents
3. Meilleurs emplois / 
opportunités 
économiques
4. Meilleure vie / avenir
5. Pour une meilleure 
société / communauté
6. Être indépendant / 
surmonter l’adversité / 
avancer
7. Autres connaissances / 
études
8. Meilleur statut social
9. Ne sait pas
10. Autre 1 (spécifier)
11. Autre 2 (spécifier)
12. Autre 3 (spécifier)
119
D’après vous, quels sont les moments les plus 
importants pour se laver les mains? (Cochez toutes les 
réponses qui s’appliquent)
(NB : Ne pas lire les réponses; Assurez-vous de 
demander s’ils connaissent d’autres raisons.)
1. Après défécation 
2. Avant de manger 
3. Après avoir lavé les 
enfants/et les couches 
culottes 
4. Après le nettoyage des 
latrines 
5. Après le nettoyage de 
pot 
6. Avant la préparation du 
repas 
7. Avant le repas de 
l’enfant 
8. Après le repas
9. Ne sait pas
10. Autre (spécifier)
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
39
120
Pour quelles raisons doit-on se laver les mains avec du 
savon? (Cochez toutes les réponses qui s’appliquent)
(NB : Ne pas lire les réponses; Assurez-vous de 
demander s’ils connaissent d’autres raisons.)
1. Prévenir la diarrhée 
2. Prévenir d’autres 
maladies 
3. Retirer les germes 
4. Empêcher la saleté de 
se mettre dans la 
bouche 
5. Empêcher la saleté de 
se mettre dans la 
nourriture 
6. Sent bon
7. Ne sait pas
8. Autre (spécifier)
121
Qu’est-ce que les familles peuvent faire pour améliorer 
la qualité de l’eau de boisson? (NB : Ne pas lire les 
réponses; Assurez-vous de demander s’ils connaissent 
d’autres manières) (Réponse multiple)
1. Bouillir l’eau 
2. Ajouter de l’eau de 
Javel 
3. Ajouter une solution 
chlorée (Sur’ Eau) 
4. Ajouter des comprimés 
de chlore (Aqua Tabs) 
5. Filtrer l’eau à travers un 
tissu 
6. La laisser reposer 
7. Utiliser un filtre en 
Céramique 
8. Utiliser un filtre bio￾Sable 
9. Désinfection solaire 
10. Garder l’eau dans un 
récipient couvert 
11. Rien 
12. Ne sait pas
13. Autre (spécifier)
122
Quels produits peuvent être ajoutés à l’eau pour la 
rendre plus potable pour la boisson? (NB: Ne pas lire 
les réponses Assurez-vous de demander s’ils 
connaissent d’autres manières)
(Réponse multiple)
1. Liquide de chlore (Sur 
‘Eau) 
2. Pastilles de chlore 
(Aqua Tabs) 
3. (PuR, Water maker) 
4. Autres présentation de 
chlore (granulés HTH, 
chlore, eau de Javel, 
autres) 
5. Iode (gouttes ou 
comprimés) 
6. Permanganate 
7. Aucun 
8. Ne sait pas 
9. Autre (spécifier)
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
40
123
La dernière fois où votre enfant le plus jeune a 
déféqué, où l’a-t-il fait? (Une réponse)
1. Toilettes
2. A utilisé les latrines 
3. A utilisé un pot 
4. A utilisé une couche￾culotte lavable 
5. A utilisé une couche￾culotte jetable 
6. Est venu dans la 
maison/cour 
7. Est allé hors des lieux 
8. A fait dans ses 
vêtements 
9. Ne sait pas
10. Autre (spécifier)
124
La dernière fois où votre enfant le plus jeune a 
déféqué, où a-t-on jeté ses excréments? (Une réponse)
1. Toilettes
2. Dans les latrines 
3. Enterrés 
4. Fosse ou bac à ordures 
5. Dans la cour 
6. Hors des lieux 
7. Latrines publiques
8. Dans l’évier ou le 
baquet 
9. Dans la conduite d’eau 
10. Ne sait pas
11. Autre (spécifier)
125
Quelle sorte de toilettes votre ménage utilise-t-il 
généralement?
1. Toilettes modernes
2. Toilettes traditionnelles
3. En plein air
Si 3 128
126
Est-ce que vous partagez ces toilettes avec d’autres 
ménages?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 127
127 Combien de ménages partagent ces toilettes, y compris 
vous-même? (NB : Peut être un approximatif) /_____/ ménages
128
Avez-vous entendu les messages radiophoniques FFE 
(CRS, WEI) en faveur de l’éducation?
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 129
129
Ces messages radio ont-ils changé votre opinion sur 
l’éducation?
1. Oui 
2. Plus ou moins
3. Non
4. Ne sait pas / Pas de 
réponse
130
Participez-vous à un système d’épargne et de crédit? 1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 131
Si 2 ou 3 133
131 Qui a organisé ce système? (Réponse multiple) 1. Gouvernement
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
41
2. CRS/FFE/World 
Education
3. SFD (système financier 
décentralisé)
4. Famille ou groupe 
d’amis
5. tontinier ambulants
6. Autre (spécifier)
132
À quelle fin avez-vous utilisé le(s) prêt(s)? (Réponse 
multiple) 
1. Éducation des enfants
2. Des raisons médicales
3. Funérailles
4. Besoins quotidiens (p. 
Ex. Nourriture)
5. Réparations à domicile
6. Entreprise
7. N’a pas pris de prêt
8. Autre (spécifier)
133
Dans l’ensemble, comment pensez-vous que le 
programme FFE a amélioré la qualité de 
l’apprentissage dans l’école fréquentée par votre 
enfant / vos enfants? En utilisant une échelle de 5 
(Énormément) à 1 (Très peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. 1 - Très peu
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
134
Dans l’ensemble, comment pensez-vous que le 
programme FFE a amélioré la qualité de la santé et de 
l’hygiène dans l’école fréquentée par votre enfant / vos 
enfants? En utilisant une échelle de 5 (Énormément) à 
1 (Très peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. 1 - Très peu
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
135
Dans l’ensemble, comment pensez-vous que le 
programme FFE a amélioré la qualité de vie de votre 
famille? En utilisant une échelle de 5 (Énormément) à 1 
(Très peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. Ne sait pas / Pas de 
réponse
6. 1 - Très peu
Seules les écoles de 
traitement
136
Si le parent ne peut pas participer, s’il vous plaît donnez-en les raisons.
NB : Question à répondre par l’agent enquêteur
Raison 1 :……………………………………………………………………………………………………………………………………………………………………..
…………………………………………………………………………………………………………………………………………………………………………………….
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 5 – PARENT SURVEY (CONTINUED) 
42
137
Faites-nous part de vos commentaires, si vous en avez éventuellement?
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
138
En tant que l’intervieweur, veuillez fournir des commentaires supplémentaires (par exemple, problèmes, 
clarifications, commentaires).
…………………………………………………………………………………………………………………………………………………………………………………
Merci beaucoup d’avoir pris le temps de remplir ce sondage. 
Fin du questionnaire 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
43
APPENDIX 6 – PRINCIPAL SURVEY 
QUESTIONNAIRE – DIRECTEUR
Cette section doit être remplie par l’intervieweur (le questionnaire en ligne aura un format déroulant):
Date: (JJ-MM-AAAA)
Nom de l’intervieweur: (spécifier)
Numéro d’enumerateur: (spécifier)
Département: (spécifier)
Commune: (spécifier)
Numéro d’identification 
unique de l’école:
(spécifier)
Nom de l’école: (spécifier)
École de Contrôle ou de 
Traitement:
(spécifier)
Voulez-vous participer à ce sondage? 1. Oui 2. Non
NB : Commencer à administrer le questionnaire si le consentement est donné
Êtes-vous le Directeur ou Adjoint au Directeur de l’école? 1. Directeur 2. Adjoint au Directeur
Bonjour. Je m’appelle __________________ et j’aide à mener l’évaluation finale du programme 
FFE mis en œuvre par CRS. Je travaille avec Advisem et CRS. Nous menons un sondage et nous 
aimerions que vous y participiez. Je voudrais vous poser des questions sur votre école et le 
personnel enseignant. Cette information nous aidera à évaluer les activités mises en œuvre par 
CRS pour améliorer les services d’éducation, fournir plus de nourriture et de médicaments, et 
améliorer la santé et l’hygiène. Le sondage prend normalement 20-30 minutes. L’information que 
vous nous donnerez sera strictement confidentielle et ne sera pas montrée à d’autres personnes. 
Votre identité ne sera pas liée à vos réponses Votre participation est volontaire et vous pouvez 
choisir de ne pas répondre à certaines questions, ou même à toutes les questions. Votre 
participation à de futurs programmes CRS ne dépend pas de vos réponses à cette enquête. 
Cependant, nous espérons que vous participerez à cette enquête puisque votre opinion est 
importante.
NB : Proposer au répondant de lui fournir les contacts de CRS s’il le désire
Maintenant, avez-vous des questions à poser sur le sondage ? 
Puis-je commencer l’interview maintenant ?
Présentation
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
44
N° Questions Réponses Saut 
2. Caractéristiques du directeur et capacités 
101 Le répondant est-il de sexe masculin ou féminin? 0. Féminin 
1. Masculin
102 Quel âge avez-vous? 
(NB : mettre 99 si la personne ne veut pas répondre)
/___/___/ ans
103
Depuis combien d’années travaillez-vous en tant que 
directeur (ou Directeur adjoint) de votre école 
actuelle?
/___/___/ ans
104
Quel est le diplôme académique le plus élevé que 
vous avez obtenu en formation initiale?
1. BEPC
2. CAP (lycée technique)
3. BAC, DEAT
4. DEUG, DUES
5. Licence
6. Maitrise
7. Master/DEA
8. Autre (à spécifier) 
105
Quel est le diplôme professionnel le plus élevé que 
vous avez obtenu?
1. Aucun
2. CEAP
3. CAP
4. Autre (à spécifier)
106
Avez-vous (ou d’autres administrateurs / officiels de 
l’école ont) reçu une formation en renforcement de 
capacité depuis septembre 2014?
(NB : Se concentrer uniquement sur la formation 
depuis septembre 2014)
1. Oui 
2. Non
3. Ne sait pas
Si 2 ou 3
Q113
107 Si oui, combien de formations avez-vous (ou d’autres 
administrateurs / officiels de l’école ont-ils) reçues?
/ ____ /
108
Qui a organisé cette formation? (Réponse multiple)
NB : Commenter le plus récent 
a. Gouvernement
b. CRS/ World Education 
/FFE
c. Autre (à spécifier)
109
Quelle est votre appréciation de cette formation?
NB : Commenter le plus récent
a. Excellente
b. Bien
c. Passable
d. Peu appréciée
e. Médiocre
110
Quelle est votre appréciation de cette formation? S’il vous plaît fournir vos commentaires.
NB : Commenter le plus récent
………………………………………………………………………………………………………………………………………………………………………..
111
A-t-elle été bien organisée?
NB : Commenter le plus récent
1. Oui 
2. Plus ou moins
3. Non
112
Le contenu de la formation vous a-t-il semblé utile?
NB : Commenter le plus récent
1. Oui 
2. Plus ou moins
3. Non
Si 3
Q201
113
Est-ce que vous ou d’autres administrateurs / officiels 
de l’école utilisez (utilisent) les nouvelles techniques 
1. Oui 
2. Plus ou moins
3. Non
Seules les écoles de 
traitement
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
45
N° Questions Réponses Saut 
ou les nouveaux outils enseignés dans la formation 
fournie par le programme FFE (CRS, WEI)?
114 S’il vous plaît fournir un exemple (des nouvelles 
techniques ou les nouveaux outils).
/ ____ / Seules les écoles de 
traitement
2. Renforcements des capacités des enseignants 
201
Le personnel enseignant a-t-il reçu une formation sur 
l’enseignement de la lecture depuis septembre 2014?
(NB : Se concentrer uniquement sur la formation 
depuis septembre 2014)
1. Oui 
2. Non
3. Ne sait pas
Si 2 ou 3
Q211
202 Si oui, environ combien de formations ont-ils reçues? / ____ /
203
Qui a organisé cette formation? (Réponse multiple)
NB : Commenter le plus récent 
1. Gouvernement
2. CRS/ World Education 
/FFE
3. Autre (à spécifier)
204
Quelle est votre appréciation de cette formation?
NB : Commenter le plus récent
1. Excellente
2. Bien
3. Passable
4. Peu appréciée
5. Médiocre
205
Quelle est votre appréciation de cette formation? S’il vous plaît fournir vos commentaires.
NB : Commenter le plus récent
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
206
A-t-elle été bien organisée?
NB : Commenter le plus récent
1. Oui 
2. Plus ou moins
3. Non
207
Le contenu de la formation vous a-t-il semblé utile?
NB : Commenter le plus récent
1. Oui 
2. Plus ou moins
3. Non
208
Cette formation cadrait-elle avec les besoins 
professionnels des enseignants en ce qui concerne la 
lecture?
1. Oui 
2. Plus ou moins
3. Non
209
Quel élément dans cette formation vous a semblé particulièrement utile pour l’enseignement de la lecture?
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
210
Utilisent-ils les nouvelles techniques en lecture? 1. Oui 
2. Plus ou moins
3. Non
211
Combien de minutes consacrent-ils par jour pour la 
lecture aux enfants au CI?
NB: Passer si non applicable
/___/___/___/ minutes
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
46
N° Questions Réponses Saut 
212
Combien de minutes consacrent-ils par jour la lecture 
aux enfants au CP?
NB: Passer si non applicable
/___/___/___/ minutes
213
Votre école suit-elle le programme national 
d’enseignement (écriture et lecture) et utilise-t-elle 
les matériels didactiques connexes?
1. Oui 
2. Non
3. Ne sait pas
214
Le personnel enseignant a-t-il suivi une formation en 
santé et hygiène scolaire?
1. Oui 
2. Non
3. Ne sait pas
Si 2 ou 3
Q301
215
Si oui, qui a organisé cette formation? a. Gouvernement
b. CRS/FFE
c. Autre (à spécifier)
3. Infrastructures de l’école
301 Combien de cours compte votre école? /______/
302
S’il vous plaît indiquer les classes de votre école. 7. Cours d’Initiation (CI)
8. Cours Préparatoire (CP)
9. Cours élémentaire 1
10. CE2
11. CM1
12. CM2
Sélectionnez tout ce qui 
s’applique.
Des questions spécifiques 
pour chaque classe 
apparaîtront si elles sont 
sélectionnées ici.
303 Le nombre de salles de classe est-il en adéquation 
avec le nombre de cours?
1. Oui
2. Non
304
Dans quel état se trouvent les salles de classe de 
votre école?
1. Bonnes conditions
2. Conditions moyennes
3. Mauvaises conditions
4. Ne sait pas / Pas de 
réponse
305
A votre avis, quelles sont les trois plus importantes 
priorités pour améliorer votre établissement scolaire?
a. Améliorer les salles de 
classe existentes
b. Améliorer les latrines 
existentes
c. Améliorer les lavabos 
existentes
d. Construire de nouvelles 
salles de classes
e. Construire de nouvelles 
latrines
f. Construire de nouveaux 
lavabos
g. Construire une clôture
a. Renforcer l’effectif du 
personnel enseignant
b. Doter (ou renforcer) la 
cantine scolaire
h. Autre (à spécifier)
4. Perception des directeurs 
401 Votre école a-t-elle une APE? 1. Oui 
2. Non
Si 1 Q403
Si 2 Q402
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
47
N° Questions Réponses Saut 
3. Ne sait pas
402
Si non, pourquoi? 1. Restriction financière 
des parents
2. Restriction de temps des 
parents
3. Aucun intérêt des 
parents
4. Pas d’intérêt des écoles 
d’avoir une association 
de parents
5. Ne sait pas
6. Autre (à spécifier)
403
Est-t-elle fonctionnelle?
NB : Fonctionnelle = L’APE organise des réunions et 
tient des procès-verbaux
1. Très active
2. Modérément active
3. Pas du tout active
4. Ne sait pas
Si 3 Q404
404
Si non, pourquoi? 1. Restriction financière 
des parents
2. Restriction de temps des 
parents
3. Aucun intérêt des 
parents
4. Pas d’intérêt des écoles 
d’avoir une association 
de parents
5. Ne sait pas
6. Autre (à spécifier)
405
Votre école a-t-elle une AME? 1. Oui 
2. Non
3. Ne sait pas
Si 1 Q407
Si 2 Q406
406
Si non, pourquoi? 1. Restriction financière 
des mères
2. Restriction de temps des 
mères
3. Aucun intérêt des mères
4. Pas d’intérêt des écoles 
d’avoir une association 
de mères
5. Ne sait pas
6. Autre (à spécifier)
407
Est-t-elle fonctionnelle? 
NB : Fonctionnelle = L’AME organise des réunions et 
tient des procès-verbaux
1. Très active
2. Modérément active
3. Pas du tout active
4. Ne sait pas
Si 3 Q408
408
Si non, pourquoi? 1. Restriction financière 
des mères
2. Restriction de temps des 
mères
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
48
N° Questions Réponses Saut 
3. Aucun intérêt des mères
4. Pas d’intérêt des écoles 
d’avoir une association 
de mères
5. Ne sait pas
6. Autre (à spécifier)
409
Les parents des élèves suivent-ils un programme 
d’alphabétisation (écriture et lecture)?
1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
410
Organisez-vous le suivi pédagogique de vos 
enseignants en lecture?
1. Oui, régulièrement
2. Oui, mais 
irrégulièrement
3. Non, pas du tout 
4. Ne sait pas / Pas de 
réponse
411
Vos enseignants sont-ils assidus à l’école? 1. Oui, toujours
2. La plupart du temps
3. Non, pas du tout 
4. Ne sait pas / Pas de 
réponse
412
Savez-vous si un logement est offert à vos 
enseignants?
1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
413
Votre école a-t-elle une cantine fonctionnelle? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1
Q415
Si 2
Q414
414
Si non, pourquoi?
NB : finir de noter les raisons éventuelles avant de réaliser le saut Q501
Raison : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
415
Votre cantine a-t-elle un comité de gestion? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
416
La cantine a-t-elle un impact sur la concentration des 
élèves? (impact=changement plus ou moins profond)
1. Oui, beaucoup
2. Oui, un peu
3. Non, pas du tout 
4. Ne sait pas / Pas de 
réponse
417
Pouvez-vous expliquer comment la cantine a eu un impact sur la concentration des élèves? Si possible, trois impacts. 
(impact=changement plus ou moins profond)
NB : l’impact peut être positif ou négatif
---------------------------------------------------------------------------------------------
Impact 1 : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
49
N° Questions Réponses Saut 
Impact 2 : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
Impact 3 : ……………….….……………………………………………………………………………………………………………………………………………….
418
La cantine a-t-elle un impact sur l’apprentissage? 
(impact=changement plus ou moins profond)
NB : l’impact peut être positif ou négatif
1. Oui, beaucoup
2. Oui, un peu
3. Non, pas du tout 
4. Ne sait pas / Pas de 
réponse
419
Pouvez-vous expliquer comment la cantine a eu un impact sur l’apprentissage? Si possible, trois impacts. 
(impact=changement plus ou moins profond)
NB : l’impact peut être positif ou négatif
Impact 1 : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
Impact 2 : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
Impact 3 : ……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
420
Votre école distribue-elle des rations sèches aux 
apprenants?
1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1
Q421
Si 2 ou 3 
Q501
421
Cette distribution a-t-elle un impact sur 
l’apprentissage?
1. Oui, beaucoup
2. Oui, un peu
3. Non, pas du tout
4. Ne sait pas / Pas de 
réponse
5. Divers équipements et matériels dont dispose l’école 
501
Avez-vous les matériels suivants pour votre cantine?
NB : uniquement pour les écoles sous intervention du 
CRS/FFE 
a. Au moins 2 marmites 
n°25 avec couvercle
b. Au moins 2 Marmites 
n°20 avec couvercle
c. Au moins 3 Bassines en 
Aluminium
d. Au moins 3 Casseroles
e. Au moins 2 Seaux 
métalliques
f. Au moins 350 
Bols/assiettes en 
plastique
g. Au moins 350 Cuillères à 
soupe importées
h. Au moins 350 Gobelets
1. oui 2. non
/___/a
/___/b
/___/c
/___/d
/___/e
/___/f
/___/g
/___/h
Si 1 pour 
413
502 Combien d’élèves sont actuellement en CI? /___/___/ élèves
503 Combien d’élèves sont actuellement en CP? /___/___/ élèves
504 Combien d’élèves sont actuellement en CE1? /___/___/ élèves
505 Combien d’élèves sont actuellement en CE2? /___/___/ élèves
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
50
N° Questions Réponses Saut 
506 Combien d’élèves sont actuellement en CM1? /___/___/ élèves
507 Combien d’élèves sont actuellement en CM2? /___/___/ élèves
508
Au CI combien d’élèves n’ont pas le manuel de 
français?
NB : Laisser vide (sauter) si le répondant ne sait pas 
(520-527).
/___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
509
Au CI, combien d’élèves n’ont pas le cahier d’activité 
en français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
510
Au CP, combien d’élèves n’ont pas le manuel de 
français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
511
Au CP, combien d’élèves n’ont pas le cahier d’activité 
en français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
512
Au CE1, combien d’élèves n’ont pas le manuel de 
français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
513
Au CE2, combien d’élèves n’ont pas le manuel de 
français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
514
Au CM1, combien d’élèves n’ont pas le manuel de 
français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
515
Au CM2, combien d’élèves n’ont pas le manuel de 
français? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
516
Au CI, combien d’élèves n’ont pas les fournitures 
nécessaires?
NB : Laisser vide (sauter) si le répondant ne sait pas 
(528-533).
/___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
517
Au CP, combien d’élèves n’ont pas les fournitures 
nécessaires? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
518
Au CE1, combien d’élèves n’ont pas les fournitures 
nécessaires? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
519
Au CE2, combien d’élèves n’ont pas les fournitures 
nécessaires? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
520
Au CM1, combien d’élèves n’ont pas les fournitures 
nécessaires? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
521
Au CM2, combien d’élèves n’ont pas les fournitures 
nécessaires? /___/___/ élèves
Doit être inférieur ou égal au 
nombre réel d’élèves dans la 
classe.
522
Les élèves dans votre école ont-ils reçu des
comprimés de vermifuges durant une campagne de 
santé l’année dernière?
1. Oui, tous
2. Oui, mais pas tous
3. Non, aucun
4. Ne sait pas / Pas de 
réponse`
Si 1 ou 2
Q535
523 Combien d’élèves ont reçu/reçoivent les comprimés? /___/___/___/ élèves
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
51
N° Questions Réponses Saut 
6. Présence des enseignants à l’école 
601
Vos enseignants ont-ils été assidus à l’école au cours 
du trimestre dernier (18 Septembre - 20 Décembre 
2017)?
1. Oui, vraiment
2. Oui, plus ou moins
3. Non
4. Ne sait pas / Pas de 
réponse`
602
Combien de jours vos enseignants ont-ils été à l’école 
au cours du trimestre dernier (18 Septembre - 20 
Décembre 2017)?
/_____/ jours
603
Combien de jours ont été perdus par l’enseignant du 
CI au cours du trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
604
Combien de jours ont été perdus par l’enseignant du 
CP au cours du trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
605
Combien de jours ont été perdus par l’enseignant du 
CE1 au cours du trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
606
Combien de jours ont été perdus par l’enseignant du 
CE2 au cours du trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
607
Combien de jours ont été perdus par l’enseignant du 
CM1 au cours du trimestre dernier (18 Septembre -
20 Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
608
Combien de jours ont été perdus par l’enseignant du 
CM2 au cours du trimestre dernier (18 Septembre -
20 Décembre 2017)?
NB : Le directeur devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait 
pas
/_____/ jours
Doit être inférieur ou égal 
au nombre total de jours 
où les enseignants auraient 
dû être présents en classe 
(602).
609
En moyenne, combien de temps les enseignants 
continuent-ils d’enseigner dans votre école?
1. Moins de six mois
2. Six mois à un an
3. Une année à deux ans
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
52
N° Questions Réponses Saut 
4. Deux ans à trois ans
5. Trois ans à quatre ans
6. Quatre ans à cinq ans
7. Plus de cinq ans
8. Ne sait pas / Pas de 
réponse`
610
Quelle est la principale raison pour laquelle un 
enseignant part?
1. Ils ne sont pas payés
2. Un travail mieux 
rémunéré ailleurs
3. Ils veulent déménager 
dans une ville plus 
grande
4. Autre (à spécifier)
611 Nombre d’enseignants de sexe masculins dans votre 
école ?
/_____/
612 Nombre d’enseignants de sexe féminines dans votre 
école ?
/_____/
7. Hygiène à l’école 
701
Chaque classe dispose-t-elle du dispositif de lavage
des mains, incluant savon (ou cendres) et eau?
1. Oui, toutes les classes
2. Oui, mais pas toutes les 
classes
3. Non, aucune classe
4. Ne sait pas / Pas de 
réponse
702
Vos élèves ont-ils accès à l’eau potable? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
703
Vos élèves se lavent-ils les mains avec du savon (ou 
cendres) avant le repas?
1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse`
Si 1 ou 3
Q705
704
Si non, pourquoi? (Une réponse, raison principale)
……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
705
Vos élèves se lavent-ils les mains avec du savon (ou 
cendres) après être allés aux toilette?
1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1 ou 3
Q707
706
Si non, pourquoi? (Une réponse, raison principale)
……………….….……………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
707
Vos élèves ont-ils accès à des latrines fonctionnelles? 1. Oui
2. Non
Si 2 ou 3
Q801
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
53
N° Questions Réponses Saut 
3. Ne sait pas / Pas de 
réponse
708
Leur nombre est-il suffisant? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
709
Les filles ont-elles leur propre latrine? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
Si 2 ou 3
Q801
710
Leur nombre est-il suffisant? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
8. Impression générale 
801
Dans l’ensemble, comment pensez-vous que le 
programme FFE a amélioré la qualité de 
l’apprentissage dans votre école? En utilisant une 
échelle de 5 (Énormément) à 1 (Très peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. 1 - Très peu
(Seules les écoles de 
traitement
802
Dans l’ensemble, comment pensez-vous que le 
programme FFE a amélioré la qualité de la santé et de 
l’hygiène dans votre école? En utilisant une échelle 
de 5 (Énormément) à 1 (Très peu).
7. 5 - Énormément 
8. 4 - Significativement
9. 3 - Assez
10. 2 - Un peu
11. 1 - Très peu
Seules les écoles de 
traitement
803
Dans quelle mesure la cantine de votre école sera-t￾elle durable une fois le programme FFE terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
Seules les écoles de 
traitement
804
S’il vous plaît fournir votre (vos) raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
805
Dans quelle mesure les améliorations de votre école 
en matière d’amélioration de la qualité de 
l’apprentissage seront-elles durables une fois que le 
programme FFE terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
Seules les écoles de 
traitement
806
S’il vous plaît fournir votre (vos) raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
807
Dans quelle mesure les améliorations apportées à 
votre école en matière de santé et d’hygiène seront￾elles durables une fois le programme FFE terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
Seules les écoles de 
traitement
808 S’il vous plaît fournir votre (vos) raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 6 – PRINCIPAL SURVEY (CONTINUED) 
54
N° Questions Réponses Saut 
809 Pendant environ combien de jours votre école a-t-elle 
été fermée à cause de la grève?
/_____/
810
Faites-nous part de vos commentaires, si vous en avez éventuellement?
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
…………………………………………………………………………………………………………………………………………………………………………………….
812 En tant que l’intervieweur, veuillez fournir des commentaires supplémentaires (par exemple, problèmes, 
clarifications, commentaires).
…………………………………………………………………………………………………………………………………………………………………………………
Merci beaucoup d’avoir pris le temps de remplir ce sondage. 
Fin du questionnaire. 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
55
APPENDIX 7 – TEACHER SURVEY 
QUESTIONNAIRE – ENSEIGNANT
Cette section doit être remplie par l’intervieweur (le questionnaire en ligne aura un format déroulant):
Date: (JJ-MM-AAAA)
Nom de l’intervieweur: (spécifier)
Numéro d’enumerateur: (spécifier)
Département: (spécifier)
Commune: (spécifier)
Numéro d’identification 
unique de l’école:
(spécifier)
Nom de l’école: (spécifier)
École de Contrôle ou de 
Traitement:
(spécifier)
École de Rural ou Urbain: (spécifier)
Voulez-vous participer à ce sondage? 1. Oui 2. Non 
NB : Commencer à administrer le questionnaire si le consentement est donné
Bonjour. Je m’appelle __________________ et j’aide à mener l’évaluation finale du programme 
FFE mis en œuvre par CRS. Je travaille avec Advisem et CRS. Nous menons un sondage et nous 
aimerions que vous y participiez. Je voudrais vous poser des questions sur votre classe et vos 
élèves. Cette information nous aidera à évaluer les activités mises en œuvre par CRS pour 
améliorer les services d’éducation, fournir plus de nourriture et de médicaments, et améliorer la 
santé et l’hygiène. Le sondage prend normalement 20 minutes. L’information que vous nous 
donnerez sera strictement confidentielle et ne sera pas montrée à d’autres personnes. Votre 
identité ne sera pas liée à vos réponses. Votre participation est volontaire et vous pouvez choisir 
de ne pas répondre à certaines questions, ou même à toutes les questions. Votre participation à 
de futurs programmes CRS ne dépend pas de vos réponses à cette enquête. Cependant, nous 
espérons que vous participerez à cette enquête puisque votre opinion est importante.
NB : Proposer au répondant de lui fournir les contacts de CRS s’il le désire
Maintenant, avez-vous des questions à poser sur le sondage? 
Puis-je commencer l’interview maintenant ?
Présentation
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
56
N° Questions Réponses Saut 
3. Caractéristiques de l’enseignant 
101Le répondant est-il de sexe masculin ou féminin ? 1. Féminin 
2. Masculin
102 Quel âge avez-vous?
(NB : mettre 99 si la personne ne veut pas répondre)
/___/___/ ans
103
Quel est la classe la plus élevée que vous avez fréquentée ? 1. 6
e
2. 5é
3. 4é
4. 3é
5. 2
nde
6. 1ére
7. Terminale
8. Université
9. Autre (à spécifier)
104
Quel est le diplôme académique le plus élevé que vous avez 
obtenu en formation initiale?
1. CEP
2. BEPC
3. CAP (lycée technique)
4. BAC, DEAT
5. DEUG, DUES
6. Licence
7. Maitrise
8. Master/DEA
9. Autre (à spécifier)
105
Quel est le diplôme professionnel le plus élevé que vous 
avez obtenu?
1. Aucun
2. CEAP
3. CAP
4. Autre (à spécifier)
106 Depuis combien d’années enseignez-vous? /_____/
107
Quel est votre statut? 1. APE
2. ACE
3. Conventionné
4. Communautaire
5. Stagiaire/volontaire
6. Autre (à spécifier)
4. Informations sur la classe gardée par l’enseignant 
201
Quelle(s) classe(s) enseignez-vous? (Réponse multiple) 1. Cours d’Initiation (CI)
2. Cours Préparatoire (CP)
3. Cours élémentaire 1
4. CE2
5. CM1
6. CM2
202 Votre classe compte combien de garçons /_____/
203 Votre classe compte combien de filles /_____/
204
Combien de garçons ont eu plus de 14 absences non 
justifiées de l’école le trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : L’enseignant devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait pas
/_____/ Doit être inférieur 
ou égal au nombre 
réel de garçons 
dans la classe.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
57
N° Questions Réponses Saut 
205
Combien de filles ont eu plus de 14 absences non justifiées 
de l’école le trimestre dernier (18 Septembre - 20 
Décembre 2017)?
NB : L’enseignant devrait se référer à son registre de 
présence. Laisser vide (sauter) si le répondant ne sait pas
/_____/ Doit être inférieur 
ou égal au nombre 
réel de filles dans la 
classe.
206
Combien de garçons ont été absents en décembre 2017?
NB : si un enfant est absent un jour, cela est considéré 
comme absent: Laisser vide (sauter) si le répondant ne sait 
pas
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
réel de garçons 
dans la classe.
207
Combien de filles ont été absentes en décembre 2017?
NB : si un enfant est absent un jour, cela est considéré 
comme absent: Laisser vide (sauter) si le répondant ne sait 
pas
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
réel de filles dans la 
classe.
208
Combien de garçons ont été absents parce qu’ils étaient 
malades en décembre 2017?
(NB : si un enfant est absent un jour, cela est considéré 
comme absent / Laisser vide (sauter) si le répondant ne sait 
pas)
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
total de garçons 
absents.
209
Combien de filles ont été absentes parce qu’elles étaient 
malades en décembre 2017?
(NB : si un enfant est absent un jour, cela est considéré 
comme absent / Laisser vide (sauter) si le répondant ne sait 
pas)
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
total de filles 
absents.
210
Combien de garçons ont été absents en décembre 2017 à 
cause des travaux champêtres ou domestiques ?
(NB : si un enfant est absent un jour, cela est considéré 
comme absent / Laisser vide (sauter) si le répondant ne sait 
pas)
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
total de garçons 
absents.
211
Combien de filles ont été absentes en décembre 2017 à 
cause des travaux champêtres ou domestiques
(NB : si un enfant est absent un jour, cela est considéré 
comme absent / Laisser vide (sauter) si le répondant ne sait 
pas)
NB : L’enseignant devrait se référer à son registre de 
présence
/_____/
Doit être inférieur 
ou égal au nombre 
total de filles 
absents.
212 Combien de garçons n’ont pas mangé durant la journée 
scolaire ou le jour de classe le plus récent?
/_____/ Doit être inférieur 
ou égal au nombre 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
58
N° Questions Réponses Saut 
(NB : la journée d’école est de 8h à 17; Laisser vide (sauter) 
si le répondant ne sait pas; Déjeuner et collations)
réel de garçons 
dans la classe.
213
Combien de filles n’ont pas mangé durant la journée 
scolaire ou le jour de classe le plus récent?
(NB : la journée d’école est de 8h à 17 ; Laisser vide (sauter) 
si le répondant ne sait pas; Déjeuner et collations)
/_____/ Doit être inférieur 
ou égal au nombre 
réel de filles dans la 
classe.
214
La salle de classe est-elle en adéquation avec le nombre 
d’élèves?
1. Oui 
2. Plus ou moins
3. Non
215
Dans quel état se trouve votre salle de classe ? 1. Bonnes conditions
2. Conditions moyennes
3. Mauvaises conditions
4. Ne sait pas / Pas de 
réponse
216
A votre avis, quelles sont les trois plus importantes priorités 
pour améliorer votre salle de classe ?
a. Plus de bureaux
b. Réparer des bureaux
c. Plus de chaises
d. Réparer des chaises
e. Plus de matériel 
didactique nécessaire 
pour enseigner
f. Plus de manuels de 
français
g. Plus de cahiers 
d’activité en français
h. Toiture
i. Peinture
j. Terrasse/cimentage
k. Portes/fenêtres
l. Autre (à spécifier)
217
A votre avis, quelles sont les trois plus importantes priorités 
pour améliorer votre établissement scolaire ?
a. Améliorer les salles de 
classe existentes
b. Améliorer les latrines 
existentes
c. Améliorer les lavabos 
existentes
d. Construire de nouvelles 
salles de classes
e. Construire de nouvelles 
latrines
f. Construire de nouveaux 
lavabos
g. Construire une clôture
h. Renforcer l’effectif du 
personnel enseignant
i. Doter (ou renforcer) la 
cantine scolaire
j. Autre (à spécifier)
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
59
N° Questions Réponses Saut 
5. Renforcements des capacités de l’enseignant 
301
Avez-vous bénéficié d’une formation spécifique concernant
l’enseignement de la lecture ? (depuis septembre 2014)
(NB : Se concentrer uniquement sur la formation depuis 
septembre 2014)
1. Oui 
2. Non
3. Ne sait pas / Pas de 
réponse
Si 2 ou 3
Q313
302 Si oui, combien de formations avez-vous reçues ? /_____/
303 Quand a eu lieu la formation la plus récente? /___/___/ mois/année
304
Qui a organisé cette formation? (Réponse multiple) a. Gouvernement
b. CRS/FFE/World 
Education
c. Autre (à spécifier)
305
Quelle est votre appréciation de cette formation ? 1. Excellent
2. Bien
3. Passable
4. Peu appréciée
5. Médiocre
306
Quelle est votre appréciation de cette formation ? S’il vous plaît fournir vos commentaires.
……………………………………………………………………………………………………………………………………………………………………………
307
A-t-elle été bien organisée? 1. Oui 
2. Plus ou moins
3. Non
308
Le contenu de la formation vous a-t-il semblé utile? 4. Oui 
1. Plus ou moins
2. Non
309
Cette formation cadrait-elle avec vos besoins 
professionnels en ce qui concerne la lecture?
1. Oui 
2. Plus ou moins
3. Non
310
Quels éléments dans cette formation vous ont semblé particulièrement utiles pour l’enseignement de la lecture?
…………………………………………………………………………………………………………………………………………………………………. 
……………………………………………………………………………………………………………………………………………………………………………….
311
Utilisez-vous ces éléments qui vous ont semblé utiles dans 
votre enseignement?
1. Oui 
2. Plus ou moins
3. Non
Si 2 ou 3
Q312
312
Si non, quelles difficultés limitent la mise en œuvre des acquis de cette formation ? pourquoi ?
…………………………………………………………………………………………………………………………………………………………………………………
….
313
Est-ce que vous utilisez les nouvelles techniques ou les 
nouveaux outils enseignés dans la formation fournie par le 
programme FFE (CRS, WEI)?
1. Oui 
2. Plus ou moins
3. Non
Si 3
Q315
Seules les écoles de 
traitement
314 S’il vous plaît fournir un exemple (des nouvelles techniques 
ou les nouveaux outils).
/ ____ / Seules les écoles de 
traitement
315 Utilisez-vous le programme national d’enseignement et les 
matériels didactiques connexes?
1. Oui 
2. Plus ou moins
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
60
N° Questions Réponses Saut 
3. Non
316
Pour combien de minutes avez-vous enseigné la 
lecture et l’écriture hier ou au cours de lecture et 
d’écriture la dernière fois? 
/______/ minutes
317
Combien de séances de lecture et d’écriture avez-vous 
par semaine?
1. 1
2. 2 
3. 3
4. 4
5. 5
6. 6
7. 7
8. 8
9. 9
10. 10
11. Autre
318
Pensez-vous que le temps officiellement alloué à 
l’apprentissage de la lecture/écriture est suffisant ?
1. Oui, largement
2. Oui, mais cela en 
prendrait un peu plus
3. Non, il manque 
beaucoup de temps
4. Ne sait pas / Pas de 
réponse
319 Pourquoi pensez-vous cela?
………………………………………………………………………………………………………………………………………………………………………
4. Cantine 
401
Votre école a-t-elle une cantine fonctionnelle? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
Si 1
Q403
Si 2
Q402
402
Si non, pourquoi ?
NB : finir de noter les raisons éventuelles avant de réaliser le saut Q501
……………….………………………………………………………………………………………………………………………..
403
Si oui, votre cantine a-t-elle un comité de gestion ? 1. Oui
1. Non
2. Ne sait pas / Pas de 
réponse
404
La cantine a-t-elle un impact sur la concentration des 
élèves? (impact=changement plus ou moins profond)
1. Oui, beaucoup
2. Oui, un peu
3. Non, pas du tout 
4. Ne sait pas / Pas de
réponse
405
Pouvez-vous expliquer comment la cantine a eu un impact sur la concentration des élèves? (impact=changement 
plus ou moins profond)
NB : l’impact peut être positif ou négatif
---------------------------------------------------------------------------------------------
Impact 1 : ……………….….…
Impact 2 : ……………….….
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
61
N° Questions Réponses Saut 
Impact 3 : ……………….…
406
La cantine a-t-elle un impact sur l’apprentissage? 
(impact=changement plus ou moins profond)
5. Oui, beaucoup
1. Oui, un peu
2. Non, pas du tout 
3. Ne sait pas / Pas de 
réponse
407
Pouvez-vous expliquer comment la cantine a eu un impact sur l’apprentissage? (impact=changement plus ou 
moins profond)
NB : l’impact peut être positif ou négatif
Impact 1 : ……………….….…
Impact 2 : ……………….….
Impact 3 : ……………….…
5. Matériel 
501
Dans quelle mesure disposez-vous du matériel didactique 
nécessaire ?
1. Le matériel est suffisant
2. Certains manquent
3. Pas de matériel
4. Ne sait pas / Pas de 
réponse
Si 1 3 ou 4
Q503
502
Quels sont les matériels qui manquent ?
NB : citez juste les plus importants (une à trois réponses)
Matériel 1 : ……………….….
Matériel 2 : …………….….……
Matériel 3 : …………….….…………
503 Dans votre classe, combien d’élèves n’ont pas le manuel de 
français?
/_____/
504 Dans votre classe, combien d’élèves n’ont pas le cahier 
d’activité en français?
/______/
505 Dans votre classe, combien d’élèves n’ont pas les 
fournitures nécessaires?
/______/
506
L’APE et/ou l’AME sont-elles actives en ce qui concerne 
l’implication des parents dans les activités éducatives
1. Oui
2. Non, pas vraiment
3. Non, pas du tout
Si 1 Q507
Si 2 ou3 Q508
507
Si oui, Quel genre d’activités éducatives favorisent-elles la participation des parents?
……………….….…………………………………………………………
508
Si non, Pourquoi pensez-vous qu’ils ne le sont pas?
……………….….…………………………………………………………
509
Vos élèves se lavent-ils les mains avec du savon (ou 
cendres) avant le repas?
4. Oui
5. Non
6. Ne sait pas / Pas de 
réponse`
Si 1 ou 3
Q511
510
Si non, pourquoi ? (Une réponse, raison principale)
………………………………………………………………………………….
511 Vos élèves se lavent-ils les mains avec du savon (ou 
cendres) avant le repas après être allés aux toilette?
1. Oui
2. Non
Si 1 ou 3
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
62
N° Questions Réponses Saut 
3. Ne sait pas / Pas de 
réponse
Q513
512
Si non, pourquoi ? (Une réponse, raison principale)
………………………………………………………………………………….
513
Vos élèves ont-ils accès à des latrines fonctionnelles ? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
514
Les filles ont-elles leur propre latrine? 1. Oui
2. Non
3. Ne sait pas / Pas de 
réponse
6. Impression générale
601
Dans l’ensemble, comment pensez-vous que le programme 
FFE a amélioré la qualité de l’apprentissage dans votre 
école? En utilisant une échelle de 5 (Énormément) à 1 (Très 
peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. 1 - Très peu
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
602
Dans l’ensemble, comment pensez-vous que le programme 
FFE a amélioré la qualité de la santé et de l’hygiène dans 
votre école? En utilisant une échelle de 5 (Énormément) à 1 
(Très peu).
1. 5 - Énormément 
2. 4 - Significativement
3. 3 - Assez
4. 2 - Un peu
5. 1 - Très peu
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
603
Dans quelle mesure la cantine de votre école sera-t-elle 
durable une fois le programme FFE terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
604 S’il vous plaît fournir votre raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
605
Dans quelle mesure les améliorations de votre école en 
matière d’amélioration de la qualité de l’apprentissage 
seront-elles durables une fois que le programme FFE 
terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
606 S’il vous plaît fournir votre raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 7 – TEACHER SURVEY (CONTINUED) 
63
N° Questions Réponses Saut 
607
Dans quelle mesure les améliorations apportées à votre 
école en matière de santé et d’hygiène seront-elles 
durables une fois le programme FFE terminé?
1. Très durable
2. Durable
3. Assez durable
4. Peu durable
5. Pas durable du tout
6. Ne sait pas / Pas de 
réponse
Seules les écoles de 
traitement
608 S’il vous plaît fournir votre raison (s) pour cela. (une réponse, raison principale)
……………….….…………………………………………………………
Seules les écoles de 
traitement
609 Faites-nous part de vos commentaires, si vous en avez éventuellement?
…………………………………………………………………………………………………………………………………………………………………………………
610 En tant que l’intervieweur, veuillez fournir des commentaires supplémentaires (par exemple, problèmes, 
clarifications, commentaires).
…………………………………………………………………………………………………………………………………………………………………………………
Merci beaucoup d’avoir pris le temps de remplir ce sondage. 
Fin du questionnaire 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
64
APPENDIX 8 – LIST OF PARTICIPANTS IN KEY INFORMANT INTERVIEWS 
The following is a list of participants in Key Informant Interviews (KIIs) outside of the schools.
Name Role Location Contact information
CRS and WEI
Cotonou
Sean Gallagher
CRS National Team - Country 
representative
Cotonou +22997975082
Marie Rongeard
CRS National Team- Head of 
Programs
Cotonou
+22997975080
Achille Kounou CRS National Team - Head of 
operation
Cotonou +22997973050
Ousman Maiga CRS Project Team-Project Director Parakou/Cotonou Ousmane.maiga@crs.org
Emmanuel 
AHOUANSINKPO 
CRS Project Team-Project MEAL 
officer
Parakou/Cotonou +229 61884103 
emmanuel.ahouansinkpo@crs.org
Augustin Govoechan CRS Project Team Project Liaison 
Officer
Parakou/Cotonou +229 61 96 65 40
augustin.govoechan@crs.org
Kevin Ray Director - WEI
Parakou +22962156308
Josue Ogouchina
Ganya Adebiyi
Programme Representative - WEI
Programme Representative - WEI
Kandi +22997164888
+22997414745
USDA
Jane Luxner Analyst of the FFE and responsible 
for its oversight
Washington D.C, USA +1888-844-9904
Traci Johnson MEAL officer Washington D.C, USA +1888-844-9904
Other stakeholders
Oussou Elisee Representative of the Directorate 
of School Feeding – Education Min
Cotonou/Porto Novo, 
Benin
+22997486801 
oussouelisee@yahoo.fr
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 8 – LIST OF PARTICIPANTS IN KEY INFORMANT 
INTERVIEWS (CONTINUED) 
65
Name Role Location Contact information
Abiba Orou Topko Child Development UNICEF Parakou +22997975109
Togou Bio Zime Gounou DDEMP
Head of School District – Alibori
Kandi – Alibori, Benin +22994641878
Kati Didier, 
Odjo Annicet, 
Agossou Jean
CRP Kandi; 
CRP Mallenville; 
CRP Gogonou.
Kandi – Alibori, Benin
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
66
APPENDIX 9 – GUIDE FOR INTERVIEWS WITH SCHOOL DIRECTORS 
Name of School: ________________________
Village/Town/City: _____________________
Time started: ________________________
Time ended: ____________________________
Facilitator: Prior to beginning the interview, reconfirm the consent form the respondent. Read out the 
contents of the consent form and ask him/her whether or not they agree to be interviewed. 
 
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here on behalf of Advisem, a consultancy firm contracted by the CRS to 
conduct the end-line evaluation for the Food For Education project. The purpose of the end-line study is to 
understand the changes that have resulted from the project. This interview with you has been requested 
because your point of view and is important. I know you are very busy, and appreciate your time. 
This interview will take about 1 hour.
 
Relevance
1. What are the main challenges preventing children from attending and learning in school?
2. What are the main health and hygiene challenges children have? How do these affect school 
attendance?
3. How does your school address these challenges to help children participate and learn in school?
4. Have you received any training since the beginning of the project on professional capacities? Who 
provided it? What did it cover? Was it useful? How?
5. Have there been any activities in your school recently to improve the quality of the school’s 
infrastructure? If yes, what were they and were they your priorities?
Effectiveness 
6. What activities have been happening in your school to improve learning and attendance and to 
improve their health and dietary practices? 
- Have they produced changes?
- Could more be done, or could these have been done better/differently to improve learning and 
attendance?
7. Do students have particular problems with paying attention during lessons? 
- If yes, what are the indicators of this? 
- Do you or teachers know the cause?
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 9 – GUIDE FOR INTERVIEWS WITH SCHOOL 
DIRECTORS (CONTINUED) 
67
8. Is there an Association des Parents d’Elèves or Association des Mères d’Elèves (APE or AME) in this 
community and working with the school? Is there a Savings and Lending Group (SILC)? A canteen 
management committee?
− What are their roles in the school, and with children? Are they effective? 
− Are they being supported effectively by the project?
 
Efficiency 
 
9. From your perspective as Director, how efficient had the project been? For example,
− Have planning and the delivery of food and infrastructure inputs happened in a timely way? 
− Has there been sufficient follow up to confirm that activities are implemented well and take 
corrective action?
− Have resources been used in a cost-effective way? 
− Should anything have been done differently?
Sustainability 
10. The FFE supports a wide range of activities: meals, school garden, hygiene & sanitation, take-home 
food, savings & loans, radio mobilizing. Are all of these equally important in your view? 
- Should some be dropped to provide better focusing or concentration of resources? 
11. What elements of the FFE does the School District expect to continue past the end of the project? 
How will these be sustained?
12. Has the government shown commitment to taking the policy action necessary to provide the 
“necessary conditions” for children’s learning in the areas being supported by the FFE? 
− If yes, in what ways?
− If no, what needs to happen to ensure improved teacher quality and children’s nutritional status 
are sustained?
Impact and Lessons 
 
13. What have been the key factors enabling the FFE to achieve results? Have there been any limiting 
factors? 
- Are there other strategies which would have worked better?
- Have there been any negative impacts? Could they have been avoided?
14. Are there any positive or negative unexpected results of the project? How did these occur?
15. Are there any other issues about the FFE implementation and/or results that you would like to 
mention?
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
68
APPENDIX 10 – GUIDE FOR INTERVIEWS WITH HEADS OF SCHOOL DISTRICTS 
Name of School: ________________________
Village/Town/City: _____________________
Time started: ________________________
Time ended: __________________________
Male/Female: ___________
Length of time in the position: ____________
Training in Education (yes/no; degree):_____________
Facilitator: Prior to beginning the interview, reconfirm the consent from the respondent. Read out the 
contents of the consent form and ask him/her whether or not they agree to be interviewed.
 
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here on behalf of Advisem, a consultancy firm contracted by the CRS to 
conduct the end-line evaluation for the Food for Education project. The purpose of the end-line study is to 
understand the changes that have resulted from the project. This interview with you has been requested 
because your point of view and is important. I know you are very busy, and appreciate your time. 
This interview will take less than 1 hour.
Background 
What has been the role of the School District in the FFE?
- Have there been any challenges, constraints in this role?
Relevance 
1. From the School District perspective, have the expected results of the FFE -- improved Gr1&2 literacy, 
attendance and parent engagement ; and improved health and dietary practices for students-- been of 
high priority to its key stakeholders: school directors, teachers, parents, community leaders? Have other 
priorities been left out?
2. Has the School District been satisfied with how the FFE has progressed in meeting its implementation 
objectives and realizing its results? If yes, please explain how. If no, please explain why not.
Effectiveness 
3. To what extent do the inputs/activities of FFE fit with your assumptions about the relationship between 
nutrition and learning?
- Are the inputs the right kind, and of sufficient intensity and duration, to lead to better 
attendance, and then to better attentiveness and learning? 
4. Were the approaches to improving reading instruction appropriate to realizing the expected 
improvements in learning outcomes?
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 10 – GUIDE FOR INTERVIEWS WITH HEADS OF SCHOOL 
DISTRICTS (CONTINUED) 
69
5. Have you seen a change of Association des Parents d’Elèves or Association des Mères d’Elèves (APE or 
AME) involvement in school management/governance since the project began? And Savings and Lending 
groups and canteen management groups?
6. Has the support provided to these associations been appropriate, sufficient for them to understand 
and take action to improve the literacy, hygiene and nutrition status of students?
Efficiency 
7. From the perspective of School District how efficient has the project been: for example, 
− Have planning and the delivery of food and infrastructure inputs happened in a timely way? 
− Has there been sufficient follow up to confirm that activities are implemented well and take 
corrective action?
− Have resources been used in a cost-effective way? 
− Should anything have been done differently?
Sustainability 
8. The FFE supports a wide range of activities: meals, school garden, hygiene & sanitation, take-home 
food, savings & loans, radio mobilizing. Are all of these equally important in your view? 
- Should some be dropped to provide better focussing or concentration of resources? 
9. What elements of the FFE does the School District expect to continue past the end of the project? How 
will these be sustained?
10. Has the government shown commitment to taking the policy action necessary to provide the 
“necessary conditions” for children’s learning in the areas being supported by the FFE? 
− If yes, in what ways?
− If no, what needs to happen to ensure improved teacher quality and children’s nutritional status
are sustained?
Impact and Lessons 
11. What have been the key factors enabling the FFE to achieve results? Have there been any limiting 
factors? 
- Are there other strategies which would have worked better?
- Have there been any negative impacts? Could they have been avoided?
12. Are there any positive or negative unexpected results of the project? How did these occur?
13. Are there any other issues about the FFE implementation and/or results that you would like to 
mention?
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
70
APPENDIX 11 – GUIDE FOR INTERVIEWS WITH SENIOR POLICY-MAKERS, 
PROGRAM MANAGERS AND IMPLEMENTERS OF FFE 
(USDA, WEI, Ministry of Preschool and Primary Education/MEMP, Direction of School Feeding, 
Department of Health, Direction of Pedagogical Inspection, National Institute for Training and Research in 
Education…) 
Name of Organization: ________________________
Role/Position in Organization:
Length of time in position: 
Male/Female: 
Village/Town/City: ______________________
Time started: _________________________
Time ended: ____________________________
Facilitator: Prior to beginning the interview, reconfirm the consent from the respondent. Read out the 
contents of the consent form and ask him/her whether or not they agree to be interviewed. 
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here on behalf of Advisem, a consultancy firm contracted by the CRS to 
conduct the end-line evaluation for the Food for Education project. The purpose of the end-line study is to 
understand the changes that have resulted from the project. This interview with you has been requested 
because your point of view and is important. I know you are very busy, and appreciate your time. 
This interview will take about 1 hour.
Background 
1. What has been the role of ______________ in the FFE?
- Have there been any challenges, constraints in this role?
Relevance 
2. From your perspective, have the expected results of the FFE -- improved Gr1&2 literacy, attendance 
and parent engagement; and improved health and dietary practices of students -- been of high priority to 
its key stakeholders: school directors, teachers, parents, community leaders? 
3. Have you been satisfied with how the FFE has progressed in meeting its implementation objectives and 
realizing its results? 
- Have the indicators it has used been accurate and sufficient as reflections of the changes 
expected?
- If yes, please explain how. If no, please explain why not. 
- What would have been better indicators of changes resulting from the project?? 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 11 – GUIDE FOR INTERVIEWS WITH SENIOR POLICY-MAKERS, 
PROGRAM MANAGERS AND IMPLEMENTERS OF FFE (CONTINUED) 
71
Effectiveness 
4. To what extent do the inputs/activities of FFE fit with your assumptions about the relationship between 
nutrition and learning? Do you agree with the “Theory of Change” expressed by the FFE? 
5. From your perspective, have the annual benchmarks been adequately set and met? 
- Has FFE shown enough flexibility in adapting the project to take into account M&E data, or 
changing circumstances?
6. Were the approaches to improving reading instruction appropriate to realizing the expected 
improvements in learning outcomes?
- What training was available for teachers as a part of this program? Has there been any challenges 
with this?
7. Has the support provided to Association des Parents d’Elèves or Association des Mères d’Elèves (APE or 
AME), the Savings and Lending Groups (SILCs), and the Canteen Management Committees, been 
appropriate, sufficient for them to understand and take action to improve the literacy, hygiene and 
nutrition status of students?
- The FFE indicator for parents’ increased engagement with their children’s learning is the ability to 
name three benefits of education. Is this an appropriate indicator?
Efficiency 
8. From your perspective, how effective has the coordination been between CRS and WEI in the 
management of the FFE?
- Have the responsibility areas been appropriately assigned?
- Have they been well enough aligned to avoid duplication, gaps?
- Have there been any particular challenges, and if so have they been satisfactorily resolved? 
9. The Midterm evaluation (MTE) made a number of recommendations as to how to adapt the FFE to 
strengthen results, especially for SO1 on improved literacy. 
- What was your reaction to the MTE and the CRS management response - its plans to follow on the 
recommendations?
- Has subsequent action been sufficient from your perspective?
- Are the wider policy and development environments “enabling” enough for making a significant 
change in literacy learning and in continuous health and nutrition levels?
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 11 – GUIDE FOR INTERVIEWS WITH SENIOR POLICY-MAKERS, 
PROGRAM MANAGERS AND IMPLEMENTERS OF FFE (CONTINUED) 
72
Sustainability 
10. The FFE supports a wide range of activities: meals, school garden, hygiene & sanitation, take-home 
food, savings & loans, radio mobilizing. Are all of these equally important in your view? 
- Should some be dropped to provide better focussing of resources? 
11. What elements of the FFE do you think can realistically be sustained past termination? In your 
experience, what are the key factors necessary for this, and are they in place?
- Has the government shown commitment to taking the policy action necessary to provide the 
“necessary conditions” for children’s learning i.e. improved teacher quality, children’s nutritional 
status?
- Is it your sense that the key stakeholders (school staff, parents, government) have taken sufficient 
“ownership” to ensure continuous attention, action?
Impact and Lessons 
12. What have been the key factors enabling progress and results of the FFE? Have there been any 
limiting factors? 
- Are there other strategies which would have worked better?
- Have there been any negative impacts? Could they have been avoided?
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
73
APPENDIX 12 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH TEACHERS 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of male, female in group: ____Male; ____Female
(There should be a maximum of 10 participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas. Also, 
please review all the notes with the note taker at the end of the FGD to ensure that the main points made 
by FGD participants during the discussion have been captured. Prior to beginning the FGD, you must obtain 
verbal consent from each of the FGD participants. Read out the contents of the consent form and ask each 
individual participant whether or not they agree to participate in the FGD.
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here on behalf of Advisem, a consultancy firm contracted by the CRS to 
conduct the end-line evaluation for the Food For Education project which you have participated in. The 
purpose of the end-line study is to understand the changes that have resulted from the project. The 
colleague with me is Hussein Faruque Aly, a consultant with Advisem.
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. We are here today to talk about your perceptions on how this project has affected 
quality of life for you, your families, and your communities.
Ground rules: 
Please feel free to talk openly. There are no right or wrong answers. Please respect the opinions of others 
– we don’t have to agree; we are interested in hearing different viewpoints. You do not have to speak in 
any particular order. When you do have something to say, please do so but please speak one at a time. 
There are many of you in the group and it is important that I get the views from each of you, so please give 
everyone an equal chance to participate in the discussion.
This focus group will take less than 1.5 hours.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 12 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH 
TEACHERS (CONTINUED) 
74
Questions FGD Teachers
Relevance
1. What are the main challenges preventing children from attending and learning in school?
2. How would you address these challenges?
3. What are the main health and hygiene challenges your children have? How do these affect school 
attendance?
4. What did this project do?
5. What training have you received for teaching reading? What did it cover? Was it useful to your 
teaching?
6. Has the support from the project (e.g. training, materials.) changed how you are teaching reading? 
Are these changes effecting learning outcomes of children in any way?
 
Effectiveness
7. How has the project affected children’s learning and attendance, and childrens’ health and dietary 
practices? In what ways, through what actions? Could these have been more effective?
8. Have you noticed changes in the children’s ability to pay attention in class since the project began?
If yes, can you explain how/why? If no, what do you think can be done?
 
Efficiency 
9. What could be done better to improve the education provided by the school?
10. Were resources of the project provided in a timely way? Was there follow-up to confirm activities 
were well implemented?
 
Impact
11. Are you satisfied with the ways you have been able to participate in the project? Why or why not?
 
Sustainability 
12. If the project was to end tomorrow, what activities would be important to continue? Do you think 
this will happen? What conditions would be necessary to allow them to continue? 
13. Has the project brought the parents and communities into closer collaboration with the schools? If 
yes, in what ways? If no, why not? Should it have tried to do this? 
 
PTAs and Savings/Lending Groups 
14. Is there an Association des Parents d’Elèves or Association des Mères d’Elèleves (APE or AME) in this 
community? If yes, how are they working with the school? Are they effective? 
15. How do you as teachers interact with either of these organisations?
16. Is there a saving and lending group in the community? If yes, has it helped provide resources for 
children’s education or welfare?
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
75
APPENDIX 13 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH STUDENTS 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of boys and girls in the group: ____Boys ____Girls
(There should be a maximum of 10 participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas. Also, 
please review all the notes with the note taker at the end of the FGD to ensure that the main points made 
by FGD participants during the discussion have been captured. Prior to beginning the FGD, you must obtain 
verbal consent from each of the student FGD participants, and their parents, who will have been selected for 
a different FGD for parents. Please ensure that during the exercise a teacher, a parent, or the school director 
is present, to ensure the protection of the children. Read out the contents of the consent form and ask each 
individual participant whether or not they agree to participate in the FGD.
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here to talk to you and learn about the Food For Education project which 
has been happening in your school; to understand the changes that have happened. The colleague with me 
is Hussein Faruque Aly, a consultant from Canada.
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. We want to understand how this project has affected quality of life for you, your 
families, your school and your community.
Ground rules: 
Please feel free to talk openly. There are no right or wrong answers. Please respect the opinions of others 
– we don’t have to agree; we are interested in hearing different viewpoints. You do not have to speak in 
any particular order. When you do have something to say, please do so but please speak one at a time. 
There are many of you in the group and it is important that I get the views from each of you, so please give 
everyone an equal chance to participate in the discussion.
This will be a participatory and fun exercise. We look forward to doing this with you. It won’t take very long.
Facilitator: create three spaces in the room, each representing “always true, sometimes true, never true” 
in the room. As an ice breaker and to get the students comfortable, you can do the following exercise: 
1) Ask students to name their favourite fruits/foods. 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 13 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH 
STUDENTS (CONTINUED) 
76
2) Choose 3.Allocate a fruit/food to each student. When you call the fruit associated with their name, they 
should come to the middle of the room and then return to their places
3) Test once then with “I love eating this fruit/food””, and ask students to go to always true, sometimes 
true, never true. 
4) Once they have understood how this works, start the exercise. When the mood meter question is asked, 
students are requested to go to the “always true, sometimes true, and never true” areas of the room. The 
notetaker should take note of the number of students in each area. 
A: Mood Meter
Statement Always true Sometimes 
true
Never true
I am happy in school; I like to come every day
I come late to class 
My teacher comes every day 
My teacher comes late to class
It is easy for me speak out in class, ask questions
It is easy for me to understand my lessons
I listen when the teacher is talking
I answer the teacher’s questions in the class.
I ask the teacher questions about the lesson 
I finish my reading assignments
The teacher reminds me to do my work
I fall asleep in the classroom
I am punished for my behaviour in class
I am hungry during the lesson
I am sick and miss class 
I feel like running and playing at recess
I like the meals in this school
I eat in the school kitchen
The teacher helps me when I have a problem in my lesson.
The teacher gives us reading tests 
I wash my hands with soap and water before eating
I wash my hands after using the bathroom
The school is organized well; I feel safe here.
The toilets are dirty; I don’t like to use them
The is a special toilet for girls
I enjoy writing stories 
I enjoy reading stories
I use lessons I learn in school when I am home
I am learning letters and words
I have difficulty with reading 
My parents are happy that I am learning at school
My parents give books I can read at home
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 13 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH 
STUDENTS (CONTINUED) 
77
Statement Always true Sometimes 
true
Never true
My parents help me with my reading homework
My parents come to the school to help the teacher
Now we would like to ask you some questions about your school and your activities in school. Each person will have 
a chance to answer so please take your time. There are no right answers. If anyone does not want to answer, that is 
also alright.
B: FGD Questions 
1. When you miss school on some days, what is the reason? 
2. Tell me about your school: what do you like best? What don’t you like?
3. Do you ever feel bored or confused in the lesson? What do you do then?
4. Do you have enough to eat during the day? Do you eat every day in the school kitchen? 
5. Tell me about what you are reading in your lessons? Do you read books at home? 
6. Tell me about the toilets and the hand-washing points in the school: Are they easy to use? Are 
there any problems with them?
7. Is there anything else you would like to say about your school and what you are learning?
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
78
APPENDIX 14 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH BEST 
PERFORMING STUDENTS 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of boys, girls in group: ____boys; ____girls
Number of Parents: ________Fathers; ____________Mothers
(There should be a maximum of 10 – ideally up to 4 with equal number of boys and girls and their parents￾participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas. When 
asking a question, please ensure that the students are able to speak first, expressing their thoughts fully, 
before engaging the parents. Also, please review all the notes with the note taker at the end of the FGD to 
ensure that the main points made by FGD participants during the discussion have been captured. Prior to 
beginning the FGD, you must obtain verbal consent from each of the FGD participants. Read out the contents 
of the consent form and ask each individual participant whether or not they agree to participate in the FGD. 
This exercise will take begin with the students and parents being interviewed in different groups to allow 
for unhindered participation from the students. 
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here on behalf of Advisem, a consultancy firm contracted by the CRS to 
conduct the end-line evaluation for the Food For Education project which you have participated in. The 
purpose of the end-line study is to understand the changes that have resulted from the project. The 
colleague with me is Hussein Faruque Aly, a consultant with Advisem.
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. We are here today to talk about your perceptions on how this project has affected 
quality of life for you, your families, and your communities.
Ground rules: 
Please feel free to talk openly. There are no right or wrong answers. Please respect the opinions of others 
– we don’t have to agree; we are interested in hearing different viewpoints. You do not have to speak in 
any particular order. When you do have something to say, please do so but please speak one at a time. 
There are many of you in the group and it is important that I get the views from each of you, so please give 
everyone an equal chance to participate in the discussion.
This focus group will take about 1.5 hours.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 14 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH BEST 
PERFORMING STUDENTS (CONTINUED) 
79
General Background 
1. What are the most important reasons for you to attend school?
2. What benefits you are getting from school now?
3. What do you enjoy the most about school? 
4. What do you dislike the most?
Family Education Background 
5. What level of education do your parents have?
6. What language do you speak to your family at home?
7. Do have brothers or sisters? Do they go to school?
8. Do you take any lessons outside school? Religious school, art school, sport school?
Home conditions (space to study, use of light, desks for their homework, etc) 
9. Do you have classwork you do after school? For how much time a day? In one sitting or in various 
sittings?
10. Do you have electricity at home?
11. Does anyone at home help you with your school work?
12. Do you have chores at home; do you help your parents with their work?
13. Do you have access to TV, Radio, Newspapers, books, internet? If so, do you use this?
Daily Routine 
14. Please describe your daily routine? Please do not forget to mention the following: 
(Time of waking up; Waking up routine; Any meal taken before school; Walking/time distance to the 
school; Play time at home? How long for? Study time at home, how much time a day? In one sitting or in 
various sittings? Any work assisting parents at home, fields, parents work? Do you see TV or listen to 
radio? What language? How long for during the day)
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
80
APPENDIX 15 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH PARENTS 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of mothers/fathers in group: Mothers____Fathers____
(There should be a maximum of 10 participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas. Also, 
please review all the notes with the note taker at the end of the FGD to ensure that the main points made 
by FGD participants during the discussion have been captured. Prior to beginning the FGD, you must obtain 
verbal consent from each of the FGD participants. Read out the contents of the consent form and ask each 
individual participant whether or not they agree to participate in the FGD.
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here to conduct the end-line evaluation for the Food For Education 
project which you have participated in. The purpose of the end-line study is to understand the changes that 
have resulted from the project. The colleague with me is Hussein Faruque Aly, a consultant from Canada.
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. We are here today to talk about your perceptions on how this project has affected 
quality of life for you, your families, and your communities.
Ground rules: 
Please feel free to talk openly. There are no right or wrong answers. Please respect the opinions of others 
– we don’t have to agree; we are interested in hearing different viewpoints. You do not have to speak in 
any particular order. When you do have something to say, please do so but please speak one at a time. 
There are many of you in the group and it is important that I get the views from each of you, so please give 
everyone an equal chance to participate in the discussion.
This focus group will take about 1.5 hours.
 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 15 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH 
PARENTS (CONTINUED) 
81
Relevance 
1. What are the main challenges preventing children from attending and learning in school?
2. How would you address these challenges?
3. What are the main health and hygiene challenges your children have? How do these affect school 
attendance?
4. What did this project do?
5. Has this helped address the challenges preventing children from attending and learning in school, 
or the health challenges children usually face?
 
Effectiveness
6. What are the most important changes to your children’s learning and attendance since the 
project started? 
7. Have you noticed changes in your children’s ability to pay attention in class since the project 
began? If yes, can you explain how/why? If no, what do you think can be done?
 
Efficiency 
8. What could be done better to improve the education provided by the school?
9. Were resources of the project provided in a timely way? Was there follow-up to confirm activities 
were well implemented?
 
Impact
10. What was the projects impact on children’s learning and attendance? What was the project 
impact on children’s health and dietary practices? 
11. Are you satisfied with the ways you have been able to participate in the project? Why or why 
not?
 
Sustainability 
12. If the project was to end tomorrow, what activities would be important to continue? Do you think 
this will happen? What conditions would be necessary to allow them to continue? 
13. Has the project brought the parents and communities into closer collaboration with the schools? 
If yes, in what ways? If no, why not? Should it have tried to do this? 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
82
APPENDIX 16 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH PARENTS OF 
CHILDREN NOT ATTENDING SCHOOL 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of mothers/fathers in group: Mothers____Fathers____
(There should be a maximum of 10 participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas – with 
particular attention to mothers. Also, please review all the notes with the note taker at the end of the FGD 
to ensure that the main points made by FGD participants during the discussion have been captured. Prior to 
beginning the FGD, you must obtain verbal consent from each of the FGD participants. Read out the contents 
of the consent form and ask each individual participant whether or not they agree to participate in the FGD.
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here to evaluate the project which is happening in your community to 
provide meals to school children. We understand that your children do not attend the school. However, we 
are still interested in learning any ideas you might have about what the project is doing; and more generally 
your opinions about the role of the school in the community. The colleague with me is Hussein Faruque 
Aly, a consultant from Canada.
Ground rules: 
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. Please feel free to talk openly. There are no right or wrong answers. Please respect 
the opinions of others – we don’t have to agree; we are interested in hearing different viewpoints. You do 
not have to speak in any particular order. When you do have something to say, please do so but please 
speak one at a time. There are many of you in the group and it is important that I get the views from each 
of you, so please give everyone an equal chance to participate in the discussion.
This focus group will take at most 1.0 hours.
 
 
Background
1. How many people are in your household? adults: children: 
2. What are the main occupations of adults in the household?
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 16 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH PARENTS OF 
CHILDREN NOT ATTENDING SCHOOL (CONTINUED) 
83
3. What are the levels of education of the adults?
4. Do any of the children go to school? Boys : Girls : 
5. How would you describe your childrens’ health condition? 
6. Do children have access to enough food during the day? 
7. For children who are not going to school, what activities do they usually do during the day? 
 
Relevance 
8. What do you see as the most important reasons for children going to school?
9. What are main challenges families face in sending them?
Effectiveness
10. In what ways can government or agencies best help families send children to school? 
11. Have you received any support to send your children to school? If yes, in what ways was it helpful 
or not helpful?
 
Efficiency 
12. What changes could the school make that would make it easier for more children to attend?
 
Impact
13. Have you noticed any changes in the community since the project began: for the children or their 
families? 
 
Sustainability 
14. If yes, do you think these changes have made a permanent difference to the well-being of the 
families, or to the health and education of the children?
 
Savings/Lending Groups 
15. Do you participate in a saving/lending group? If not, why not? If yes, has it been helpful to your 
family – in what ways?
Radio Messages 
16. Have you heard any radio messages on the importance of education lately? Do you think they 
were relevant messages or shows? Have they made you reflect about your opinion of children’s
education?
The radio shows were broadcasted by the partner radio stations on the following languages: 
- Kandi FM : Dendi et Monkolé ; Nonsina FM de Bembèrèkè : Bariba et Peulh ; Su Tii dèra FM de Nikki : 
Bariba et Boo
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
84
APPENDIX 17 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH MEMBERS OF 
PARENT-TEACHER ASSOCIATIONS, SAVINGS AND INTERNAL LENDING 
COMMUNITIES AND CANTEEN MANAGEMENT COMMITTEES 
Name of School: ________________________
Village/Town/City: _____________________
Time focus group started: ________________________
Time focus group ended: ____________________________
Number of Participants: ______________
Record number of mothers/fathers in group: Mothers____Fathers____
(There should be a maximum of 10 participants to keep the FGD manageable) 
 
Facilitator: As a facilitator, you should encourage all those present to share their views and ideas. Also, 
please review all the notes with the note taker at the end of the FGD to ensure that the main points made 
by FGD participants during the discussion have been captured. Prior to beginning the FGD, you must obtain 
verbal consent from each of the FGD participants. Read out the contents of the consent form and ask each 
individual participant whether or not they agree to participate in the FGD.
Note taker: The note taker should note down, without judgement, everything that is mentioned by FGD 
participants during the discussion. Please record this information in the note taking template provided to 
you.
Introduction:
Hello. Thank you for agreeing to take part in this focus group discussion. My name is 
____________________ and I am here to conduct the end-line evaluation for the Food For Education 
project which you have participated in. The purpose of the end-line study is to understand the changes that 
have resulted from the project. The colleague with me is Hussein Faruque Aly, a consultant from Canada.
You have been asked here today since your point of view is important. I know everyone is very busy, and 
we appreciate your time. We are here today to talk about your perceptions on how this project has affected 
quality of life for you, your families, and your communities.
Ground rules: 
Please feel free to talk openly. There are no right or wrong answers. Please respect the opinions of others 
– we don’t have to agree; we are interested in hearing different viewpoints. You do not have to speak in 
any particular order. When you do have something to say, please do so but please speak one at a time. 
There are many of you in the group and it is important that I get the views from each of you, so please give 
everyone an equal chance to participate in the discussion.
This focus group will take about 1.5 hours.
 
 
 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 17 – GUIDE FOR FOCUS GROUP DISCUSSIONS WITH MEMBERS OF 
PARENT-TEACHER ASSOCIATIONS, SAVINGS AND INTERNAL LENDING 
COMMUNITIES AND CANTEEN MANAGEMENT COMMITTEES (CONTINUED) 
85
Relevance
1. What are the main reasons for your children to attend school?
2. What are the main health and hygiene challenges your children have to be able to attend school?
3. How has the project worked with Parents Associations (APEs/AMEs)?
4. How has the project worked with Savings and Lending Groups (SILCs)?
5. How has the project supported the parents working with the Canteens?
Effectiveness 
6. What have the Parents Associations, Savings and Lending groups and others learnt through the 
project? 
7. What was the approach used to engage with the PTAs (APEs/AMEs) and SILCs? Was it effective?
8. Has your understanding of education, health and hygiene practices changed because of the 
project? How?
 
Efficiency
9. If you were in charge of the project, what would you have done differently to engage with the 
PTAs (APEs/AMEs) and with the SILCs?
Sustainability
10. If the project was to end tomorrow, what activities would continue? How would these activities 
continue?
11. What could help activities started by the project to continue or not?
12. How did the project bring communities into closer collaboration with schools?
Impact
13. What was the projects impact on children’s learning and attendance? What was the project 
impact on children’s health and dietary practices? 
14. How have parents’ and the communities’ participation in the school’s activities been impacted by 
the project? 
15. What is the PTAs (APEs/AMEs) role and the SILCs role in the good functioning of the school and 
improving the students’ education?
Lessons learned and best practices 
16. What have been the key lessons learnt from implementing this project? 
17. What would you recommend to improve future versions of the project?
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
86
APPENDIX 18 – HAND-WASHING STATION OBSERVATION TOOL 
QUESTIONNAIRE – STATIONS DE LAVAGE DES MAINS 
OUTIL D’OBSERVATION
Remplissez ceci vers la fin de la journée. Prenez le temps d’observer les élèves pendant leurs pauses et 
leurs déjeuners pour examiner l’utilisation des stations de lavage des mains. La pause déjeuner (de midi à 
15h00) pourrait être un bon moment pour faire des observations. Vous aurez besoin de visiter toutes les 
stations pour voir si elles ont de l’eau et du savon / cendres.
Pour le projet, les "stations de lavage des mains" sont souvent “tippy taps”.
Cette section doit être remplie par l’intervieweur (le questionnaire en ligne aura un format déroulant):
Date: (JJ-MM-AAAA)
Nom de l’intervieweur: (spécifier)
Département: (spécifier)
Commune: (spécifier)
Numéro d’identification 
unique de l’école:
(spécifier)
Nom de l’école: (spécifier)
École de Contrôle ou de 
Traitement:
(spécifier)
École de Rural ou Urbain: (spécifier)
N° Questions Réponses Saut 
101 L’école a-t-elle des stations de lavage des mains? 3. Oui
4. Non
Si 2 Fin
102
Si oui, combien de stations de lavage des mains 
l’école dispose-t-elle?
/______/
103
Les stations de lavage des mains ont-elles de l’eau? 1. Toutes (100%)
2. Certaines (50% - 100%)
3. Peu (1-49%)
4. Aucune (0%)
104
Les stations de lavage des mains ont-elles du savon 
(ou les cendres de remplacement localement 
acceptées)?
1. Toutes (100%)
2. Certaines (50% - 100%)
3. Peu (1-49%)
4. Aucune (0%)
105
Pensez-vous que les stations de lavage des mains 
sont couramment utilisées ? (p. ex., voir des élèves 
qui utilisent des stations, la présence du savon ou 
des cendres utilisées, de l’eau utilisée dans un seau 
ou sur le sol par des élèves à l’école).
1. Oui, très bien
2. Quelque peu
3. Non pas du tout
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 18 – HAND-WASHING STATION OBSERVATION TOOL (CONTINUED) 
87
N° Questions Réponses Saut 
106
Personnellement, avez-vous vu des élèves qui 
utilisaient des stations de lavage des mains ?
1. Oui, plusieurs
2. Oui, quelques uns
3. Non
107
S’il vous plaît fournir tous les commentaires pertinents.
Par exemple, quand avez-vous vu des personnes qui utilisaient des stations de lavage des mains - après avoir utilisé les 
latrines, avant de manger? Est-ce qu’une des stations de lavage des mains avait besoin d’être réparée?
………………………………………………………………………………………………………………………………………………………………………..
MGD Illustrative: Percent of schools with soap and water at hand washing stations commonly used by students.
To be considered a ‘yes’; 50% or more of hand washing stations must have soap and water (Q: 103 and 104) and the 
enumerator must provide the answer ‘yes, very much’ (Q: 105) to whether there is evidence that students are using 
the stations.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
88
APPENDIX 19 – PHYSICAL ENVIRONMENT OBSERVATION TOOL 
QUESTIONNAIRE – OUTIL D’OBSERVATION DE 
L’ENVIRONNEMENT D’APPRENTISSAGE 
Veuillez remplir cette page vers la fin de la journée une fois que vous avez eu l’occasion d’observer 
l’école. La pause déjeuner (de midi à 15h00) pourrait être un bon moment pour faire des observations. 
Meilleur si vous pouvez voir l’école avec le directeur ou un enseignant.
N° Questions Réponses Saut 
101L’école a-telle une cantine? 1. Oui
2. Non Si 2 105
102
Si oui, la cantine semble être bien équipée? 1. Oui, très bien
2. Quelque peu
3. Non pas du tout
103
La cantine est-elle propre? 1. Oui, très propre
2. Un peu sale
3. Très sale
104L’école dispose-t-elle d’un magasin? 1. Oui
2. Non Si 2 107
105
Si oui, le magasin est-il propre et bien rangé? 1. Oui, très bien rangé
2. Un peu désordonné
3. Très désordonné
106L’école dispose-telle des latrines? 1. Oui
2. Non Si 2 112
107
Si oui, le nombre de latrines semble-t-il suffisant 
pour l’école? 
NB : Il y a suffisamment de toilettes disponibles -
1 par 25 filles ou personnel féminin, et 1 toilette 
plus 1 urinoir (ou 50 centimètres de mur 
d’urinoir) par 50 garçons ou personnel 
masculin.9
Utiliser la population connue 
d’étudiants et d’enseignants 
masculins et féminins; et peut 
compter le nombre de toilettes 
désignées. Peut estimer la 
longueur de mur d’urinoir.
1. Oui
1. Non
108L’école a-t-elle des latrines spécifiquement pour les 
femmes?
2. Oui
3. Non Si 2 111
 
9 Taken from UNICEF’s Water, Sanitation and Hygiene (WASH) in Schools. 
https://www.unicef.org/publications/files/CFS_WASH_E_web.pdf
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 19 – PHYSICAL ENVIRONMENT OBSERVATION TOOL (CONTINUED) 
89
N° Questions Réponses Saut 
109
Si oui, le nombre de latrines réservées aux femmes 
semble-t-il suffisant pour l’école?
NB : Il y a suffisamment de toilettes disponibles -
1 par 25 filles ou personnel féminin, et 1 
toilette.
Utiliser la population connue d’étudiantes et 
d’enseignantes féminins; et peut compter le nombre 
de toilettes désignées.
4. Oui
1. Non
110
Les latrines sont-elles propres? 1. Oui, très propres
2. Un peu sales
3. Très sales
111L’école a-t-elle accès à de l’eau potable? 1. Oui
2. Non
112L’école a-t-elle un jardin? 1. Oui
2. Non Si 2 117
113Le jardin est-t-il utilisé? 1. Oui
2. Non Si 2 116
114
Si oui, Quelles sont les trois principales cultures du jardin?
Culture 1 : ……………………..
Culture 2 : ……………………..
Culture 3 : ……………………..
115Y a-t-il de l’espace pour développer davantage le 
jardin?
1. Oui
2. Non 
116
S’il vous plaît fournir tous les commentaires pertinents.
………………………………………………………………………………………………………………………………………………………………
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
90
APPENDIX 20 – ATTENTIVENESS OBSERVATION TOOL 
QUESTIONNAIRE - OUTIL D’OBSERVATION DE 
L’ATTENTION DE L’ÉLÈVE
Cette section doit être remplie par l’intervieweur (le questionnaire en ligne aura un format déroulant):
Date: (JJ-MM-AAAA)
Nom de l’intervieweur: (spécifier)
Département: (spécifier)
Commune: (spécifier)
Numéro d’identification 
unique de l’école:
(spécifier)
Nom de l’école: (spécifier)
École de Contrôle ou de 
Traitement:
(spécifier)
Rural ou Urbain (spécifier)
Instructions pour l’énumérateur:
• Deux salles de classe (de la 3ème à la 6ème année) seront sélectionnées de façon aléatoire pour 
observation dans chaque école visitée (RNG) ;
• Notez que chaque classe représente un “sondage” distinct ;
• Vous devez observer un cours d’alphabétisation en français ;
• Procédez à la sélection aléatoire de 10 étudiants en utilisant les grilles de sélection aléatoire ;
• Il doit y avoir un rapport égal entre les sexes (5 filles et 5 garçons) ;
• Attentif / Inattentif: Les observations doivent être faites à un moment choisi pendant le cours. 
L’observateur doit observer les étudiants pendant le temps nécessaire pour affirmer s’ils sont 
attentifs ou non (suggestion de 3 minutes pour chaque élève).
• Demandez si l’élève reçoit une ration à emporter (take home ration) et des repas scolaires APRÈS 
que toutes les observations ont été faites ;
 
Attentif: 
• Tous ou la plupart des comportements de l’élève sont liés à la réalisation active de la tâche 
assignée par l’enseignant ;
• L’élève écoute l’enseignant ou d’autres élèves pendant qu’ils participent ;
• L’élève prend des notes qui semblent être liées au contenu de la classe ; ou
• L’élève lève la main pour répondre à la question d’une manière sincère.
 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 20 – ATTENTIVENESS OBSERVATION TOOL (CONTINUED) 
91
Inattentif: 
• L’élève est rappelé à l’ordre par l’enseignant pour lui rappeler l’objectif de la classe ;
• L’élève fait des commentaires qui perturbent la classe ;
• L’élève parle à d’autres élèves pendant une période qu’il/elle devrait écouter ;
• L’élève “bouge” sur son siège, au point qu’il/elle perturbe les autres autour de lui/elle ;
• L’élève est endormi ;
• L’élève semble distrait ou oublie des choses évidentes dans les activités en classe ;
N° Questions Réponses Saut 
108
Niveau scolaire: 5. CE1 (3e année)
6. CE2 (4e année)
7. CM1 (5e année)
8. CM2 (6e année)
9. Autre (spécifier)
109 Sujet enseigné 3. Français
4. Autre (spécifier)
110 Nombre d’élèves / ________/
111 Nombre de filles / ________/
112 Nombre de garçons / ________/
Élèves # fille 
/garçon
Mange des 
repas scolaires 
(oui / non)
Reçoit la ration 
à emporter 
(oui / non)
Évaluation des Atténuation (Attentif / Inattentif)
1
2
3
4
5
6
7
8
9
10
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
92
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS 
Intervention Schools 
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
12 Alibori Gogounou Alafiarou R 74 65 139
28 Alibori Gogounou Bèlètchoutè R 62 38 100
13 Alibori Gogounou Goubéra R 34 32 66
29 Alibori Gogounou Lafiarou R 69 75 144
21 Alibori Gogounou Ouèrè/A R 160 117 277
22 Alibori Gogounou Ouèrè/B R 145 83 228
8 Alibori Gogounou Soukarou/A R 157 107 264
9 Alibori Gogounou Soukarou/B R 136 99 235
18 Alibori Gogounou Sounon-Gah R 88 99 187
10 Alibori Gogounou Wara/A R 230 147 377
11 Alibori Gogounou Wara/B R 197 117 314
7 Alibori Gogounou Worou-Wodou R 63 80 143
30 Alibori Gogounou Zougou-P/A R 213 176 389
1 Alibori Gogounou Bagou/A U 173 185 358
2 Alibori Gogounou Bagou/B U 252 258 510
3 Alibori Gogounou Bagou/C U 93 104 197
14 Alibori Gogounou Gogounou-Centre U 168 202 370
23 Alibori Gogounou Gogounou-Qtier/A U 151 144 295
24 Alibori Gogounou Gogounou-Qtier/B U 163 151 314
25 Alibori Gogounou Gounarou/A U 187 191 378
26 Alibori Gogounou Gounarou/B U 175 133 308
27 Alibori Gogounou Gounarou/C U 117 120 237
15 Alibori Gogounou Gourou-Bansou/A U 254 210 464
16 Alibori Gogounou Gourou-Bansou/B U 213 227 440
17 Alibori Gogounou Gourou-Bansou/C U 154 176 330
4 Alibori Gogounou Sori/A U 147 150 297
5 Alibori Gogounou Sori/B U 148 150 298
6 Alibori Gogounou Sori/C U 151 179 330
19 Alibori Gogounou Sori-Qtier/A U 131 108 239
20 Alibori Gogounou Sori-Qtier/B U 125 153 278
31 Alibori Kandi ABOUKOTO R 64 63 127
32 Alibori Kandi AKOÏTCHAOU R 29 38 67
33 Alibori Kandi ALFAKOARA R 202 226 428
37 Alibori Kandi BEKOUNTAROU R 72 97 169
39 Alibori Kandi BOROGAMAGOU R 41 36 77
42 Alibori Kandi DEOU R 55 43 98
43 Alibori Kandi DODOPANE R 71 81 152
46 Alibori Kandi DONWARIBE R 34 33 67
47 Alibori Kandi FIRI BIONORO R 33 67 100
48 Alibori Kandi GALA R 39 40 79
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
93
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
52 Alibori Kandi 
GBEROUKPANIN￾PEULH R 69 46 115
53 Alibori Kandi GUEDAROU R 64 71 135
54 Alibori Kandi KADJERE R 65 52 117
55 Alibori Kandi KAMA R 44 74 118
61 Alibori Kandi KANDI-FÔ R 123 110 233
65 Alibori Kandi KPESSAKO R 36 26 62
69 Alibori Kandi MONGO R 109 81 190
70 Alibori Kandi PADE R 241 193 434
71 Alibori Kandi PADE-PEULH R 54 62 116
72 Alibori Kandi PEDE-A R 154 103 257
73 Alibori Kandi PEDE-B R 131 146 277
74 Alibori Kandi PEDIGUI R 124 151 275
75 Alibori Kandi PODO-LAFIAROU R 134 123 257
82 Alibori Kandi SAKATOUSSA R 38 62 100
85 Alibori Kandi SINAWONGOUROU/A R 109 124 233
86 Alibori Kandi SINAWONGOUROU/B R 57 55 112
87 Alibori Kandi SONKPARE R 45 45 90
89 Alibori Kandi TANKONGOU R 200 185 385
90 Alibori Kandi TCHOKA R 35 50 85
91 Alibori Kandi TERI R 48 72 120
92 Alibori Kandi THUY R 178 185 363
93 Alibori Kandi THYA / A R 185 179 364
94 Alibori Kandi TISSAROU R 169 126 295
95 Alibori Kandi TISSAROU-PEULH R 55 65 120
96 Alibori Kandi 
WOYOROUKOU￾BANSOU R 33 59 92
97 Alibori Kandi YEGOUROU R 77 71 148
34 Alibori Kandi ANGARADEBOU A U 179 198 377
35 Alibori Kandi ANGARADEBOU B U 69 72 141
36 Alibori Kandi BAKAYERE-FIRI U 204 143 347
38 Alibori Kandi BENSEKOU U 147 164 311
40 Alibori Kandi DAMADI-A U 156 179 335
41 Alibori Kandi DAMADI-B U 111 128 239
44 Alibori Kandi DONWARI-A U 172 144 316
45 Alibori Kandi DONWARI-B U 169 135 304
49 Alibori Kandi GANSOSSO /A U 186 174 360
50 Alibori Kandi GANSOSSO /B U 225 286 511
51 Alibori Kandi GANSOSSO /C U 210 222 432
56 Alibori Kandi KANDI / A U 202 279 481
57 Alibori Kandi KANDI / B U 178 150 328
58 Alibori Kandi KANDI / C U 162 174 336
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
94
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
59 Alibori Kandi KANDI / D U 176 186 362
60 Alibori Kandi KANDI / E U 140 169 309
62 Alibori Kandi KASSAKOU/A U 180 188 368
63 Alibori Kandi KASSAKOU/B U 121 137 258
64 Alibori Kandi KASSANIN U 98 167 265
66 Alibori Kandi MADINA / A U 158 175 333
67 Alibori Kandi MADINA / B U 193 170 363
68 Alibori Kandi MADINA / C U 151 153 304
76 Alibori Kandi SAKA / A U 223 286 509
77 Alibori Kandi SAKA / B U 229 262 491
78 Alibori Kandi SAKA / C U 222 252 474
79 Alibori Kandi SAKA /D U 197 197 394
80 Alibori Kandi SAKA /E U 185 211 396
81 Alibori Kandi SAKA /F U 49 88 137
83 Alibori Kandi SAM A U 193 180 373
84 Alibori Kandi SAM B U 78 88 166
88 Alibori Kandi SONSORO-A U 232 260 492
112 Alibori Malanville BODJECALI/A R 180 274 454
113 Alibori Malanville BODJECALI/B R 117 112 229
114 Alibori Malanville BODJECALI/C R 86 107 193
98 Alibori Malanville FADAMA R 83 99 182
99 Alibori Malanville GAROU/A R 180 275 455
100 Alibori Malanville GAROU/B R 126 241 367
103 Alibori Malanville GOUN-GOUN R 207 195 402
104 Alibori Malanville GOURE-BOIFFO R 66 45 111
107 Alibori Malanville GUENE QUARTIER R 147 193 340
105 Alibori Malanville GUENE/A R 186 183 369
106 Alibori Malanville GUENE/B R 175 132 307
108 Alibori Malanville ILOUA R 77 111 188
109 Alibori Malanville KOALEROU R 44 63 107
117 Alibori Malanville KOCHI R 62 141 203
110 Alibori Malanville MADECALI R 234 523 757
101 Alibori Malanville MONKASSA R 213 248 461
111 Alibori Malanville SENDE-KOARATEDJI R 62 81 143
102 Alibori Malanville WANDA R 39 69 108
129 Alibori Malanville WANZAN-KOARA R 101 112 213
128 Alibori Malanville WOLLO R 118 213 331
115 Alibori Malanville HOURO-YESSO/A U 124 192 316
116 Alibori Malanville HOURO-YESSO/B U 229 368 597
118 Alibori Malanville MVILLE CENTRE/A U 204 240 444
119 Alibori Malanville MVILLE CENTRE/B U 248 301 549
120 Alibori Malanville MVILLE CENTRE/C U 252 315 567
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
95
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
121 Alibori Malanville MVILLE CENTRE/D U 263 318 581
122 Alibori Malanville MVILLE QUARTIER/A U 196 331 527
123 Alibori Malanville MVILLE QUARTIER/B U 193 245 438
124 Alibori Malanville MVILLE QUARTIER/C U 292 392 684
125 Alibori Malanville TASSI-TEDJI/A U 219 302 521
126 Alibori Malanville TASSI-TEDJI/B U 254 290 544
127 Alibori Malanville TASSI-TEDJI/C U 346 443 789
139 Borgou Kalalé BOUCA/A R 281 332 613
140 Borgou Kalalé BOUCA/B R 249 250 499
141 Borgou Kalalé BOUCA-QUARTIER/A R 282 253 535
142 Borgou Kalalé BOUCA-QUARTIER/B R 300 257 557
143 Borgou Kalalé DERASSI R 406 334 740
134 Borgou Kalalé KIDAROUKPEROU/A R 166 176 342
135 Borgou Kalalé KIDAROUKPEROU/B R 106 100 206
136 Borgou Kalalé NASSICONZI/A R 109 115 224
137 Borgou Kalalé NASSICONZI/B R 266 223 489
138 Borgou Kalalé PEONGA R 355 380 735
132 Borgou Kalalé KALALE QUARTIER/A U 180 215 395
133 Borgou Kalalé KALALE QUARTIER/B U 177 212 389
130 Borgou Kalalé KALALE/A U 186 185 371
131 Borgou Kalalé KALALE/B U 160 169 329
Total 21228 23119 44347
Control Schools 
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
1 ALIBORI BANIKOARA EPP GNIGOUWIROU R 42 51 93
2 ALIBORI BANIKOARA EPP ABIDJAN R 39 37 86
3 ALIBORI BANIKOARA EPP KINGAROU R 52 47 99
4 ALIBORI BANIKOARA EPP TANTANI R 28 52 80
5 ALIBORI BANIKOARA EPP TISSOUA R 40 37 77
6 ALIBORI BANIKOARA EPP TOUMAROU R 62 76 138
7 ALIBORI BANIKOARA EPP BANIGOUROU R 48 37 85
8 ALIBORI BANIKOARA
EPP SAHADOUROU￾BOURO R 25 30 55
9 ALIBORI BANIKOARA
EPP GOUNON NON 
SOUAROU R 26 33 59
10 ALIBORI BANIKOARA
EPP KPAKO 
WONGNAGOU R 32 15 47
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
96
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
11 ALIBORI BANIKOARA EPP KOKIBOROU-PEULH R 27 31 58
15 ALIBORI BANIKOARA EPP GABE-GBEROU R 49 56 105
17 ALIBORI BANIKOARA EPP TIGANSON R 56 78 134
18 ALIBORI BANIKOARA EPP YOSSINENDE R 47 54 101
19 ALIBORI BANIKOARA EPP BAM-BOUH R 29 27 56
20 ALIBORI BANIKOARA EPP BONNI R 77 58 135
21 ALIBORI BANIKOARA EPP DOMBOURE-PEULH R 38 49 87
22 ALIBORI BANIKOARA EPP DONDANGOU R 38 52 90
23 ALIBORI BANIKOARA EPP GANDARA R 34 43 77
24 ALIBORI BANIKOARA EPP MAMA GBÉROU R 55 81 136
25 ALIBORI BANIKOARA
EPP PATATCHAGOU￾DOUROU R 37 29 66
26 ALIBORI BANIKOARA EPP BIROU WIRI R 66 62 128
27 ALIBORI BANIKOARA EPP GOUBAGOU R 37 27 64
28 ALIBORI BANIKOARA EPP MOUROUSSE R 23 45 68
29 ALIBORI BANIKOARA EPP SARÉGBÉROU R 55 29 84
30 ALIBORI BANIKOARA EPP SONWARI R 74 87 161
31 ALIBORI BANIKOARA EPP SOUMABOUGOU R 100 77 177
32 ALIBORI BANIKOARA EPP YAMBEROU R 78 116 194
33 ALIBORI BANIKOARA EPP SIRIKOU R 85 97 182
34 ALIBORI BANIKOARA EPP WABAGOU R 49 43 92
35 ALIBORI BANIKOARA EPP BONIKI R 95 90 185
36 ALIBORI BANIKOARA EPP KIHOUHOU R 38 28 66
37 ALIBORI BANIKOARA EPP MONDARA R 53 32 85
38 ALIBORI BANIKOARA EPP SOMBI-KEREGOU R 69 47 116
39 ALIBORI BANIKOARA EPP ANGARADEBOU R 26 29 55
40 ALIBORI BANIKOARA EPP DAGBAROU R 47 36 83
41 ALIBORI BANIKOARA EPP DEROU-BOUH R 47 56 103
42 ALIBORI BANIKOARA EPP PARIS R 41 29 70
43 ALIBORI BANIKOARA EPP SOMBEDOU R 61 66 127
44 ALIBORI BANIKOARA EPP YABADOU R 23 40 63
45 ALIBORI BANIKOARA EPP KEREMOU R 39 56 95
46 ALIBORI BANIKOARA EPP MEKROU R 91 70 161
47 ALIBORI BANIKOARA EPP BADAGOU R 42 69 111
48 ALIBORI BANIKOARA
EPP BIBOUBA￾WANRAGOU R 52 82 134
49 ALIBORI BANIKOARA EPP KAKOUROGOU R 45 114 159
50 ALIBORI BANIKOARA EPP KANTANKOUNIN R 28 39 67
51 ALIBORI BANIKOARA EPP KOGNINKOU R 58 70 128
52 BORGOU KALALE EPP BAGOUDOURE R 16 18 34
53 BORGOU KALALE EPP BARAGANA R 40 53 93
54 BORGOU KALALE EPP BESSASSI/A R 130 150 280
55 BORGOU KALALE EPP BESSASSI/B R 125 139 264
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
97
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
56 BORGOU KALALE EPP CAMP-KALALE R 78 44 122
57 BORGOU KALALE EPP DADAGOURGNIAN R 38 47 85
58 BORGOU KALALE EPP DANTCHA R 56 55 111
59 BORGOU KALALE EPP DIAMBADI R 42 52 94
60 BORGOU KALALE EPP DJILIDJALARE R 37 30 67
61 BORGOU KALALE EPP DOUKOM R 36 30 66
62 BORGOU KALALE EPP DUNKASSA/A R 115 140 255
63 BORGOU KALALE EPP DUNKASSA/B R 145 133 278
64 BORGOU KALALE EPP GNEL-DEMON R 59 66 125
65 BORGOU KALALE EPP GNEL-DOGARI R 44 43 87
66 BORGOU KALALE EPP GNEL- GNABOUCA R 49 59 108
67 BORGOU KALALE EPP GNEL GOUISSA R 39 42 81
68 BORGOU KALALE EPP GNEL KELE R 49 57 106
69 BORGOU KALALE EPP GNEL KIAGOU R 47 61 108
70 BORGOU KALALE EPP GNEL MERE R 30 16 46
71 BORGOU KALALE EPP GNEL YAKAN R 20 54 74
72 BORGOU KALALE EPP GNEL YINCOUBLE R 37 48 85
73 BORGOU KALALE EPP KALEL GOUDEL R 50 31 81
74 BORGOU KALALE EPP KOPRE R 81 46 127
75 BORGOU KALALE EPP LOU R 119 134 253
76 BORGOU KALALE EPP NINZILE R 23 34 57
77 BORGOU KALALE EPP SONKOROU R 42 33 75
78 BORGOU KALALE EPP TEPARE SORE R 74 73 147
79 BORGOU KALALE EPP YEKENA R 79 48 127
80 BORGOU BEMBEREKE
EPP BAGOUDOU￾KPAROU R 30 25 55
81 BORGOU BEMBEREKE EPP GAMARE R 114 81 195
83 BORGOU BEMBEREKE EPP KENEKOU R 74 81 155
84 BORGOU BEMBEREKE
EPP KEROUKPOGOH (EX 
WARANKEROU) R 147 108 255
85 BORGOU BEMBEREKE EPP PEDAROU GANDO R 19 15 34
86 BORGOU BEMBEREKE EPP PEDAROU/A R 152 118 270
87 BORGOU BEMBEREKE EPP PEDAROU/B R 130 126 256
88 BORGOU BEMBEREKE EPP SAORE R 238 173 411
89 BORGOU BEMBEREKE EPP WANRAROU/A R 182 137 319
90 BORGOU BEMBEREKE EPP DARIGOUROU R 116 108 224
91 BORGOU BEMBEREKE EPP KPREGOUROU A R 146 112 258
92 BORGOU BEMBEREKE EPP KPREGOUROU B R 106 87 193
93 BORGOU BEMBEREKE EPP SANSE R 23 33 56
94 BORGOU BEMBEREKE EPP SOMBOUAN 1 R 129 106 235
95 BORGOU BEMBEREKE EPP SOMBOUAN 2 R 88 76 164
96 BORGOU BEMBEREKE EPP DONWARI R 24 17 41
97 BORGOU BEMBEREKE EPP GUERA N’KALI R 195 153 348
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
98
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
98 BORGOU BEMBEREKE EPP TASSE R 125 116 241
99 BORGOU BEMBEREKE EPP KONAROU R 14 13 27
100 BORGOU BEMBEREKE EPP SONGOURA R 160 198 358
104 BORGOU BEMBEREKE EPP GANRO R 147 147 294
105 BORGOU BEMBEREKE EPP KORE R 32 44 76
106 BORGOU BEMBEREKE EPP MANI R 59 55 114
107 BORGOU BEMBEREKE EPP NASSA R 135 148 283
108 BORGOU BEMBEREKE EPP GANDO-BAGA R 45 47 92
109 BORGOU BEMBEREKE EPP GANDO-BOU R 142 155 297
111 BORGOU BEMBEREKE EPP KOUHOURA R 16 21 37
112 BORGOU BEMBEREKE SOMBOUROU-SON R 99 90 189
113 BORGOU BEMBEREKE EPP WANRAROU/B R 92 91 183
12 ALIBORI BANIKOARA EPP BONHANROU/A U 110 111 221
13 ALIBORI BANIKOARA EPP BONHANROU/B U 76 62 138
14 ALIBORI BANIKOARA
EPP GABAGA (EX￾GOMPAROU PEULH) U 72 98 170
16 ALIBORI BANIKOARA EPP KPESSANROU U 36 70 106
82 BORGOU BEMBEREKE EPP GUERE U 163 151 314
101 BORGOU BEMBEREKE EPP BEREKE/B U 141 131 272
102 BORGOU BEMBEREKE EPP BEREKE-SUD A U 139 128 267
103 BORGOU BEMBEREKE EPP BEREKE-SUD B U 107 99 206
110 BORGOU BEMBEREKE EPP BEREKE/A U 89 139 228
1 ALIBORI BANIKOARA EPP GNIGOUWIROU R 42 51 93
2 ALIBORI BANIKOARA EPP ABIDJAN R 39 37 86
3 ALIBORI BANIKOARA EPP KINGAROU R 52 47 99
4 ALIBORI BANIKOARA EPP TANTANI R 28 52 80
5 ALIBORI BANIKOARA EPP TISSOUA R 40 37 77
6 ALIBORI BANIKOARA EPP TOUMAROU R 62 76 138
7 ALIBORI BANIKOARA EPP BANIGOUROU R 48 37 85
8 ALIBORI BANIKOARA
EPP SAHADOUROU￾BOURO R 25 30 55
9 ALIBORI BANIKOARA
EPP GOUNON NON 
SOUAROU R 26 33 59
10 ALIBORI BANIKOARA
EPP KPAKO 
WONGNAGOU R 32 15 47
11 ALIBORI BANIKOARA EPP KOKIBOROU-PEULH R 27 31 58
15 ALIBORI BANIKOARA EPP GABE-GBEROU R 49 56 105
17 ALIBORI BANIKOARA EPP TIGANSON R 56 78 134
18 ALIBORI BANIKOARA EPP YOSSINENDE R 47 54 101
19 ALIBORI BANIKOARA EPP BAM-BOUH R 29 27 56
20 ALIBORI BANIKOARA EPP BONNI R 77 58 135
21 ALIBORI BANIKOARA EPP DOMBOURE-PEULH R 38 49 87
22 ALIBORI BANIKOARA EPP DONDANGOU R 38 52 90
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 21 – COMPLETE LIST OF INTERVENTION SCHOOLS AND CONTROL 
SCHOOLS (CONTINUED) 
99
N° Department Commune School Name Rural/Urban
School Population
Female Male Total
23 ALIBORI BANIKOARA EPP GANDARA R 34 43 77
24 ALIBORI BANIKOARA EPP MAMA GBÉROU R 55 81 136
25 ALIBORI BANIKOARA
EPP PATATCHAGOU￾DOUROU R 37 29 66
26 ALIBORI BANIKOARA EPP BIROU WIRI R 66 62 128
27 ALIBORI BANIKOARA EPP GOUBAGOU R 37 27 64
28 ALIBORI BANIKOARA EPP MOUROUSSE R 23 45 68
29 ALIBORI BANIKOARA EPP SARÉGBÉROU R 55 29 84
30 ALIBORI BANIKOARA EPP SONWARI R 74 87 161
31 ALIBORI BANIKOARA EPP SOUMABOUGOU R 100 77 177
32 ALIBORI BANIKOARA EPP YAMBEROU R 78 116 194
33 ALIBORI BANIKOARA EPP SIRIKOU R 85 97 182
34 ALIBORI BANIKOARA EPP WABAGOU R 49 43 92
Total 7915 7915 15840
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
100
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM 
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
100 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 10 Fewer than 16 
tested (see 
General 
Comments)
101 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 13 Fewer than 16 
tested (see 
General 
Comments)
102 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
103 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 1 Only one class (CE2) 
was sampled, 
because this school 
has only 3 classes (CI, 
CP and CE2) and 
Classroom 
Observation only 
deals with CE1, CE2, 
CM1 and CM2
Completed Completed 16
104 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
105 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
106 Completed 6 10 2 Completed Completed 16
107 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
108 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
109 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
110 Completed 0 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
111 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
112 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
113 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 1 Time limitations Completed Completed 16
114 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
115 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
101
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
116 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 15 Fewer than 16 
tested (see 
General 
Comments)
117 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
119 Completed 6 10 2 Completed Completed 16
121 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
122 Not completed The director 
was absent
5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 14 Fewer than 16 
tested (see 
General 
Comments)
123 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
124 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
125 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
126 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
128 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 14 Fewer than 16 
tested (see 
General 
Comments)
129 Completed 6 10 2 Completed Completed 16
130 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
131 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
132 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
133 Completed 1 Fewer than 6 
sampled (see 
General Comments)
9 Fewer than 
10 sampled 
(see 
General 
Comments)
1 Classroom 
Observation was 
done only in one 
class because this 
school has only one 3 
classes (CM1, CP and 
CE1) and one teacher
Completed Completed 16
134 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 0 No students were 
present on the 
sampling day in the 
classes concerned 
with the observations 
Completed Completed 16
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
102
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
135 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
136 Not completed School not 
sampled (see 
General 
Comments)
0 School not sampled 
(see General 
Comments)
0 School not 
sampled 
(see 
General 
Comments)
0 School not sampled 
(see General 
Comments)
Not completed School not 
sampled (see 
General 
Comments)
Not completed School not 
sampled (see 
General 
Comments)
0 School not 
sampled (see 
General 
Comments)
137 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
138 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
139 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
140 Completed 6 10 2 Completed Completed 16
141 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 0 No Grade 2 in 
this school 
143 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
144 Completed 6 10 2 Completed Completed 16
145 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
146 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
147 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
148 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
149 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 1 Completed Completed 16
151 Completed 0 This school has only 
one class (CM2 with 
only 7 students 
present), and the 
school director is the 
one teaching in that 
class
10 1 Only one class is 
operational officially 
in this school (CM2). 
Completed Completed 0 No Grade 2 in 
this school 
152 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
153 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
103
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
154 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 0 No Grade 2 in 
this school 
155 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 8 Fewer than 16 
tested (see 
General 
Comments)
156 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 1 Only CM1 was 
sampled because 
only one teacher was 
available (teachers in 
other classes were no 
longer present since 
they had not been 
paid)
Completed Completed 2 Fewer than 16 
tested (see 
General 
Comments)
157 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 0 No Grade 2 in 
this school 
158 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
159 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 11 Fewer than 16 
tested (see 
General 
Comments)
161 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
163 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
165 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 6 Fewer than 16 
tested (see 
General 
Comments)
166 Completed 2 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 0 No Grade 2 in 
this school 
167 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
168 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 14 Fewer than 16 
tested (see 
General 
Comments)
169 Completed 5 Fewer than 6 
sampled (see 
General Comments)
9 Fewer than 
10 sampled 
(see 
General 
Comments)
2 Completed Completed 16
170 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
104
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
171 Completed 2 Fewer than 6 
sampled (see 
General Comments)
9 Fewer than 
10 sampled 
(see 
General 
Comments)
1 Time limitations Not completed Completed 15 One student 
did not give 
consent.
172 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
173 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
174 Completed 0 There were no 
teachers present the 
day the survey was 
conducted
10 1 Time limitations Completed Completed 1 Fewer than 16 
tested (see 
General 
Comments)
175 Not completed School not 
sampled (see 
General 
Comments)
0 School not sampled 
(see General 
Comments)
0 School not 
sampled 
(see 
General 
Comments)
0 School not sampled 
(see General 
Comments)
Not completed School not 
sampled (see 
General 
Comments)
Not completed School not 
sampled (see 
General 
Comments)
0 School not 
sampled (see 
General 
Comments)
176 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 1 Time limitations Completed Completed 16
177 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
178 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
179 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
180 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 15 Fewer than 16 
tested (see 
General 
Comments)
181 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 1 Both CE1 and CE2 
classes are mixed 
(study together) but 
their teacher was 
absent; CM2 
students went to 
attend the white 
examination, which is 
why Classroom 
Observation was 
done in only one 
class (CM1)
Completed Completed 15 Fewer than 16 
tested (see 
General 
Comments)
182 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
183 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 15 Fewer than 16 
tested (see 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
105
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
General 
Comments)
184 Completed 0 The school has only 
3 classes (CP, CE1, 
CM2), and only one 
teacher was present
10 2 Completed Completed 8 Fewer than 16 
tested (see 
General 
Comments)
185 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Not completed Survey not 
uploaded to 
database
Completed 16
186 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
187 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
188 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
189 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
190 Completed 1 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
191 Completed 2 Fewer than 6 
sampled (see 
General Comments)
7 Fewer than 
10 sampled 
(see 
General 
Comments)
1 Only one class (CE1) 
was sampled, 
because this school 
has only 3 classes (CI, 
CE1 and CM1) and 
Classroom 
Observation only 
deals with CE1, CE2, 
CM1 and CM2, and 
because CM1 
students were not in 
class on the day of 
sampling 
Completed Completed 0 No Grade 2 in 
this school 
192 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
193 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
194 Completed 0 The school has only 
2 teachers and one 
director who also 
teaches; the 2 
teachers were not 
present the day the 
survey was 
conducted (working 
session out of the 
school)
10 2 Completed Completed 16
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
106
ID 
École
Director Survey Comment 
(Director 
Survey)
# of 
Teacher 
Surveys 
Completed
Comment (Teacher 
Surveys)
# of Parent 
Surveys 
Completed
Comment 
(Parent 
Surveys)
# of 
Classroom 
Observations 
Completed
Comment (Classroom 
Observations)
Environmental 
Observation 
Survey
Comment 
(Environmental 
Observation 
Survey)
Handwashing 
Observation 
Survey
Comment 
(Handwashing 
Observation 
Survey)
# of EGRA 
Tests 
Completed
Comment 
(EGRA Tests)
195 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
196 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 15 Fewer than 16 
tested (see 
General 
Comments)
197 Completed 0 The teachers were 
absent at sampling 
10 1 This school has only 4 
classes (CI, CP, CE1 
and CE2), but 
because the classes 
targeted by 
Classroom 
Observation (CE1 and 
CE2) are mixed (i.e. 
study together), only 
one class was 
sampled.
Completed Completed 2 Fewer than 16 
tested (see 
General 
Comments)
198 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
199 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
200 Completed 3 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
201 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
220 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
242 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
250 Completed 4 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
260 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
264 Completed 5 Fewer than 6 
sampled (see 
General Comments)
10 2 Completed Completed 16
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 22 – DETAILS ON SURVEYS DONE BY THE EVALUATION TEAM (CONTINUED) 
107
General Comments
Data Collection 
Aspect
Comment Impact/mitigation measures
School Sampling Two schools were not sampled. They were supposed to be 
sampled the day we noticed the inappropriate sampling method 
used, and enumerators were asked to stop collecting data until 
everything was sorted out. 
Still had 48 treatment schools and 49 control schools; thus, 
greater than or equal to the number necessary (48). A total of 
96 Principals were surveyed, which was greater than the 95 
sampled at baseline. 
Teacher Surveys Generally, when fewer than 6 were sampled it was because the 
number sampled were present at the school. For example, only 
about 65% of schools had all 6 grades present. Also, some 
schools have the director teaching in up to 4 classes (sometimes 
no other teacher present in the school).
A total of 339 teachers were surveyed, which is somewhat less 
than the 399 sampled at baseline. It should be noted that the 
strike may have affected the number of teachers still working at 
the schools. 
Parent Surveys In those rare cases where 10 parents were not surveyed, it was 
because not all parents who had been called for the interview 
could come due to the agricultural season (they had to go to 
their fields).
Targets were almost always met. No impact on results. 
EGRA Tests There were several schools that had fewer than 16 students in 
the CP class
More than 1400 students were sampled. Power will be 
sufficient to detect change (see letter from Statlog).
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
108
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION 
To address concerns regarding the potential bias introduced by the non-use of RNG for sample selection 
for the parent and EGRA surveys, ADVISEM suggested using a specialized external firm to get a feedback 
on how to address those concerns and still use the data collected. This specialized firm through ADVISEM 
sent their suggestions to CRS. CRS agreed to a series of statistical tests. For the tests below, we utilized 
the basic equation annd R-code (with some adjustments for aspects such as number of communes and 
enumerators) sent by Stacy Prieto from CRS Headquarters. For example:
𝐶ℎ𝑖𝑙𝑑_𝑠𝑖𝑐𝑘𝑖𝑗𝑘 = 𝛽0 + 𝛽2𝑆𝑢𝑠𝑝𝑒𝑐𝑡𝐷𝑎𝑦𝑗 + 𝛽4𝑇𝑟𝑒𝑎𝑡𝑒𝑑𝑘 + 𝛽3𝐹𝑒𝑚𝑎𝑙𝑒𝑖 + 𝛽3𝑈𝑟𝑏𝑎𝑛𝑆𝑐ℎ𝑜𝑜𝑙𝑘 + 𝜹𝑚𝑬𝑗𝑚
+ 𝜶𝑝𝑪𝑘𝑝 + 𝜀𝑖𝑗𝑘
where Child_sick [0,1] is the dependent variable that takes the value 1 if child i in family j in school district 
k was sick in May 2018, and 0 otherwise. SuspectDay is a binary variable [0,1] that takes the value 1 if 
parent j was surveyed on one of the first 3 days, and zero otherwise. Treated is a binary variable [0,1] that 
takes the value 1 if school k was a treated school, and zero otherwise. Female is a binary variable [0,1] 
that takes the value 1 if child i is female, and zero otherwise. Urban is a binary variable [0,1] that takes the 
value 1 if school k is classified as urban, and zero otherwise. Ejm
10 is a vector of binary variables [0,1] 
denoting which of the M enumerators surveyed family j, excluding enumerator XX... Ckp is a vector of 
binary variables [0,1] denoting in which of the P communes school k is located, excluding commune XX... 
𝜀𝑖𝑗𝑘𝑚𝑝is the error term.
CRS Benin asked for different tests to be done on the parents’ survey. Some of these tests resulted in an 
immense number of explanatory variables (interpretation is limited).
Below are the results of such tests.
PARENT SURVEY BIAS CHECK 
After removing missing values some variables have only one modality. In addition, the number of 
explanatory variables is immense so the models manage '' NA '' in places
Model 1 
Dependent variable : child_sick (If a child in the household was sick)
Call:
glm(formula = child_sick ~ ., family = binomial(link = "logit"), 
 data = data.bias.check)
Deviance Residuals: 
 Min 1Q Median 3Q Max 
-1.6189 -0.7982 -0.7025 0.9928 2.3122 
 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
109
Coefficients: (1 not defined because of singularities)
 Estimate Std. Error z value Pr(>|z|) 
(Intercept) 3.516e+13 1.904e+13 1.846 0.0649 . 
SuspectDay1 -2.179e-01 1.696e-01 -1.285 0.1988 
Treated1 -4.190e-02 3.670e-01 -0.114 0.9091 
Female1 8.558e-03 1.109e-01 0.077 0.9385 
UrbanSchool1 -4.119e-01 1.941e-01 -2.123 0.0338 * 
commune11 -2.176e-01 5.521e-01 -0.394 0.6934 
commune21 -5.148e-02 4.455e-01 -0.116 0.9080 
cpmmune31 4.517e-02 2.742e-01 0.165 0.8691 
commune41 4.769e-02 4.797e-01 0.099 0.9208 
commune51 1.594e-02 2.333e-01 0.068 0.9455 
commune61 NA NA NA NA 
`Enumerator 1`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 2`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 3`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 4`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 5`1 -4.539e+15 1.904e+13 -238.325 <2e-16 ***
`Enumerator 6`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 7`1 -4.539e+15 1.904e+13 -238.325 <2e-16 ***
`Enumerator 8`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 9`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 10`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 11`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 12`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 13`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 14`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 15`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 16`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 17`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 18`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 19`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 20`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 21`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 22`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 23`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 24`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 25`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 26`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 27`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 28`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 29`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 30`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 31`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 32`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 33`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 34`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 73`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 75`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
`Enumerator 99`1 -3.516e+13 1.904e+13 -1.846 0.0649 . 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
110
Model 2 
Dependent variable: Diet (minimum acceptable diet score: Constructed from questions 11 through 17)
Call:
glm(formula = diet ~ ., family = binomial(link = "logit"), data = data.bias.check)
Deviance Residuals: 
 Min 1Q Median 3Q Max 
-2.7609 -0.5868 0.4663 0.7757 2.3153 
Coefficients: (1 not defined because of singularities)
 Estimate Std. Error z value Pr(>|z|) 
(Intercept) -3.773e+13 2.113e+13 -1.786 0.07411 . 
SuspectDay1 1.848e-01 2.012e-01 0.919 0.35833 
Treated1 1.051e+00 4.190e-01 2.509 0.01211 * 
Female1 7.978e-02 1.216e-01 0.656 0.51169 
UrbanSchool1 -2.097e-01 2.122e-01 -0.988 0.32306 
commune11 6.444e-01 6.141e-01 1.049 0.29403 
commune21 6.993e-01 5.138e-01 1.361 0.17349 
cpmmune31 -8.529e-01 3.003e-01 -2.840 0.00451 ** 
commune41 1.309e+00 5.249e-01 2.494 0.01264 * 
commune51 -1.011e-01 2.583e-01 -0.392 0.69540 
commune61 NA NA NA NA 
`Enumerator 1`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 2`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 3`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 4`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 5`1 4.541e+15 2.113e+13 214.972 < 2e-16 ***
`Enumerator 6`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 7`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 8`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 9`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 10`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 11`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 12`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 13`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 14`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 15`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 16`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 17`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 18`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 19`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 20`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 21`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 22`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 23`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 24`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 25`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 26`1 3.773e+13 2.113e+13 1.786 0.07411 . 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
111
`Enumerator 27`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 28`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 29`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 30`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 31`1 -4.466e+15 2.113e+13 -211.400 < 2e-16 ***
`Enumerator 32`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 33`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 34`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 73`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 75`1 3.773e+13 2.113e+13 1.786 0.07411 . 
`Enumerator 99`1 3.773e+13 2.113e+13 1.786 0.07411 . 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 3 
Dependent variable: using microfinance (If parents use savings/ credit for the education of their children: 
Construct a binary variable [0,1])
Call:
glm(formula = finance ~ ., family = binomial(link = "logit"), 
 data = data.bias.check[, -c(33, 41, 45, 47, 48)])
Deviance Residuals: 
 Min 1Q Median 3Q Max 
-8.49 0.00 0.00 0.00 8.49 
Coefficients: (2 not defined because of singularities)
 Estimate Std. Error z value Pr(>|z|) 
(Intercept) 9.925e+14 7.799e+07 12726982 <2e-16 ***
SuspectDay1 -3.266e+14 1.418e+07 -23039268 <2e-16 ***
Treated1 -9.171e+14 3.026e+07 -30308386 <2e-16 ***
Female1 -9.730e+13 8.441e+06 -11527334 <2e-16 ***
UrbanSchool1 -4.384e+14 1.344e+07 -32621311 <2e-16 ***
commune11 4.257e+14 4.576e+07 9303064 <2e-16 ***
commune21 -8.075e+14 3.609e+07 -22372163 <2e-16 ***
cpmmune31 1.563e+14 1.645e+07 9500222 <2e-16 ***
commune41 3.317e+15 3.368e+07 98480487 <2e-16 ***
commune51 4.602e+14 1.632e+07 28199069 <2e-16 ***
commune61 NA NA NA NA 
`Enumerator 1`1 -3.516e+15 7.587e+07 -46348035 <2e-16 ***
`Enumerator 2`1 -3.814e+15 7.457e+07 -51148269 <2e-16 ***
`Enumerator 3`1 -1.118e+14 7.985e+07 -1399802 <2e-16 ***
`Enumerator 4`1 -1.892e+15 7.269e+07 -26030310 <2e-16 ***
`Enumerator 5`1 -2.078e+14 9.706e+07 -2140540 <2e-16 ***
`Enumerator 6`1 -7.105e+15 7.571e+07 -93849550 <2e-16 ***
`Enumerator 7`1 -7.948e+15 1.026e+08 -77506929 <2e-16 ***
`Enumerator 8`1 -8.867e+14 7.150e+07 -12401533 <2e-16 ***
`Enumerator 9`1 -1.052e+15 7.709e+07 -13641627 <2e-16 ***
`Enumerator 10`1 3.436e+15 7.220e+07 47584364 <2e-16 ***
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
112
`Enumerator 11`1 -1.043e+14 7.194e+07 -1449420 <2e-16 ***
`Enumerator 12`1 4.998e+15 8.045e+07 62119034 <2e-16 ***
`Enumerator 13`1 -4.218e+15 8.117e+07 -51957301 <2e-16 ***
`Enumerator 14`1 -1.224e+15 7.117e+07 -17203608 <2e-16 ***
`Enumerator 15`1 1.614e+15 7.134e+07 22627656 <2e-16 ***
`Enumerator 16`1 -2.739e+15 8.332e+07 -32869263 <2e-16 ***
`Enumerator 17`1 4.488e+15 7.267e+07 61762733 <2e-16 ***
`Enumerator 18`1 1.902e+15 7.446e+07 25541491 <2e-16 ***
`Enumerator 19`1 -2.409e+14 7.149e+07 -3369615 <2e-16 ***
`Enumerator 20`1 -1.382e+15 7.222e+07 -19133014 <2e-16 ***
`Enumerator 21`1 -4.329e+14 7.213e+07 -6001214 <2e-16 ***
`Enumerator 23`1 7.123e+14 7.316e+07 9736163 <2e-16 ***
`Enumerator 24`1 1.307e+15 7.223e+07 18091495 <2e-16 ***
`Enumerator 25`1 2.035e+14 7.719e+07 2636579 <2e-16 ***
`Enumerator 26`1 -2.310e+15 7.285e+07 -31708260 <2e-16 ***
`Enumerator 27`1 2.562e+14 7.988e+07 3207404 <2e-16 ***
`Enumerator 28`1 -1.046e+15 7.826e+07 -13363835 <2e-16 ***
`Enumerator 29`1 1.591e+15 7.472e+07 21292997 <2e-16 ***
`Enumerator 31`1 -2.965e+15 7.972e+07 -37188346 <2e-16 ***
`Enumerator 32`1 -6.357e+14 7.293e+07 -8716778 <2e-16 ***
`Enumerator 33`1 3.085e+15 8.743e+07 35290709 <2e-16 ***
`Enumerator 73`1 NA NA NA NA 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 4 
Dependent variable: Quality learning (If parents felt that the FFE program improved the quality of 
children’s learning)
t test of coefficients:
 OR 2.5 % 97.5 % p 
1/(Intercept) 1.4799e+40 9.8865e-70 2.2154e+149 0.470814 
1/SuspectDay1 4.8430e+14 1.4229e-137 1.6483e+166 0.849354 
1/Female1 2.3877e-64 NA NA NA 
1/UrbanSchool1 1.9357e-111 1.9357e-111 0.0000e+00 < 2.2e-16 ***
1/commune21 1.0624e+05 NA NA NA 
1/cpmmune31 5.0449e-37 9.0327e-80 2.8177e+06 0.096066 . 
1/commune41 3.5033e+60 4.4154e-92 2.7796e+212 0.434681 
1/commune51 2.6907e+44 2.6728e+44 2.7088e+44 < 2.2e-16 ***
1/commune61 2.9293e-34 2.8102e-34 0.0000e+00 < 2.2e-16 ***
1/`Enumerator 1`1 2.8225e+57 8.6443e-95 9.2157e+208 0.457376 
1/`Enumerator 2`1 1.5546e+49 6.3976e+48 3.7778e+49 < 2.2e-16 ***
1/`Enumerator 3`1 4.0844e+13 4.0844e+13 4.0844e+13 < 2.2e-16 ***
1/`Enumerator 4`1 7.6061e+14 7.6061e+14 7.6061e+14 < 2.2e-16 ***
1/`Enumerator 5`1 2.6784e+11 2.6784e+11 2.6784e+11 < 2.2e-16 ***
1/`Enumerator 6`1 3.8261e+04 NA NA NA 
1/`Enumerator 8`1 5.0105e+30 4.9771e+30 5.0441e+30 < 2.2e-16 ***
1/`Enumerator 9`1 8.3703e+13 NA NA NA 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
113
1/`Enumerator 10`1 1.3830e+63 1.3268e+63 1.4415e+63 < 2.2e-16 ***
1/`Enumerator 11`1 2.8737e+56 2.8737e+56 2.8737e+56 < 2.2e-16 ***
1/`Enumerator 12`1 6.2754e+46 6.2754e+46 6.2754e+46 < 2.2e-16 ***
1/`Enumerator 13`1 8.2894e+12 8.2894e+12 8.2894e+12 < 2.2e-16 ***
1/`Enumerator 14`1 5.0891e-85 5.0891e-85 0.0000e+00 < 2.2e-16 ***
1/`Enumerator 15`1 8.6467e+113 1.5481e+71 4.8294e+156 1.751e-07 ***
1/`Enumerator 16`1 5.7847e+35 NA NA NA 
1/`Enumerator 17`1 5.1497e+34 5.1497e+34 5.1497e+34 < 2.2e-16 ***
1/`Enumerator 18`1 5.5398e+17 5.5398e+17 5.5398e+17 < 2.2e-16 ***
1/`Enumerator 19`1 3.2040e+22 3.2040e+22 3.2040e+22 < 2.2e-16 ***
1/`Enumerator 20`1 2.0136e+27 2.0136e+27 2.0136e+27 < 2.2e-16 ***
1/`Enumerator 21`1 1.5453e-245 NA NA NA 
1/`Enumerator 22`1 2.3836e+14 NA NA NA 
1/`Enumerator 23`1 3.6913e+42 3.6913e+42 3.6913e+42 < 2.2e-16 ***
1/`Enumerator 24`1 6.0494e+41 6.0491e+41 6.0498e+41 < 2.2e-16 ***
1/`Enumerator 25`1 4.5648e-136 4.5648e-136 0.0000e+00 < 2.2e-16 ***
1/`Enumerator 26`1 8.0023e+30 8.0023e+30 8.0023e+30 < 2.2e-16 ***
1/`Enumerator 27`1 4.6616e+10 4.6616e+10 4.6616e+10 < 2.2e-16 ***
1/`Enumerator 28`1 4.9584e+07 NA NA NA 
1/`Enumerator 29`1 9.9895e+16 9.9895e+16 9.9895e+16 < 2.2e-16 ***
1/`Enumerator 30`1 4.0936e+05 NA NA NA 
1/`Enumerator 31`1 4.4746e+02 4.4746e+02 4.4746e+02 < 2.2e-16 ***
1/`Enumerator 32`1 4.2848e+24 NA NA NA 
1/`Enumerator 34`1 6.2377e-157 NA NA NA 
1/`Enumerator 73`1 1.5645e-173 1.5645e-173 0.0000e+00 < 2.2e-16 ***
1/`Enumerator 75`1 1.0535e+06 NA NA NA 
1/`Enumerator 99`1 4.6587e+02 NA NA NA 
2/(Intercept) 8.0240e+17 8.4457e+12 7.6233e+22 1.792e-12 ***
2/SuspectDay1 8.8315e-136 5.2099e-151 0.0000e+00 < 2.2e-16 ***
2/Female1 1.0649e+00 7.6884e-02 1.4750e+01 0.962592 
2/UrbanSchool1 4.8695e-21 4.8660e-21 0.0000e+00 < 2.2e-16 ***
2/commune21 2.2557e-160 NA NA NA 
2/cpmmune31 1.0438e+121 8.2233e+110 1.3249e+131 < 2.2e-16 ***
2/commune41 3.6471e-91 NA NA NA 
2/commune51 3.0332e+72 2.6356e+57 3.4908e+87 < 2.2e-16 ***
2/commune61 3.0808e+74 3.0785e+74 3.0830e+74 < 2.2e-16 ***
2/`Enumerator 1`1 3.8302e+238 5.8358e+200 2.5139e+276 < 2.2e-16 ***
2/`Enumerator 2`1 3.8890e+52 NA NA NA 
2/`Enumerator 3`1 2.7452e+00 2.7452e+00 2.7452e+00 < 2.2e-16 ***
2/`Enumerator 4`1 9.6714e+52 9.6643e+52 9.6784e+52 < 2.2e-16 ***
2/`Enumerator 5`1 9.9477e+05 NA NA NA 
2/`Enumerator 6`1 9.3193e+53 9.3193e+53 9.3193e+53 < 2.2e-16 ***
2/`Enumerator 8`1 2.4720e-14 NA NA NA 
2/`Enumerator 9`1 3.2370e+00 3.2370e+00 3.2370e+00 < 2.2e-16 ***
2/`Enumerator 10`1 9.0299e+48 9.0299e+48 9.0299e+48 < 2.2e-16 ***
2/`Enumerator 11`1 3.1451e+21 3.1451e+21 3.1451e+21 < 2.2e-16 ***
2/`Enumerator 12`1 6.5608e+37 NA NA NA 
2/`Enumerator 13`1 4.1763e+06 NA NA NA 
2/`Enumerator 14`1 9.2387e-83 NA NA NA 
2/`Enumerator 15`1 4.9002e-94 NA NA NA 
2/`Enumerator 16`1 4.6875e-01 NA NA NA 
2/`Enumerator 17`1 4.2190e+07 NA NA NA 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
114
2/`Enumerator 18`1 2.7612e+16 NA NA NA 
2/`Enumerator 19`1 3.5711e-94 NA NA NA 
2/`Enumerator 20`1 1.3743e-104 1.3743e-104 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 21`1 7.2836e-247 7.2836e-247 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 22`1 4.3787e+03 4.3787e+03 4.3787e+03 < 2.2e-16 ***
2/`Enumerator 23`1 5.8243e+128 3.1425e+118 1.0795e+139 < 2.2e-16 ***
2/`Enumerator 24`1 8.2414e+38 NA NA NA 
2/`Enumerator 25`1 7.3496e-04 5.7902e-14 9.3289e+06 0.543250 
2/`Enumerator 26`1 2.5347e+07 NA NA NA 
2/`Enumerator 27`1 6.2617e-04 6.2617e-04 6.0000e-04 < 2.2e-16 ***
2/`Enumerator 28`1 2.5983e-07 2.5983e-07 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 29`1 4.4282e+239 1.3847e+227 1.4161e+252 < 2.2e-16 ***
2/`Enumerator 30`1 3.8859e-02 3.8859e-02 3.8900e-02 < 2.2e-16 ***
2/`Enumerator 31`1 3.0002e+00 3.0002e+00 3.0002e+00 < 2.2e-16 ***
2/`Enumerator 32`1 1.0384e+18 1.0384e+18 1.0384e+18 < 2.2e-16 ***
2/`Enumerator 34`1 1.4658e-104 1.4658e-104 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 73`1 4.1988e-160 4.1988e-160 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 75`1 3.0708e-50 3.0708e-50 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 99`1 6.1629e-04 6.1629e-04 6.0000e-04 < 2.2e-16 ***
3/(Intercept) 3.0985e+30 1.1213e+28 8.5619e+32 < 2.2e-16 ***
3/SuspectDay1 7.3371e+55 3.4766e+39 1.5484e+72 1.981e-11 ***
3/Female1 4.7527e+00 8.0365e-01 2.8108e+01 0.085631 . 
3/UrbanSchool1 6.1461e-101 2.7546e-103 0.0000e+00 < 2.2e-16 ***
3/commune21 2.2755e-28 2.2755e-28 0.0000e+00 < 2.2e-16 ***
3/cpmmune31 6.2202e-64 1.2006e-75 0.0000e+00 < 2.2e-16 ***
3/commune41 2.0465e+160 1.6735e+132 2.5026e+188 < 2.2e-16 ***
3/commune51 7.7413e+24 3.5840e+22 1.6721e+27 < 2.2e-16 ***
3/commune61 1.3818e-64 2.7024e-76 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 1`1 4.4473e-252 4.4473e-252 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 2`1 6.5033e-111 6.5033e-111 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 3`1 2.3163e-99 2.3163e-99 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 4`1 4.0611e+77 6.7834e+72 2.4314e+82 < 2.2e-16 ***
3/`Enumerator 5`1 4.5295e-39 4.5295e-39 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 6`1 1.1653e-40 9.5289e-69 0.0000e+00 0.005326 ** 
3/`Enumerator 8`1 1.7507e-22 NA NA NA 
3/`Enumerator 9`1 6.7954e+119 2.5184e+119 1.8336e+120 < 2.2e-16 ***
3/`Enumerator 10`1 2.5623e-19 NA NA NA 
3/`Enumerator 11`1 4.9579e+15 4.9579e+15 4.9579e+15 < 2.2e-16 ***
3/`Enumerator 12`1 8.1888e-52 8.1888e-52 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 13`1 3.1867e-61 3.1867e-61 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 14`1 8.6697e-243 8.6697e-243 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 15`1 1.2993e+149 5.3214e+134 3.1725e+163 < 2.2e-16 ***
3/`Enumerator 16`1 6.9735e+96 2.7172e+96 1.7897e+97 < 2.2e-16 ***
3/`Enumerator 17`1 4.8753e+112 1.6184e+112 1.4686e+113 < 2.2e-16 ***
3/`Enumerator 18`1 1.9233e-67 1.9233e-67 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 19`1 2.4376e+147 9.1435e+146 6.4985e+147 < 2.2e-16 ***
3/`Enumerator 20`1 5.8462e+121 2.2035e+121 1.5510e+122 < 2.2e-16 ***
3/`Enumerator 21`1 2.6035e-24 2.3150e-31 0.0000e+00 5.536e-11 ***
3/`Enumerator 22`1 3.1125e+126 8.6346e+125 1.1220e+127 < 2.2e-16 ***
3/`Enumerator 23`1 2.2206e-52 2.2206e-52 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 24`1 4.5849e-22 NA NA NA 
3/`Enumerator 25`1 1.5744e-24 1.3365e-31 0.0000e+00 4.178e-11 ***
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
115
3/`Enumerator 26`1 2.6758e+101 6.3188e+100 1.1331e+102 < 2.2e-16 ***
3/`Enumerator 27`1 1.1115e+122 3.3981e+121 3.6357e+122 < 2.2e-16 ***
3/`Enumerator 28`1 4.0172e+42 7.7204e+41 2.0903e+43 < 2.2e-16 ***
3/`Enumerator 29`1 6.8042e+83 2.9411e+77 1.5741e+90 < 2.2e-16 ***
3/`Enumerator 30`1 1.6526e-115 1.6526e-115 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 31`1 1.6960e-45 1.6960e-45 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 32`1 2.7322e-54 2.7322e-54 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 34`1 0.0000e+00 0.0000e+00 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 73`1 9.2685e+44 4.3590e+42 1.9708e+47 < 2.2e-16 ***
3/`Enumerator 75`1 1.6770e+199 5.3993e+198 5.2088e+199 < 2.2e-16 ***
3/`Enumerator 99`1 2.2800e+212 2.2800e+212 2.2800e+212 < 2.2e-16 ***
4/(Intercept) 2.9703e+67 2.7100e+56 3.2557e+78 < 2.2e-16 ***
4/SuspectDay1 2.3201e+55 1.1121e+39 4.8405e+71 2.942e-11 ***
4/Female1 2.2893e+00 4.2649e-01 1.2288e+01 0.334031 
4/UrbanSchool1 4.5663e-101 2.1296e-103 0.0000e+00 < 2.2e-16 ***
4/commune21 1.0435e+112 2.0828e+111 5.2280e+112 < 2.2e-16 ***
4/cpmmune31 5.3312e-52 3.8179e-72 0.0000e+00 6.085e-07 ***
4/commune41 1.4188e+22 1.5058e-40 1.3369e+84 0.483574 
4/commune51 2.7206e+36 4.8739e+25 1.5186e+47 3.036e-11 ***
4/commune61 1.3833e-52 9.9834e-73 0.0000e+00 4.503e-07 ***
4/`Enumerator 1`1 1.3200e+33 4.5809e-24 3.8033e+89 0.250242 
4/`Enumerator 2`1 1.1122e-92 NA NA NA 
4/`Enumerator 3`1 2.3683e+50 5.6932e+49 9.8517e+50 < 2.2e-16 ***
4/`Enumerator 4`1 1.4772e+30 1.5927e+25 1.3701e+35 < 2.2e-16 ***
4/`Enumerator 5`1 1.6303e-71 NA NA NA 
4/`Enumerator 6`1 4.7921e+62 1.7120e+48 1.3414e+77 < 2.2e-16 ***
4/`Enumerator 8`1 2.6079e+20 8.7584e+19 7.7654e+20 < 2.2e-16 ***
4/`Enumerator 9`1 1.7355e+72 6.4316e+71 4.6829e+72 < 2.2e-16 ***
4/`Enumerator 10`1 1.7019e+22 5.5954e+21 5.1764e+22 < 2.2e-16 ***
4/`Enumerator 11`1 5.0647e+56 1.2996e+56 1.9739e+57 < 2.2e-16 ***
4/`Enumerator 12`1 2.0950e+56 5.4133e+55 8.1076e+56 < 2.2e-16 ***
4/`Enumerator 13`1 1.4826e+84 3.6190e+83 6.0734e+84 < 2.2e-16 ***
4/`Enumerator 14`1 4.7067e-72 2.5255e-79 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 15`1 3.4613e+101 9.0008e+86 1.3311e+116 < 2.2e-16 ***
4/`Enumerator 16`1 2.3150e-51 NA NA NA 
4/`Enumerator 17`1 2.0232e+65 6.9224e+64 5.9131e+65 < 2.2e-16 ***
4/`Enumerator 18`1 2.2639e+103 6.6409e+102 7.7179e+103 < 2.2e-16 ***
4/`Enumerator 19`1 8.1957e-38 NA NA NA 
4/`Enumerator 20`1 3.0893e-16 NA NA NA 
4/`Enumerator 21`1 6.1024e-296 6.1024e-296 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 22`1 1.4359e+78 2.9089e+77 7.0876e+78 < 2.2e-16 ***
4/`Enumerator 23`1 3.2948e-73 3.2948e-73 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 24`1 2.4208e-48 NA NA NA 
4/`Enumerator 25`1 8.3782e-72 4.8186e-79 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 26`1 4.8475e+53 1.2225e+53 1.9221e+54 < 2.2e-16 ***
4/`Enumerator 27`1 4.6632e+74 1.4256e+74 1.5253e+75 < 2.2e-16 ***
4/`Enumerator 28`1 7.8283e-06 1.3486e-06 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 29`1 1.3912e+36 9.5855e+29 2.0191e+42 < 2.2e-16 ***
4/`Enumerator 30`1 4.1145e+59 7.9581e+58 2.1272e+60 < 2.2e-16 ***
4/`Enumerator 31`1 1.6048e+173 1.6048e+173 1.6048e+173 < 2.2e-16 ***
4/`Enumerator 32`1 1.1010e+33 3.1530e+32 3.8443e+33 < 2.2e-16 ***
4/`Enumerator 34`1 0.0000e+00 0.0000e+00 0.0000e+00 < 2.2e-16 ***
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
116
4/`Enumerator 73`1 3.2610e-04 2.4460e-06 4.3500e-02 0.001300 ** 
4/`Enumerator 75`1 7.4667e+150 2.4040e+150 2.3191e+151 < 2.2e-16 ***
4/`Enumerator 99`1 1.0561e-73 1.0561e-73 0.0000e+00 < 2.2e-16 ***
5/(Intercept) 5.2201e+63 2.1928e+53 1.2427e+74 < 2.2e-16 ***
5/SuspectDay1 1.2648e+55 6.0950e+38 2.6244e+71 3.623e-11 ***
5/Female1 1.3003e+00 2.5697e-01 6.5796e+00 0.750931 
5/UrbanSchool1 4.0955e-101 1.9047e-103 0.0000e+00 < 2.2e-16 ***
5/commune21 5.8011e+109 1.1579e+109 2.9064e+110 < 2.2e-16 ***
5/cpmmune31 9.0071e-52 4.7786e-72 0.0000e+00 8.038e-07 ***
5/commune41 1.1544e+21 1.6622e-41 8.0170e+82 0.504431 
5/commune51 1.3750e+36 1.8228e+25 1.0372e+47 7.438e-11 ***
5/commune61 6.2941e-53 3.3565e-73 0.0000e+00 4.495e-07 ***
5/`Enumerator 1`1 4.3637e+39 2.4540e-18 7.7595e+96 0.174757 
5/`Enumerator 2`1 1.7223e+73 7.0875e+72 4.1852e+73 < 2.2e-16 ***
5/`Enumerator 3`1 2.8111e+55 6.7578e+54 1.1694e+56 < 2.2e-16 ***
5/`Enumerator 4`1 1.1424e+34 7.4879e+29 1.7430e+38 < 2.2e-16 ***
5/`Enumerator 5`1 2.5312e+92 NA NA NA 
5/`Enumerator 6`1 8.7544e+67 1.9437e+54 3.9429e+81 < 2.2e-16 ***
5/`Enumerator 8`1 1.2589e+25 4.2042e+24 3.7694e+25 < 2.2e-16 ***
5/`Enumerator 9`1 2.8578e-187 2.8578e-187 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 10`1 2.6961e+27 8.5717e+26 8.4803e+27 < 2.2e-16 ***
5/`Enumerator 11`1 5.1256e+59 1.3152e+59 1.9976e+60 < 2.2e-16 ***
5/`Enumerator 12`1 1.1332e+62 2.9280e+61 4.3854e+62 < 2.2e-16 ***
5/`Enumerator 13`1 7.7091e+87 1.8818e+87 3.1581e+88 < 2.2e-16 ***
5/`Enumerator 14`1 1.1054e-66 3.7790e-73 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 15`1 1.6632e+106 2.6954e+92 1.0262e+120 < 2.2e-16 ***
5/`Enumerator 16`1 4.8582e+53 1.8930e+53 1.2468e+54 < 2.2e-16 ***
Model 5 
Dependent variable: Health Quality (If parents felt that the FFE program improved the child’s health)
 OR 2.5 % 97.5 % p 
2/(Intercept) 2.5977e+00 1.1225e+00 6.0120e+00 0.0257613 * 
2/SuspectDay1 5.8728e-07 1.7692e-07 0.0000e+00 < 2.2e-16 ***
2/Treated1 2.5977e+00 1.1225e+00 6.0120e+00 0.0257613 * 
2/Female1 5.8016e+00 3.3306e-01 1.0106e+02 0.2278653 
2/UrbanSchool1 6.2658e+05 2.1832e+05 1.7983e+06 < 2.2e-16 ***
2/commune11 1.0000e+00 1.0000e+00 1.0000e+00 1.0000000 
2/commune21 2.7681e+00 2.7681e+00 2.7681e+00 < 2.2e-16 ***
2/cpmmune31 9.7666e+00 1.5762e+00 6.0516e+01 0.0143281 * 
2/commune41 1.6117e-03 1.6117e-03 1.6000e-03 < 2.2e-16 ***
2/commune51 4.1902e+01 1.7984e+01 9.7626e+01 < 2.2e-16 ***
2/commune61 1.4228e+00 1.9603e-01 1.0327e+01 0.7273227 
2/`Enumerator 1`1 7.8077e+04 NA NA NA 
2/`Enumerator 2`1 4.8535e-02 4.8535e-02 4.8500e-02 < 2.2e-16 ***
2/`Enumerator 3`1 3.9575e+02 3.9575e+02 3.9575e+02 < 2.2e-16 ***
2/`Enumerator 4`1 2.1887e-10 2.1887e-10 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 5`1 9.3276e+01 9.3276e+01 9.3276e+01 < 2.2e-16 ***
2/`Enumerator 6`1 6.1858e-04 6.1858e-04 6.0000e-04 < 2.2e-16 ***
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
117
2/`Enumerator 7`1 1.0000e+00 1.0000e+00 1.0000e+00 1.0000000 
2/`Enumerator 8`1 7.8889e-02 7.8889e-02 7.8900e-02 < 2.2e-16 ***
2/`Enumerator 9`1 1.2329e+20 4.8469e+19 3.1360e+20 < 2.2e-16 ***
2/`Enumerator 10`1 3.2483e+04 3.2483e+04 3.2483e+04 < 2.2e-16 ***
2/`Enumerator 11`1 1.0753e+02 1.0753e+02 1.0753e+02 < 2.2e-16 ***
2/`Enumerator 12`1 5.6938e+00 5.6938e+00 5.6938e+00 < 2.2e-16 ***
2/`Enumerator 13`1 7.9738e+00 7.9738e+00 7.9738e+00 < 2.2e-16 ***
2/`Enumerator 14`1 2.9258e+04 3.6571e+03 2.3407e+05 < 2.2e-16 ***
2/`Enumerator 15`1 9.0766e+10 3.5222e+10 2.3390e+11 < 2.2e-16 ***
2/`Enumerator 16`1 2.5732e+00 2.5732e+00 2.5732e+00 < 2.2e-16 ***
2/`Enumerator 17`1 4.1601e+00 4.1601e+00 4.1601e+00 < 2.2e-16 ***
2/`Enumerator 18`1 4.1372e+00 NA NA NA 
2/`Enumerator 19`1 1.2296e+05 1.2296e+05 1.2296e+05 < 2.2e-16 ***
2/`Enumerator 20`1 3.0832e+03 3.0832e+03 3.0832e+03 < 2.2e-16 ***
2/`Enumerator 21`1 1.7619e-22 1.7619e-22 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 22`1 5.6704e-01 NA NA NA 
2/`Enumerator 23`1 2.9092e+01 2.9092e+01 2.9092e+01 < 2.2e-16 ***
2/`Enumerator 24`1 5.3124e+06 5.3124e+06 5.3124e+06 < 2.2e-16 ***
2/`Enumerator 25`1 2.1356e+04 2.3119e+03 1.9727e+05 < 2.2e-16 ***
2/`Enumerator 26`1 4.1827e+01 4.1827e+01 4.1827e+01 < 2.2e-16 ***
2/`Enumerator 27`1 5.6014e+00 NA NA NA 
2/`Enumerator 28`1 1.3500e-03 NA NA NA 
2/`Enumerator 29`1 8.3086e+02 NA NA NA 
2/`Enumerator 30`1 8.1439e-02 8.1439e-02 8.1400e-02 < 2.2e-16 ***
2/`Enumerator 31`1 4.4605e-01 4.4605e-01 4.4600e-01 < 2.2e-16 ***
2/`Enumerator 32`1 1.2442e-02 1.2442e-02 1.2400e-02 < 2.2e-16 ***
2/`Enumerator 33`1 1.0000e+00 1.0000e+00 1.0000e+00 1.0000000 
2/`Enumerator 34`1 1.1960e-13 NA NA NA 
2/`Enumerator 73`1 9.2997e-24 9.2997e-24 0.0000e+00 < 2.2e-16 ***
2/`Enumerator 75`1 5.9059e-02 5.9059e-02 5.9100e-02 < 2.2e-16 ***
2/`Enumerator 99`1 7.7523e-01 NA NA NA 
3/(Intercept) 1.3641e+02 7.6074e+01 2.4460e+02 < 2.2e-16 ***
3/SuspectDay1 3.6941e+01 1.2699e+01 1.0746e+02 3.476e-11 ***
3/Treated1 1.3641e+02 7.6074e+01 2.4460e+02 < 2.2e-16 ***
3/Female1 5.2441e+00 9.1953e-01 2.9907e+01 0.0621087 . 
3/UrbanSchool1 9.1588e-09 2.1898e-09 0.0000e+00 < 2.2e-16 ***
3/commune11 1.0000e+00 NA NA NA 
3/commune21 8.8040e-01 8.8040e-01 8.8040e-01 < 2.2e-16 ***
3/cpmmune31 1.9905e-03 4.3769e-04 9.1000e-03 8.882e-16 ***
3/commune41 1.1226e+04 2.2405e+03 5.6245e+04 < 2.2e-16 ***
3/commune51 1.4455e+04 5.8890e+03 3.5479e+04 < 2.2e-16 ***
3/commune61 4.7970e-04 1.2677e-04 1.8000e-03 < 2.2e-16 ***
3/`Enumerator 1`1 5.9081e-07 5.9081e-07 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 2`1 1.3654e-09 NA NA NA 
3/`Enumerator 3`1 8.3899e-03 NA NA NA 
3/`Enumerator 4`1 2.4127e-12 2.4127e-12 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 5`1 6.2363e-03 NA NA NA 
3/`Enumerator 6`1 1.3466e+09 3.2316e+08 5.6114e+09 < 2.2e-16 ***
3/`Enumerator 7`1 1.0000e+00 1.0000e+00 1.0000e+00 1.0000000 
3/`Enumerator 8`1 1.4832e+06 3.7461e+05 5.8724e+06 < 2.2e-16 ***
3/`Enumerator 9`1 3.8324e-07 3.8324e-07 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 10`1 8.9589e-08 8.9589e-08 0.0000e+00 < 2.2e-16 ***
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
118
3/`Enumerator 11`1 7.4458e-05 7.4458e-05 1.0000e-04 < 2.2e-16 ***
3/`Enumerator 12`1 2.7666e-07 2.7666e-07 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 13`1 1.8397e-07 1.8397e-07 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 14`1 7.6150e-04 1.0505e-04 5.5000e-03 1.207e-12 ***
3/`Enumerator 15`1 1.8993e+11 7.8703e+10 4.5836e+11 < 2.2e-16 ***
3/`Enumerator 16`1 2.9995e+10 1.2653e+10 7.1108e+10 < 2.2e-16 ***
3/`Enumerator 17`1 2.7198e+10 1.0732e+10 6.8928e+10 < 2.2e-16 ***
3/`Enumerator 18`1 1.7039e-07 1.7039e-07 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 19`1 1.5291e-02 1.5291e-02 1.5300e-02 < 2.2e-16 ***
3/`Enumerator 20`1 5.5419e+13 2.8862e+13 1.0641e+14 < 2.2e-16 ***
3/`Enumerator 21`1 2.2377e-04 3.7755e-05 1.3000e-03 < 2.2e-16 ***
3/`Enumerator 22`1 6.6477e+09 2.0398e+09 2.1664e+10 < 2.2e-16 ***
3/`Enumerator 23`1 5.8766e+08 2.0985e+08 1.6457e+09 < 2.2e-16 ***
3/`Enumerator 24`1 3.7123e-02 3.7123e-02 3.7100e-02 < 2.2e-16 ***
3/`Enumerator 25`1 2.0657e-21 2.0657e-21 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 26`1 5.9839e+07 1.3927e+07 2.5712e+08 < 2.2e-16 ***
3/`Enumerator 27`1 3.7690e+10 1.1311e+10 1.2560e+11 < 2.2e-16 ***
3/`Enumerator 28`1 3.1004e+05 6.2501e+04 1.5380e+06 < 2.2e-16 ***
3/`Enumerator 29`1 6.7215e+09 2.9095e+09 1.5528e+10 < 2.2e-16 ***
3/`Enumerator 30`1 5.8288e-08 5.8288e-08 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 31`1 8.0640e-04 8.0640e-04 8.0000e-04 < 2.2e-16 ***
3/`Enumerator 32`1 5.7462e-08 5.7462e-08 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 33`1 1.0000e+00 NA NA NA 
3/`Enumerator 34`1 7.5521e-25 7.5521e-25 0.0000e+00 < 2.2e-16 ***
3/`Enumerator 73`1 1.7448e-01 3.2005e-02 9.5120e-01 0.0436088 * 
3/`Enumerator 75`1 1.1016e+09 3.2107e+08 3.7798e+09 < 2.2e-16 ***
3/`Enumerator 99`1 2.4493e+21 2.4493e+21 2.4493e+21 < 2.2e-16 ***
4/(Intercept) 4.6132e+03 2.6412e+03 8.0577e+03 < 2.2e-16 ***
4/SuspectDay1 2.6663e+01 1.0320e+01 6.8888e+01 1.205e-11 ***
4/Treated1 4.6132e+03 2.6412e+03 8.0577e+03 < 2.2e-16 ***
4/Female1 1.3316e+00 2.5857e-01 6.8576e+00 0.7320038 
4/UrbanSchool1 2.8929e-09 1.0159e-09 0.0000e+00 < 2.2e-16 ***
4/commune11 1.0000e+00 1.0000e+00 1.0000e+00 NA 
4/commune21 4.4186e-07 4.4186e-07 0.0000e+00 < 2.2e-16 ***
4/cpmmune31 1.5833e-01 3.4681e-02 7.2290e-01 0.0173669 * 
4/commune41 2.6284e+06 9.2015e+05 7.5081e+06 < 2.2e-16 ***
4/commune51 1.0012e+06 4.4614e+05 2.2466e+06 < 2.2e-16 ***
4/commune61 2.5058e-02 7.5633e-03 8.3000e-02 1.622e-09 ***
4/`Enumerator 1`1 3.8272e+01 1.8329e+01 7.9912e+01 < 2.2e-16 ***
4/`Enumerator 2`1 1.1520e+05 3.6422e+04 3.6438e+05 < 2.2e-16 ***
4/`Enumerator 3`1 3.3705e-09 3.3705e-09 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 4`1 1.7338e+02 3.5017e+01 8.5842e+02 2.670e-10 ***
4/`Enumerator 5`1 9.1842e-09 9.1842e-09 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 6`1 1.1390e+05 3.9839e+04 3.2566e+05 < 2.2e-16 ***
4/`Enumerator 7`1 1.0000e+00 1.0000e+00 1.0000e+00 NA 
4/`Enumerator 8`1 1.9051e+02 6.7073e+01 5.4111e+02 < 2.2e-16 ***
4/`Enumerator 9`1 5.1497e+02 2.0245e+02 1.3099e+03 < 2.2e-16 ***
4/`Enumerator 10`1 6.4045e+06 3.3337e+06 1.2304e+07 < 2.2e-16 ***
4/`Enumerator 11`1 7.1947e+10 3.9365e+10 1.3150e+11 < 2.2e-16 ***
4/`Enumerator 12`1 1.4842e+10 7.0657e+09 3.1175e+10 < 2.2e-16 ***
4/`Enumerator 13`1 4.6596e+08 1.6024e+08 1.3549e+09 < 2.2e-16 ***
4/`Enumerator 14`1 1.8174e-08 1.6943e-09 0.0000e+00 < 2.2e-16 ***
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
119
4/`Enumerator 15`1 1.3076e+07 5.4795e+06 3.1203e+07 < 2.2e-16 ***
4/`Enumerator 16`1 6.4912e+05 2.3703e+05 1.7776e+06 < 2.2e-16 ***
4/`Enumerator 17`1 3.3726e+06 1.6542e+06 6.8759e+06 < 2.2e-16 ***
4/`Enumerator 18`1 6.0416e+08 2.5187e+08 1.4492e+09 < 2.2e-16 ***
4/`Enumerator 19`1 1.3174e-14 1.3174e-14 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 20`1 8.5245e-21 8.5245e-21 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 21`1 1.2664e-32 1.2664e-32 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 22`1 1.3673e+05 3.3808e+04 5.5298e+05 < 2.2e-16 ***
4/`Enumerator 23`1 8.7356e+03 2.2687e+03 3.3636e+04 < 2.2e-16 ***
4/`Enumerator 24`1 3.7367e-14 3.7367e-14 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 25`1 2.3209e-08 1.7381e-09 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 26`1 2.0616e+03 4.8491e+02 8.7648e+03 < 2.2e-16 ***
4/`Enumerator 27`1 3.8741e+05 1.1626e+05 1.2910e+06 < 2.2e-16 ***
4/`Enumerator 28`1 2.1355e+01 4.3442e+00 1.0497e+02 0.0001647 ***
4/`Enumerator 29`1 1.5872e+05 6.8705e+04 3.6669e+05 < 2.2e-16 ***
4/`Enumerator 30`1 4.4099e+05 1.1230e+05 1.7318e+06 < 2.2e-16 ***
4/`Enumerator 31`1 1.3241e+15 1.3241e+15 1.3241e+15 < 2.2e-16 ***
4/`Enumerator 32`1 2.5591e+09 1.1243e+09 5.8250e+09 < 2.2e-16 ***
4/`Enumerator 33`1 1.0000e+00 1.0000e+00 1.0000e+00 NA 
4/`Enumerator 34`1 1.5737e-15 2.0356e-16 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 73`1 6.7913e-06 1.0213e-06 0.0000e+00 < 2.2e-16 ***
4/`Enumerator 75`1 2.7814e+04 9.1911e+03 8.4172e+04 < 2.2e-16 ***
4/`Enumerator 99`1 9.8332e-09 9.8332e-09 0.0000e+00 < 2.2e-16 ***
5/(Intercept) 5.5280e+05 3.3552e+05 9.1080e+05 < 2.2e-16 ***
5/SuspectDay1 4.6940e+01 1.9958e+01 1.1040e+02 < 2.2e-16 ***
5/Treated1 5.5280e+05 3.3552e+05 9.1080e+05 < 2.2e-16 ***
5/Female1 1.2099e+00 2.5004e-01 5.8548e+00 0.8127377 
5/UrbanSchool1 2.6835e-09 1.0060e-09 0.0000e+00 < 2.2e-16 ***
5/commune11 1.0000e+00 1.0000e+00 1.0000e+00 NA 
5/commune21 9.6352e+05 9.6352e+05 9.6352e+05 < 2.2e-16 ***
5/cpmmune31 6.5913e-04 1.7276e-04 2.5000e-03 < 2.2e-16 ***
5/commune41 4.3862e+03 1.6791e+03 1.1458e+04 < 2.2e-16 ***
5/commune51 2.9511e+03 1.3754e+03 6.3319e+03 < 2.2e-16 ***
5/commune61 6.7247e-05 2.3278e-05 2.0000e-04 < 2.2e-16 ***
5/`Enumerator 1`1 2.0367e+00 9.7542e-01 4.2527e+00 0.0582627 . 
5/`Enumerator 2`1 5.9960e+04 1.8957e+04 1.8965e+05 < 2.2e-16 ***
5/`Enumerator 3`1 1.0124e+07 1.0124e+07 1.0124e+07 < 2.2e-16 ***
5/`Enumerator 4`1 1.9358e+00 4.1025e-01 9.1345e+00 0.4040378 
5/`Enumerator 5`1 5.0310e+07 5.0310e+07 5.0310e+07 < 2.2e-16 ***
5/`Enumerator 6`1 4.6087e+03 1.8405e+03 1.1541e+04 < 2.2e-16 ***
5/`Enumerator 7`1 1.0000e+00 1.0000e+00 1.0000e+00 NA 
5/`Enumerator 8`1 1.2196e+01 5.0576e+00 2.9411e+01 2.561e-08 ***
5/`Enumerator 9`1 2.8864e-16 2.8864e-16 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 10`1 5.8910e+05 3.0664e+05 1.1318e+06 < 2.2e-16 ***
5/`Enumerator 11`1 1.9111e+08 1.0457e+08 3.4930e+08 < 2.2e-16 ***
5/`Enumerator 12`1 1.6814e+09 8.0046e+08 3.5318e+09 < 2.2e-16 ***
5/`Enumerator 13`1 1.3720e+06 4.7182e+05 3.9894e+06 < 2.2e-16 ***
5/`Enumerator 14`1 4.6120e-09 8.4916e-10 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 15`1 3.2834e+05 1.6361e+05 6.5892e+05 < 2.2e-16 ***
5/`Enumerator 16`1 6.2437e+04 3.0761e+04 1.2673e+05 < 2.2e-16 ***
5/`Enumerator 17`1 2.0831e+04 9.2284e+03 4.7024e+04 < 2.2e-16 ***
5/`Enumerator 18`1 2.4175e+07 1.0079e+07 5.7989e+07 < 2.2e-16 ***
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
120
5/`Enumerator 19`1 1.1580e+09 1.1580e+09 1.1580e+09 < 2.2e-16 ***
5/`Enumerator 20`1 2.8674e+08 1.4933e+08 5.5059e+08 < 2.2e-16 ***
5/`Enumerator 21`1 3.1509e-10 6.5613e-11 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 22`1 1.2695e+04 4.8773e+03 3.3042e+04 < 2.2e-16 ***
5/`Enumerator 23`1 1.9477e+03 8.3545e+02 4.5405e+03 < 2.2e-16 ***
5/`Enumerator 24`1 3.4840e+08 3.4840e+08 3.4840e+08 < 2.2e-16 ***
5/`Enumerator 25`1 4.1508e-09 5.6373e-10 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 26`1 3.7028e+02 1.3693e+02 1.0013e+03 < 2.2e-16 ***
5/`Enumerator 27`1 3.3598e-16 3.3598e-16 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 28`1 3.1540e-01 6.5119e-02 1.5276e+00 0.1516900 
5/`Enumerator 29`1 8.1034e-19 8.1034e-19 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 30`1 6.5132e+03 1.6585e+03 2.5577e+04 < 2.2e-16 ***
5/`Enumerator 31`1 1.5744e-10 1.5744e-10 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 32`1 3.1569e+07 1.3870e+07 7.1857e+07 < 2.2e-16 ***
5/`Enumerator 33`1 1.0000e+00 1.0000e+00 1.0000e+00 NA 
5/`Enumerator 34`1 1.7579e-16 3.9424e-17 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 73`1 7.3581e-07 1.9225e-07 0.0000e+00 < 2.2e-16 ***
5/`Enumerator 75`1 1.0952e+02 2.6312e+01 4.5589e+02 1.089e-10 ***
5/`Enumerator 99`1 4.1945e-13 4.1945e-13 0.0000e+00 < 2.2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Conclusion: Of the 5 tests, 3 found that ‘SuspectDay’ was significant; however, almost all variables in 
these tests were significant (interpretation is not reliable). Thus, there is little proof that sampling was 
performed the first days had an influence on the data collected.
EGRA TEST BIAS CHECK 
Model 1 
Dependent variable: s2_comp_SUM (oral comprehension)
t test of coefficients:
 Estimate Std. Error t value Pr(>|t|) 
(Intercept) 13.90216 0.98810 14.0696 < 2.2e-16 ***
SuspectDay1 0.88377 0.56019 1.5776 0.1148822 
Treated1 0.65844 0.81402 0.8089 0.4187277 
Female1 -0.13726 0.16400 -0.8370 0.4027527 
UrbanSchool1 1.18801 0.36239 3.2783 0.0010709 ** 
commune11 0.91127 1.17512 0.7755 0.4382013 
commune21 1.87300 0.90063 2.0797 0.0377451 * 
cpmmune31 0.20781 0.64391 0.3227 0.7469536 
commune41 -0.53313 0.81593 -0.6534 0.5136142 
commune51 0.82002 0.44685 1.8351 0.0667081 . 
`Enumerator 1`1 -6.43111 1.15533 -5.5665 3.131e-08 ***
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
121
`Enumerator 2`1 -2.20906 1.18257 -1.8680 0.0619747 . 
`Enumerator 3`1 -3.63363 1.24143 -2.9270 0.0034802 ** 
`Enumerator 4`1 -0.63419 0.75164 -0.8437 0.3989613 
`Enumerator 5`1 -3.69300 1.33951 -2.7570 0.0059123 ** 
`Enumerator 6`1 -4.83851 0.95562 -5.0632 4.692e-07 ***
`Enumerator 7`1 -2.57085 0.54775 -4.6935 2.959e-06 ***
`Enumerator 8`1 -2.21277 0.96262 -2.2987 0.0216741 * 
`Enumerator 9`1 -1.86007 0.66039 -2.8166 0.0049237 ** 
`Enumerator 10`1 -2.07285 0.89653 -2.3121 0.0209226 * 
`Enumerator 11`1 -3.06639 0.99265 -3.0891 0.0020485 ** 
`Enumerator 12`1 -3.34912 0.68072 -4.9200 9.720e-07 ***
`Enumerator 13`1 -2.14212 0.54598 -3.9234 9.169e-05 ***
`Enumerator 14`1 -3.10663 0.57966 -5.3594 9.807e-08 ***
`Enumerator 15`1 -1.88782 0.46988 -4.0177 6.201e-05 ***
`Enumerator 16`1 -3.31841 0.70620 -4.6989 2.882e-06 ***
`Enumerator 17`1 -3.10786 0.95876 -3.2415 0.0012179 ** 
`Enumerator 18`1 -3.44753 0.62242 -5.5389 3.653e-08 ***
`Enumerator 19`1 -3.09489 1.41503 -2.1872 0.0289021 * 
`Enumerator 20`1 -2.88938 0.66233 -4.3624 1.384e-05 ***
`Enumerator 21`1 -1.69173 0.88014 -1.9221 0.0548003 . 
`Enumerator 22`1 -4.32691 0.73417 -5.8936 4.765e-09 ***
`Enumerator 23`1 -4.44414 0.65260 -6.8099 1.466e-11 ***
`Enumerator 24`1 -11.88493 0.90953 -13.0671 < 2.2e-16 ***
`Enumerator 25`1 -5.73340 1.20546 -4.7562 2.184e-06 ***
`Enumerator 26`1 -3.54611 0.70547 -5.0266 5.662e-07 ***
`Enumerator 27`1 -2.09294 0.79352 -2.6375 0.0084471 ** 
`Enumerator 30`1 -2.99373 1.81626 -1.6483 0.0995240 . 
`Enumerator 31`1 -5.39556 1.57166 -3.4330 0.0006149 ***
`Enumerator 32`1 -4.24990 1.28122 -3.3171 0.0009338 ***
`Enumerator 33`1 -4.05531 1.15994 -3.4961 0.0004873 ***
`Enumerator 34`1 -3.04248 1.04281 -2.9176 0.0035859 ** 
`Enumerator 35`1 -2.25552 1.05741 -2.1331 0.0331000 * 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 2 
Dependent variable: s5_initial_SUM (letters names)
 Estimate Std. Error t value Pr(>|t|) 
(Intercept) 3.892802 1.275284 3.0525 0.0023137 ** 
SuspectDay1 -0.629739 0.519902 -1.2113 0.2260055 
Treated1 1.383699 0.882879 1.5673 0.1172888 
Female1 -0.070689 0.229152 -0.3085 0.7577630 
UrbanSchool1 0.207232 0.463999 0.4466 0.6552205 
commune11 2.240939 1.303231 1.7195 0.0857477 . 
commune21 0.147243 1.204229 0.1223 0.9027023 
cpmmune31 1.663487 0.815308 2.0403 0.0415128 * 
commune41 -2.224506 1.105734 -2.0118 0.0444403 * 
commune51 0.551612 0.620867 0.8885 0.3744550 
`Enumerator 1`1 -0.160773 1.199725 -0.1340 0.8934160 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
122
`Enumerator 2`1 0.025059 1.073084 0.0234 0.9813728 
`Enumerator 3`1 0.735805 1.244672 0.5912 0.5545099 
`Enumerator 4`1 0.924767 1.121661 0.8245 0.4098226 
`Enumerator 5`1 -2.122250 0.854293 -2.4842 0.0131040 * 
`Enumerator 6`1 -2.075423 0.942891 -2.2011 0.0278957 * 
`Enumerator 7`1 0.970701 0.856549 1.1333 0.2573021 
`Enumerator 8`1 -1.936167 0.888702 -2.1786 0.0295304 * 
`Enumerator 9`1 -0.676205 1.047691 -0.6454 0.5187621 
`Enumerator 10`1 -0.846384 1.321083 -0.6407 0.5218433 
`Enumerator 11`1 1.591705 1.222878 1.3016 0.1932732 
`Enumerator 12`1 -1.016320 1.221886 -0.8318 0.4056893 
`Enumerator 13`1 1.132671 0.897880 1.2615 0.2073482 
`Enumerator 14`1 0.651566 0.802610 0.8118 0.4170440 
`Enumerator 15`1 0.338454 1.099075 0.3079 0.7581720 
`Enumerator 16`1 -1.639729 0.941545 -1.7415 0.0818183 . 
`Enumerator 17`1 -0.552320 1.020463 -0.5412 0.5884280 
`Enumerator 18`1 -0.561884 0.980811 -0.5729 0.5668234 
`Enumerator 19`1 -3.537780 0.812688 -4.3532 1.443e-05 ***
`Enumerator 20`1 -1.443587 1.036077 -1.3933 0.1637523 
`Enumerator 21`1 -3.740045 1.090610 -3.4293 0.0006233 ***
`Enumerator 22`1 -3.243121 0.759448 -4.2704 2.088e-05 ***
`Enumerator 23`1 -1.183182 1.085933 -1.0896 0.2761040 
`Enumerator 24`1 -0.304840 1.139989 -0.2674 0.7891972 
`Enumerator 25`1 -5.525635 0.657999 -8.3976 < 2.2e-16 ***
`Enumerator 26`1 -0.090383 1.123029 -0.0805 0.9358664 
`Enumerator 27`1 1.316379 0.984347 1.3373 0.1813458 
`Enumerator 30`1 0.081197 1.593057 0.0510 0.9593573 
`Enumerator 31`1 -4.607252 1.425471 -3.2321 0.0012585 ** 
`Enumerator 32`1 -2.779682 1.151748 -2.4134 0.0159349 * 
`Enumerator 33`1 -3.707804 0.973809 -3.8075 0.0001467 ***
`Enumerator 34`1 -3.534264 0.994389 -3.5542 0.0003922 ***
`Enumerator 35`1 -4.429429 0.763947 -5.7981 8.341e-09 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 3 
Dependent variable: s8a_read_SUM (words in context)
t test of coefficients:
 Estimate Std. Error t value Pr(>|t|) 
(Intercept) 14.756851 3.430766 4.3013 1.820e-05 ***
SuspectDay1 -1.015291 1.303478 -0.7789 0.4361697 
Treated1 2.199840 2.117258 1.0390 0.2989887 
Female1 0.130863 0.515119 0.2540 0.7994999 
UrbanSchool1 0.095593 1.258195 0.0760 0.9394494 
commune11 4.934107 3.080250 1.6019 0.1094219 
commune21 -0.600096 2.421391 -0.2478 0.8043026 
cpmmune31 -0.071441 1.627953 -0.0439 0.9650032 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
123
commune41 -4.957515 2.741750 -1.8082 0.0708043 . 
commune51 3.382285 1.894977 1.7849 0.0745069 . 
`Enumerator 1`1 -9.899627 3.492823 -2.8343 0.0046613 ** 
`Enumerator 2`1 -8.072868 3.350794 -2.4092 0.0161188 * 
`Enumerator 3`1 -7.676446 4.169888 -1.8409 0.0658519 . 
`Enumerator 4`1 -6.971190 4.173911 -1.6702 0.0951151 . 
`Enumerator 5`1 -15.184071 1.889107 -8.0377 1.981e-15 ***
`Enumerator 6`1 -11.970951 2.686185 -4.4565 9.020e-06 ***
`Enumerator 7`1 -9.374520 2.640069 -3.5509 0.0003972 ***
`Enumerator 8`1 -14.393021 2.621371 -5.4906 4.778e-08 ***
`Enumerator 9`1 -13.860139 2.427472 -5.7097 1.390e-08 ***
`Enumerator 10`1 -10.841071 3.024970 -3.5839 0.0003506 ***
`Enumerator 11`1 -8.485190 3.272450 -2.5929 0.0096193 ** 
`Enumerator 12`1 -7.077972 2.635059 -2.6861 0.0073182 ** 
`Enumerator 13`1 -7.064810 2.817239 -2.5077 0.0122683 * 
`Enumerator 14`1 -8.804728 2.211394 -3.9815 7.212e-05 ***
`Enumerator 15`1 -9.278583 3.868546 -2.3985 0.0165987 * 
`Enumerator 16`1 -13.058036 2.350452 -5.5555 3.329e-08 ***
`Enumerator 17`1 -12.354703 2.923977 -4.2253 2.546e-05 ***
`Enumerator 18`1 -4.067975 4.525027 -0.8990 0.3688156 
`Enumerator 19`1 -11.597070 3.227933 -3.5927 0.0003390 ***
`Enumerator 20`1 -12.985156 2.694287 -4.8195 1.602e-06 ***
`Enumerator 21`1 -15.451675 2.176778 -7.0984 2.033e-12 ***
`Enumerator 22`1 -12.571624 2.737009 -4.5932 4.772e-06 ***
`Enumerator 23`1 -14.822108 2.439006 -6.0771 1.589e-09 ***
`Enumerator 24`1 -13.946455 2.261567 -6.1667 9.192e-10 ***
`Enumerator 25`1 -17.783107 2.120779 -8.3852 < 2.2e-16 ***
`Enumerator 26`1 -11.309141 2.508919 -4.5076 7.125e-06 ***
`Enumerator 27`1 -6.299550 3.003501 -2.0974 0.0361434 * 
`Enumerator 30`1 -3.335954 8.391929 -0.3975 0.6910474 
`Enumerator 31`1 -15.290893 4.029928 -3.7943 0.0001546 ***
`Enumerator 32`1 -12.318228 2.664161 -4.6237 4.131e-06 ***
`Enumerator 33`1 -13.792000 2.808966 -4.9100 1.022e-06 ***
`Enumerator 34`1 -14.707154 2.680259 -5.4872 4.870e-08 ***
`Enumerator 35`1 -17.320293 2.173474 -7.9689 3.376e-15 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 4 
Dependent variable: “s1_vocab_SUM” (vocabulary)
t test of coefficients:
 Estimate Std. Error t value Pr(>|t|) 
(Intercept) 8.3488639 0.7152931 11.6719 < 2.2e-16 ***
SuspectDay1 0.2935529 0.3612680 0.8126 0.4166120 
Treated1 0.9808847 0.5477761 1.7907 0.0735705 . 
Female1 0.0036326 0.0935183 0.0388 0.9690211 
UrbanSchool1 0.6845097 0.2445223 2.7994 0.0051930 ** 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
124
commune11 1.4588835 0.8055673 1.8110 0.0703627 . 
commune21 1.5499299 0.6774081 2.2880 0.0222897 * 
cpmmune31 -0.0144713 0.4500238 -0.0322 0.9743518 
commune41 -0.0235190 0.5398507 -0.0436 0.9652570 
commune51 0.9839253 0.2424836 4.0577 5.240e-05 ***
`Enumerator 1`1 -1.3236019 0.8099686 -1.6341 0.1024625 
`Enumerator 2`1 0.8016780 0.7004000 1.1446 0.2525775 
`Enumerator 3`1 0.6069129 0.7570059 0.8017 0.4228512 
`Enumerator 4`1 2.2222929 0.5131319 4.3308 1.595e-05 ***
`Enumerator 5`1 0.3783440 0.6439971 0.5875 0.5569706 
`Enumerator 6`1 0.7261370 0.3630701 2.0000 0.0457011 * 
`Enumerator 7`1 1.4860777 0.4368473 3.4018 0.0006889 ***
`Enumerator 8`1 1.7958444 0.4846239 3.7056 0.0002194 ***
`Enumerator 9`1 -0.2410620 0.4975463 -0.4845 0.6281085 
`Enumerator 10`1 1.1857225 0.5667240 2.0922 0.0366033 * 
`Enumerator 11`1 0.5187170 0.6581437 0.7882 0.4307463 
`Enumerator 12`1 1.1446516 0.3772581 3.0341 0.0024583 ** 
`Enumerator 13`1 1.5689139 0.3129753 5.0129 6.072e-07 ***
`Enumerator 14`1 0.7906671 0.4369907 1.8093 0.0706196 . 
`Enumerator 15`1 1.5313529 0.5494025 2.7873 0.0053893 ** 
`Enumerator 16`1 0.5527062 0.4575355 1.2080 0.2272559 
`Enumerator 17`1 1.0117620 0.5210098 1.9419 0.0523538 . 
`Enumerator 18`1 1.2126656 0.3671638 3.3028 0.0009823 ***
`Enumerator 19`1 1.0706636 0.5071520 2.1111 0.0349445 * 
`Enumerator 20`1 1.3657793 0.3972586 3.4380 0.0006038 ***
`Enumerator 21`1 0.4406971 0.5797954 0.7601 0.4473330 
`Enumerator 22`1 -0.1698804 0.5324478 -0.3191 0.7497337 
`Enumerator 23`1 0.1675107 0.3612763 0.4637 0.6429635 
`Enumerator 24`1 0.3925018 0.6430066 0.6104 0.5416887 
`Enumerator 25`1 0.0503872 0.7772891 0.0648 0.9483235 
`Enumerator 26`1 0.5860403 0.4224684 1.3872 0.1656153 
`Enumerator 27`1 1.3212287 0.4733314 2.7913 0.0053229 ** 
`Enumerator 30`1 0.2681305 1.3306765 0.2015 0.8403384 
`Enumerator 31`1 -1.0044973 0.7322181 -1.3719 0.1703362 
`Enumerator 32`1 0.3704312 0.8819957 0.4200 0.6745580 
`Enumerator 33`1 0.5591793 0.5942257 0.9410 0.3468619 
`Enumerator 34`1 0.3518568 0.7726001 0.4554 0.6488810 
`Enumerator 35`1 -0.2240954 0.5163728 -0.4340 0.6643724 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 5 
Dependent variable: “s3_letter_name_SUM”
t test of coefficients:
 
(Intercept) 23.063855 4.370864 5.2767 1.530e-07 ***
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
125
SuspectDay1 -1.858538 1.342427 -1.3845 0.1664459 
Treated1 6.417933 2.585247 2.4825 0.0131663 * 
Female1 0.064272 0.713375 0.0901 0.9282241 
UrbanSchool1 1.493297 1.532672 0.9743 0.3300772 
commune11 9.154932 4.337700 2.1105 0.0349944 * 
commune21 0.612493 3.307938 0.1852 0.8531324 
cpmmune31 2.336145 3.136809 0.7448 0.4565511 
commune41 -4.632406 3.317449 -1.3964 0.1628308 
commune51 4.492778 2.492324 1.8026 0.0716665 . 
`Enumerator 1`1 -12.398987 4.248240 -2.9186 0.0035741 ** 
`Enumerator 2`1 -11.913692 3.544698 -3.3610 0.0007983 ***
`Enumerator 3`1 -7.430752 5.304219 -1.4009 0.1614696 
`Enumerator 4`1 -7.520495 4.106112 -1.8315 0.0672406 . 
`Enumerator 5`1 -22.117710 2.175056 -10.1688 < 2.2e-16 ***
`Enumerator 6`1 -16.229327 3.199506 -5.0724 4.474e-07 ***
`Enumerator 7`1 -7.355904 3.525610 -2.0864 0.0371277 * 
`Enumerator 8`1 -16.694141 2.755106 -6.0593 1.770e-09 ***
`Enumerator 9`1 -17.614253 2.682240 -6.5670 7.304e-11 ***
`Enumerator 10`1 -19.135273 4.068039 -4.7038 2.815e-06 ***
`Enumerator 11`1 -10.896625 4.171891 -2.6119 0.0091035 ** 
`Enumerator 12`1 -13.195669 2.916368 -4.5247 6.579e-06 ***
`Enumerator 13`1 1.209189 3.648779 0.3314 0.7403972 
`Enumerator 14`1 -1.673040 2.344747 -0.7135 0.4756430 
`Enumerator 15`1 -2.501691 5.160244 -0.4848 0.6278961 
`Enumerator 16`1 -13.695904 2.854149 -4.7986 1.775e-06 ***
`Enumerator 17`1 -12.107860 3.804049 -3.1829 0.0014914 ** 
`Enumerator 18`1 -3.859019 3.920838 -0.9842 0.3251770 
`Enumerator 19`1 -8.140101 3.768039 -2.1603 0.0309247 * 
`Enumerator 20`1 -13.148041 3.579286 -3.6734 0.0002487 ***
`Enumerator 21`1 -10.266254 3.214648 -3.1936 0.0014376 ** 
`Enumerator 22`1 -11.506495 4.897583 -2.3494 0.0189459 * 
`Enumerator 23`1 -13.445673 2.845357 -4.7255 2.535e-06 ***
`Enumerator 24`1 -18.367550 2.993559 -6.1357 1.112e-09 ***
`Enumerator 25`1 -20.246401 3.172686 -6.3815 2.405e-10 ***
`Enumerator 26`1 -16.142684 2.953465 -5.4657 5.485e-08 ***
`Enumerator 27`1 -12.897966 3.235065 -3.9869 7.052e-05 ***
`Enumerator 30`1 -3.692135 9.468903 -0.3899 0.6966556 
`Enumerator 31`1 -22.565893 2.367039 -9.5334 < 2.2e-16 ***
`Enumerator 32`1 -14.198837 4.340197 -3.2715 0.0010969 ** 
`Enumerator 33`1 -19.613808 3.163487 -6.2001 7.485e-10 ***
`Enumerator 34`1 -16.519419 3.900419 -4.2353 2.437e-05 ***
`Enumerator 35`1 -24.540514 2.173822 -11.2891 < 2.2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
 
Model 6 
Dependent variable: “s4_letter_sound_SUM”
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
126
t test of coefficients:
 Estimate Std. Error t value Pr(>|t|) 
(Intercept) 12.34456 4.07310 3.0307 0.002486 ** 
SuspectDay1 -1.80126 1.58094 -1.1394 0.254754 
Treated1 4.10244 2.20175 1.8633 0.062641 . 
Female1 -0.37687 0.59081 -0.6379 0.523658 
UrbanSchool1 -0.16928 1.43256 -0.1182 0.905956 
commune11 5.55886 3.93987 1.4109 0.158497 
commune21 -0.77054 3.10715 -0.2480 0.804181 
cpmmune31 0.55733 2.20492 0.2528 0.800486 
commune41 -6.93034 3.06399 -2.2619 0.023864 * 
commune51 2.41071 2.25357 1.0697 0.284933 
`Enumerator 1`1 -0.81282 3.46373 -0.2347 0.814503 
`Enumerator 2`1 0.82705 3.55993 0.2323 0.816323 
`Enumerator 3`1 0.60008 3.88465 0.1545 0.877259 
`Enumerator 4`1 5.91420 4.32868 1.3663 0.172078 
`Enumerator 5`1 -7.85466 2.48943 -3.1552 0.001639 ** 
`Enumerator 6`1 -4.32079 2.98845 -1.4458 0.148457 
`Enumerator 7`1 -2.75683 3.40538 -0.8096 0.418341 
`Enumerator 8`1 -6.75770 3.05242 -2.2139 0.027003 * 
`Enumerator 9`1 -6.26101 2.99414 -2.0911 0.036707 * 
`Enumerator 10`1 -3.91319 4.46517 -0.8764 0.380979 
`Enumerator 11`1 4.02825 3.55784 1.1322 0.257745 
`Enumerator 12`1 -0.99942 2.63832 -0.3788 0.704890 
`Enumerator 13`1 2.90841 2.53469 1.1474 0.251402 
`Enumerator 14`1 0.47311 2.56699 0.1843 0.853802 
`Enumerator 15`1 -0.87601 4.05956 -0.2158 0.829184 
`Enumerator 16`1 -6.64838 2.68536 -2.4758 0.013416 * 
`Enumerator 17`1 -6.28647 3.27694 -1.9184 0.055272 . 
`Enumerator 18`1 7.87158 4.99450 1.5761 0.115248 
`Enumerator 19`1 -3.61647 2.84579 -1.2708 0.204013 
`Enumerator 20`1 -2.61773 3.46096 -0.7564 0.449565 
`Enumerator 21`1 -7.91169 2.55334 -3.0986 0.001985 ** 
`Enumerator 22`1 -6.10749 3.38796 -1.8027 0.071657 . 
`Enumerator 23`1 -4.86471 2.95592 -1.6458 0.100048 
`Enumerator 24`1 -3.43001 3.47777 -0.9863 0.324180 
`Enumerator 25`1 -9.42406 3.01189 -3.1290 0.001792 ** 
`Enumerator 26`1 -2.68155 2.76147 -0.9711 0.331693 
`Enumerator 27`1 -0.92783 2.90879 -0.3190 0.749795 
`Enumerator 30`1 3.59482 7.35504 0.4888 0.625094 
`Enumerator 31`1 -11.04406 4.01585 -2.7501 0.006037 ** 
`Enumerator 32`1 -8.35954 2.78940 -2.9969 0.002777 ** 
`Enumerator 33`1 -5.98219 3.51074 -1.7040 0.088617 . 
`Enumerator 34`1 -4.39027 4.42429 -0.9923 0.321224 
`Enumerator 35`1 -11.27085 2.40326 -4.6898 3.011e-06 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Model 7 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
127
Dependent variable: Indicator (65% of boy students who, by the end of two grades can read and 
understand the meaning of grade level text, based on EGRA sections 8, 9 and 10) (is a binary variable of 
1=pass, 0=fail)
z test of coefficients:
 Estimate Std. Error z value Pr(>|z|) 
(Intercept) -1.539557 1.195297 -1.2880 0.197742 
SuspectDay1 -0.421996 0.509779 -0.8278 0.407783 
Treated1 0.687357 0.994352 0.6913 0.489402 
Female1 -0.048225 0.199660 -0.2415 0.809142 
UrbanSchool1 0.242401 0.413772 0.5858 0.557988 
commune11 1.516437 1.205515 1.2579 0.208422 
commune21 0.010083 1.084771 0.0093 0.992584 
cpmmune31 -0.057198 0.489009 -0.1170 0.906886 
commune41 -1.611754 0.857674 -1.8792 0.060215 . 
commune51 0.633911 0.504009 1.2577 0.208487 
`Enumerator 1`1 -1.750809 1.400189 -1.2504 0.211150 
`Enumerator 2`1 0.133456 1.011498 0.1319 0.895033 
`Enumerator 3`1 -15.815768 0.902150 -17.5312 < 2.2e-16 ***
`Enumerator 4`1 -0.217049 0.992704 -0.2186 0.826927 
`Enumerator 5`1 -1.667877 0.639405 -2.6085 0.009094 ** 
`Enumerator 6`1 -1.765976 1.111447 -1.5889 0.112083 
`Enumerator 7`1 -0.414671 0.625192 -0.6633 0.507158 
`Enumerator 8`1 -2.246486 1.094933 -2.0517 0.040198 * 
`Enumerator 9`1 -2.870425 1.130264 -2.5396 0.011098 * 
`Enumerator 10`1 -1.692624 0.847060 -1.9982 0.045691 * 
`Enumerator 11`1 -0.523170 0.823547 -0.6353 0.525256 
`Enumerator 12`1 -0.511219 0.604347 -0.8459 0.397607 
`Enumerator 13`1 -0.313542 0.624944 -0.5017 0.615870 
`Enumerator 14`1 -0.698869 0.638498 -1.0946 0.273713 
`Enumerator 15`1 -0.341635 0.770600 -0.4433 0.657523 
`Enumerator 16`1 -2.539632 0.900374 -2.8206 0.004793 ** 
`Enumerator 17`1 -0.931992 1.090859 -0.8544 0.392902 
`Enumerator 18`1 0.425581 0.627216 0.6785 0.497440 
`Enumerator 19`1 -0.935734 0.870572 -1.0748 0.282442 
`Enumerator 20`1 -1.735813 0.860562 -2.0171 0.043688 * 
`Enumerator 21`1 -18.264736 0.848255 -21.5321 < 2.2e-16 ***
`Enumerator 22`1 -1.081488 0.689656 -1.5682 0.116845 
`Enumerator 23`1 -2.447330 1.163032 -2.1043 0.035355 * 
`Enumerator 24`1 -1.276504 0.815084 -1.5661 0.117325 
`Enumerator 25`1 -18.215383 0.773266 -23.5564 < 2.2e-16 ***
`Enumerator 26`1 -0.793533 0.754356 -1.0519 0.292829 
`Enumerator 27`1 0.084176 0.738801 0.1139 0.909288 
`Enumerator 30`1 0.917537 1.305387 0.7029 0.482127 
`Enumerator 31`1 -1.453022 1.360084 -1.0683 0.285370 
`Enumerator 32`1 -1.708470 1.126150 -1.5171 0.129244 
`Enumerator 33`1 -1.433385 0.909631 -1.5758 0.115075 
`Enumerator 34`1 -1.287536 0.905459 -1.4220 0.155035 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 23 – CHECKING FOR BIAS IN DATA COLLECTION (CONTINUED) 
128
`Enumerator 35`1 -3.355790 1.182424 -2.8381 0.004539 ** 
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
 
Conclusion: Of the 7 tests, none found that ‘SuspectDay’ was significant. This, it can be said that how 
sampling was performed the first days had little and probably no influence on the data collected.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
129
APPENDIX 24 – INDICATOR CALCULATIONS 
Indicator Definition Source Description Calculation Comments 
Improved 
Literacy: % boy 
students who, by 
the end of two 
grades of can 
read and 
understand the 
meaning of 
grade level text.
% girl students 
who, by the end
of two grades of 
can read and 
understand the 
meaning of 
grade level text.
The national education 
curriculum defines 
reading comprehension 
standards for each grade. 
The project will use these 
standards to design an 
early-grade assessment 
to test reading levels.
Unit of Measure: percent 
grade 2 boys / girls (a 
statistical sample that is 
representative of that 
population is adequate)
EGRA 
(sections 8, 9 
and 10)
Improved literacy 
defined as obtaining 
2 out of 5 points 
‘PASS’ (less than 
50%) on the reading 
and comprehension 
parts (sections 8, 9 
and 10) of the test 
(same as Mid-term). 
Usually, 3 points out 
of 5 is considered to 
be the lowest 
acceptable 
performance to be 
considered 
“competent” in any 
academic 
performance in 
Benin, so the 
selection of 2 points 
out of 5 is very 
conservative. 
𝐼1,2 =
∑ 𝐴𝑠𝑖𝑗 𝐸𝐺𝑅𝐴 ∗ 𝑝𝑎𝑠𝑠𝑖𝑗
𝑁𝑠
𝑖
∑ 𝐴𝑠𝑖𝑗 𝐸𝐺𝑅𝐴
𝑁𝑠
𝑖
⁄
𝑁𝑠 = Total sample size across all 
schools
𝑝𝑎𝑠𝑠𝑖𝑗= is a binary variable {0,1} 
taking the value one if student i 
obtained 2 out of 5 points ‘PASS’
*Evaluate girls and boys in the 
same way, just separating the 
sample
Not calculated in 
baseline. Advisem 
conducted 
baseline 
calculations. 
However, 
baseline 
calculation could 
not be weighted 
due to lack of 
class size data for 
all schools.
Improved 
Attentiveness: % 
of students in 
target schools 
who are 
identified as 
attentive during 
class / 
instruction
DEFINITION: A 
statistically 
representative sample of 
target Students observed
for attentiveness
Unit of Measure: Percent
Classroom 
Observation 
Percentage of 
students identified 
as attentive divided 
by all students 
observed. 
Will be 
disaggregated by 
gender.
𝐼3,4
=
∑ 𝐴𝑠𝑖𝑗 𝑐𝑙 ∗ 𝑎𝑡𝑡𝑖𝑗
𝑁𝑠
𝑖
∑ 𝐴𝑠𝑖𝑗 𝑐𝑙
𝑁𝑠
𝑖
⁄
𝑁𝑠 = Total sample size across all 
schools
𝑎𝑡𝑡𝑖𝑗= is a binary variable {0,1} 
taking the value one if student 
was attentive and 0 if the student 
was inattentive
*Evaluate girls and boys in the 
same way, just separating the 
sample
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
130
Indicator Definition Source Description Calculation Comments 
Reduced Short￾Term Hunger: % 
of parents in 
target schools 
indicate that 
their children 
were “hungry" 
during the 
school day 
DEFINITION: Percent of 
parents whose children 
(taken from a statistically 
representative sample) 
state they were "hungry" 
during the school day.
Unit of Measure: Percent 
(one parent is counted as 
many times as the 
number of his/her 
hungry children)
Parent Survey 
(section 108; 
Q 7)
Percent of parents 
who indicate that a 
child was hungry at 
school the week 
before.
One parent counted 
as many times for 
the number of their 
hungry children
𝐼5 =
𝑃𝑎
𝑃𝑜
∗ 100
𝑃𝑎= # parents whose children 
were hungry
𝑃𝑜= # questioned parents 
𝐼5= % of parents whose children 
were hungry
Baseline 
incorrectly 
derived one value 
for each parent (a 
binomial indicator 
of responded that 
any of his or her 
children were 
hungry during the 
school day last 
week. Not 
possible to 
correctly answer 
indicator from 
baseline as only 
one value was 
obtained (not for 
each child).
Reduced Health￾Related 
Absences: % of 
students who 
report a 
decrease in 
health-related 
absences
DEFINTION: Statistically 
representative sample of 
students who self-report 
on health-related 
absences 
Unit of Measure: Percent
(number of students 
absent due to 
illness/total number of 
absent student)
Teacher 
Survey (Qs 
206-209)
This will be 
calculated by 
determining the 
percentage of 
students absent due 
to illness divided by 
the total number of 
absent students and 
the total number of 
students in the 
classroom (to have 
comparable results 
with the baseline).
𝐼6 = ((
𝐻𝑎𝑖
𝐻𝑎
) ÷ 𝑇) ∗ 100
𝐼6= % of health-related absences 
in school j
𝐻𝑎𝑖= # students absent due to 
illness in a class
𝐻𝑎= Total # absent students (or 
total # of students in class) in a 
class
𝑇 = Total # of teachers 
interviewed
Not calculated 
correctly in 
baseline (number 
of students in 
class c of school j 
who had health 
related absences 
divided by is the 
total number of 
students in class c 
of school j). 
However, to have 
comparable 
results the 
endline will 
calculate this 
way.
Increased 
Community 
Understanding 
of Benefits of 
Education: % of 
parents in target 
communities 
who can name 
at least three 
benefits of 
primary 
education
DEFINITION: Percent of 
parents who can name at 
least 3 benefits of 
schooling (socio￾economic, 
personal/socialization of 
children, society gains, 
etc.)
Unit of Measure: 
Percent, disaggregated 
by sex, target community
Parent Survey 
(Q 118)
Each parent will be 
classified as having 
met the goal (by 
naming three or 
more reasons), or 
not (by naming less 
than three reasons). 
Other responses 
were checked and 
removed only if they 
did not answer the 
question.
Disaggregation by 
sex.
𝐼7 =
𝑃𝑎
𝑃
∗ 100
𝐼7= % of parents in who can name 
these benefits
𝑃𝑎= # of parents who can name 
these benefits
𝑃 =all parents surveyed
Baseline also 
include a negative 
answer if the 
respondent said 
that his or her 
child’s education 
is not important. 
However, this 
question is not 
utilized in the 
endline.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
131
Indicator Definition Source Description Calculation Comments 
Increased 
Capacity of 
Government 
Institutions: 
Students in 144 
schools assessed 
using Early 
Grade Reading 
Assessment tool
DEFINITION: Number of 
classrooms in target 
schools where children's 
literacy levels are 
assessed by INFRE￾trained enumerators 
using Early Grade 
Literacy Assessment 
Instrument.
Unit of Measure: 
Number
EGRA Will be derived 
through program 
records.
Will be derived through program 
records.
Increased Use of 
Health and 
Dietary 
Practices: % of 
students in 
schools 
receiving USDA 
assistance 
receiving a 
minimum 
acceptable diet
DEFINITION: Percent of 
statistically 
representative sample of 
students (male/female) 
receiving a minimum 
acceptable diet from 
USDA supported canteen 
activities and family 
meals.
Dietary diversity is 
defined as the number of 
individual food items or 
food groups consumed 
over a given period of 
time. It is measured by 
counting the number of 
food groups rather than 
food Items consumed. At 
the individual bevel it 
reflects dietary quality 
(mainly micronutrient 
adequacy of the diet).
Unit of Measure: Percent
Parent Survey 
(Section 108, 
Qs 9, 11-17)
Foods were 
classified into the 
seven food groups 
as specified by 
USDA. Parents will 
also be asked how 
often their children 
ate the day before. 
Any answer greater 
than 7 times will be 
classified as 7 times. 
A minimum 
acceptable diet is 
defined as eating 4 
or more of these 
food groups, and a 
minimum frequency 
of 3 times a day.
𝐼9 =
𝐷𝑎
𝐷
∗ 100
𝐼9= % of students in school j with 
minimum acceptable diet
𝐷𝑎= # of parents in who testify 
their children eat 4+ food groups 
and 3 times a day 
𝐷 = # of parents surveyed (parent 
needs to be counted as many 
times as children they have)
Note that 
baseline results 
provided only 
contains one 
value per parent. 
Final evaluation 
correctly focuses 
on each school 
child of the 
parent.
Percent of 
schools with 
soap and water 
at hand washing
stations 
commonly used 
by students
DEFINITION: This 
indicator allows to verify 
the application of one of 
the three key hygiene 
practices 
Observation:
-Presence of water at 
hand
washing station
-Presence of soap or ash
-Signs of use
Unit of measure: Percent
Observational 
Tool
(Q: 103, 104 
and 105)
To be considered a 
‘yes’; 50% or more 
of hand washing 
stations must have 
soap and water and 
the enumerator 
must provide the 
answer ‘yes, very 
much’ to whether 
there is evidence 
that students are 
using the stations.
𝐼10𝑟 =
𝑁10𝑟
𝑁𝑟
∗ 100
.𝐼10𝑢 =
𝑁10𝑢
𝑁𝑢
∗ 100
𝐼10 = (
41
48 ∗ 𝐼10𝑟) +
 (
7
48
∗ 𝐼10𝑢)
𝐼10𝑟= % of rural schools with soap 
and water at hand washing.
𝑁10𝑟= # rural schools with soap 
and water at hand washing.
𝑁𝑟= total # of rural schools
First time being 
measured.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
132
Indicator Definition Source Description Calculation Comments 
𝐼10𝑢= % of urban schools with 
soap and water at hand washing.
𝑁10𝑢= # urban schools with soap 
and water at hand washing.
𝑁𝑢= total # of urban schools
𝐼10= Percent of schools with soap 
and water at hand washing
Increased 
capacity of 
Government 
Institutions:
Number of 
government 
staff in relevant 
ministries/office
s implicated in 
canteen 
/commodity 
management 
training
DEFINITION; Relevant 
government staff with an 
active role in training 
canteen committees in 
proper handling of 
commodities
Unit of Measure: 
Number
Project 
reports. If 
possible, 
confirmed 
through KIIs.
Confirmation will be 
determined through 
questioning in KIIs. 
However, will be 
dependent on the 
ability of 
consultants to meet 
with these 
individuals.
Project reports.
Improved 
Quality of 
Literacy 
Instruction: % of 
teachers who 
devote at least 
an average of 45 
minutes a day to 
literacy 
instruction (in 
the PMP - 45 
minutes twice a 
day to literacy)
DEFINITION: Measure 
the number of teachers 
devoting time and 
attention to daily literacy 
instruction
Unit of Measure: 
Number of teachers
Project 
records as 
PMP states 
Project M&E 
& literacy 
staff in 
collaboration 
with DIP 
School 
Inspectors & 
INFRE 
pedagogical 
experts are 
responsible 
for data 
collection.
Will be 
confirmed 
through: 
Teacher 
Survey (Qs 
315 and 316)
Teachers were 
asked to report how 
long their last 
literacy instruction 
period was and how 
many times per 
week they teach 
literacy.
The indicator relates 
to 225 minutes of 
literacy instruction a 
week.
Baseline: had 
passed the target of
70%.
𝐼12 =
𝑇𝑎
𝑇
∗ 100
𝐼12= % teachers average of 45 
minutes a day in school j
𝑇𝑎= # of teachers average of 45 
minutes a day
𝑇= total # teachers surveyed 
Percentage is a 
more appropriate 
estimation as the 
consultants will 
not know the 
global population 
of teachers.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
133
Indicator Definition Source Description Calculation Comments 
Improved 
Literacy 
Instructional 
Materials: 
Number of 
teachers 
correctly using 
the national 
literacy 
curriculum and 
the related 
instructional 
materials
Number of teachers of 1st
and 2nd grades using the 
national literacy 
curriculum.
Will be reported as a 
percentage.
Project 
records as 
PMP states 
Project M&E 
& literacy 
staff in 
collaboration 
with DIP 
School 
Inspectors & 
INFRE 
pedagogical 
experts are 
responsible 
for data 
collection.
Will be 
confirmed 
through: 
Teacher 
Survey (Q 
314)
Or 
Principal 
Survey (Q 
213)
Teachers will have 
to respond ‘yes’ to 
this question to be a 
positive answer, not 
‘somewhat’ or ‘no’. 
Principals have to 
respond ‘yes’ to this 
question to be a 
positive answer, not 
‘no’. 
𝐼13 =
𝑇𝑐
𝑇
∗ 100
𝑇𝑐 = Number of teachers (1st and 
2
nd grade) correctly using the 
national literacy curriculum and 
the related instructional 
𝑇 = total # teachers (1st and 2nd
grade) surveyed 
𝐼13 =% of 1st and 2nd grade 
teachers correctly using the 
national literacy curriculum and 
the related instructional
Target is assumed 
to be 144 
teachers in each 
of the two 
grades.
Improved 
Average Teacher 
Attendance Rate 
Average teacher 
attendance rate 
for each school 
DEFINITION: Teacher 
attendance based on 
sign-in registers kept by
the director at each 
school, Unit is the 
number of days present 
in the school year. The 
rate is the number of 
days present in the
school year divided by 
total number of school 
days in the school year, 
and then the average
attendance per school 
and for all schools in the 
program.
Principal 
Survey (Qs 
603 to 608)
Done by trimester 
like in the baseline. 
Had to use the 
September –
December 2017 
trimester due to the 
strike. 
𝑟𝑗𝑡 =
67 − 𝑎𝑗𝑡
67
𝑟𝑗 =
∑ 𝑟𝑗𝑡
𝑛𝑗𝑡
𝑡=1
𝑛𝑗𝑡
𝐼14 =
𝑟𝑗
𝑃
∗ 100
The principal reported the 
number of days ( ) teacher t was 
absent. The reported number of 
days was subtracted from the 
total number of days last 
trimester (67) and divided by the 
total number of days to achieve 
the rate reported 𝒓𝒋𝒕.
Teacher attendance rates (𝒓𝒋) 
were then averaged by school by 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
134
Indicator Definition Source Description Calculation Comments 
summing all the rates and dividing 
by the number of teachers 
surveyed in school (𝒏𝒋𝒕). 
𝑃 = number of principals surveyed
𝐼14 = average teacher attendance 
rate (reported as a %)
Increased Skills 
and 
Knowledge of 
Teachers: 
Number of 
teachers in 
target schools 
who 
demonstrate 
use of new and 
quality teaching 
techniques or 
tools.
Will focus on CI 
and CP teachers.
Number of teachers 
/teaching assistants who 
are using improved 
techniques and tools in 
their classrooms as a 
result of USDA 
assistance.
Successful application 
requires that teachers, 
educators, teaching 
assistance have 
incorporated the learned 
methods into their 
curriculum and are 
actively applying these 
methods in their daily 
classroom Instruction.
Unit of Measure: number
Will be reported as a 
percentage, because we 
do not know total 
population.
Project 
records as 
PMP states 
Project and 
INFRE trained 
and 
supervised 
enumerators 
and field 
managers are 
responsible 
for data 
collection
Will be 
confirmed 
through: 
Teacher 
Survey. Self 
reflection of 
teachers (Q 
313).
Verified 
through
Principal 
Survey (Q 
210)
Teachers will have 
to respond ‘yes’ to 
this question to be a 
positive answer, not 
‘somewhat’ or ‘no’.
Principals will have 
to respond ‘yes’ to 
this question to be a 
positive answer, not 
‘somewhat’ or ‘no’.
𝐼15 =
𝑇𝑑
𝑇
∗ 100
𝑇𝑑 = Number of teachers (1st and 
2
nd grade) who demonstrate use 
of new and quality teaching 
techniques or tools
𝑇 = total # teachers (1st and 2nd
grade) surveyed 
𝐼13 =% of 1st and 2nd grade 
teachers who demonstrate use of 
new and quality teaching 
techniques or tools
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 24 – INDICATOR CALCULATIONS (CONTINUED) 
135
Indicator Definition Source Description Calculation Comments 
Increased Skills 
and 
Knowledge of 
School 
Administrators: 
Number of 
school 
administrators 
in targeted 
schools who 
demonstrate 
use of new 
techniques or 
tools
DEFINITION: Number of 
school administrators 
who are applying the 
new knowledge and skills 
received in USDA 
supported training and 
certification programs.
School administrators 
should demonstrate the 
use of at least One new 
technique or technology 
in their standard 
practices or quality 
assurance of instruction,
Unit of measure: number
Project 
records as 
PMP states 
INFRE, WEI 
Training 
Coordinator, 
NGO field 
agents are 
responsible 
for data 
collection
Will be 
confirmed 
through:
Director 
Survey. Self 
reflection of 
directors (Q
113).
Principals will have 
to respond ‘yes’ to 
this question to be a 
positive answer, not 
‘somewhat’ or ‘no’. 
A ‘please give an 
example’ question 
has been added so 
principals provide 
proof.
𝐼16 =
𝑃𝑑
𝑃
∗ 100
𝑃𝑑 = Number of principals who 
demonstrate use of new and 
quality teaching techniques or 
tools
P = total # principals surveyed 
𝐼16 =% of principals who 
demonstrate use of new and 
quality teaching techniques or 
tools
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
136
APPENDIX 25 – FIELD MISSION SCHEDULE 
Below is the schedule for the evaluation team’s field visit in Benin.
Day# Date Qualitative Qualitative 
Consultant Activity Location Consultant Activity Location 
00 Feb Sunday 25 Hussein Inception 
report and 
evaluation 
discussions
Cotonou Satoguina Inception 
report and 
evaluation 
discussions
Cotonou
00 Mar Wed 7 Hussein Departure Cotonou
1 June Saturday 2 Hussein Departure for 
Cotonou
Satoguina, 
Issaou Gado
2 Sunday 3 Hussein Departure 
Cotonou￾Kandi
Satoguina, 
Issaou Gado
3 Monday 4 Hussein Training of 
Facilitators
Kandi Satoguina, 
Gado, 
Training of 
Enumerators
Kandi
4 Tuesday 5 Hussein Conducting 
Exercises
Soukarou/A, 
Gogonou, Kalale
Satoguina, 
Gado. Grace 
Arrives 
Cotonou, 
bring tablets
Training of 
Enumerators
Kandi
5 Wednesday 6 Hussein Conducting 
Exercises
Donwari A, 
Kandi, Alibori, 
Satoguina, 
Gado. Grace 
arrives Kandi
Training of 
Enumerators
Kandi
6 Thursday 7 Hussein Conducting 
Exercises
Kidakperou, 
Bourgou-Kalale
Satoguina, 
Gado, and 
Grace
Training of 
Enumerators
Kandi
7 Friday 8 Hussein Conducting 
Exercises
Tassi-Tedji C
Mallenville
Satoguina 
and Grace
Data Gathering 
in Kandi, 
Debriefing, 
cleaning Data, 
addressing 
challenges /
Kandi
8 Saturday 9 Hussein Transcribing 
Exercises
Kandi Satoguina 
and Grace
Debriefing, 
cleaning Data, 
addressing 
challenges / 
Data 
Gathering. 
9 Sunday 10 Hussein Transcribing 
Exercises
Kandi Satoguina 
and Grace
Preparations 
for Field Work
10 Monday 11 Hussein KIIs Kandi Satoguina 
and Grace
Data Gathering All 
regions
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 25 – FIELD MISSION SCHEDULE (CONTINUED) 
137
Day# Date Qualitative Qualitative 
Consultant Activity Location Consultant Activity Location 
11 Tuesday 12 Hussein KIIs/Travel to 
Cotonou
Parakou/Kandi Satoguina 
and Grace
Data Gathering All 
regions
12 Wednesday 13 Hussein KIIs Cotonou/Porto 
Novo
Satoguina 
and Grace
Data Gathering All 
regions
13 Thursday 14 Hussein KIIs Cotonou/Porto 
Novo. Debriefing 
with CRS/Travel 
out
Satoguina 
and Grace
Data Gathering All 
regions
14 June Friday 15 Hussein Travel Satoguina 
and Grace
Travelling to 
Cotonou
15 Saturday 16 Satoguina 
and Grace
Debriefing, 
cleaning Data, 
addressing
Cotonou
16 Sunday 17 Satoguina 
and Grace
Debriefing, 
cleaning Data, 
addressing
Cotonou
17 Monday 18 Grace travels 
out. 
18 Wednesday 27 Satoguina Travels 
overnight to 
Kandi
19 Thursday 28 Satoguina Data Gathering All 
regions
20 Friday 29 Satoguina Data Gathering All 
regions
21 Saturday 30 Satoguina Travel Kandi￾Cotonou
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
138
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK 
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
MGD SO 1: Improved Literacy of School-Age Children 
 
• 65% of boy students who, by the end of two grades can read 
and understand the meaning of grade level text
EGRA Yes Evaluation 
team
• 22,621 males benefiting directly from USDA-funded 
interventions
Project records No CRS
• 60% of girl students who, by the end of two grades, can read 
and understand the meaning of grade-level text
EGRA Yes Evaluation 
team
• 22,879 females benefiting directly from USDA-funded 
interventions
Project records No CRS Will be verified 
through FGDs 
with students
• 220,783 individuals benefiting indirectly from USDA-funded 
interventions
Project records No CRS Will be verified 
through FGDs 
with students
1.1 IMPROVED 
QUALITY OF 
LITERACY 
INSTRUCTION 
70% of teachers who devote at least an average 
of 45 minutes a day to literacy instruction
Program 
observation, 
interviews, sites 
visits. Who: 
Project M&E 
and literacy 
staff with DIP 
School 
Inspectors and 
WERE 
pedagogical 
experts
Yes, but 
should also 
be validated 
through 
means in 
PMP
CRS / 
Evaluation 
team
Will be verified 
through Teacher 
Survey.
1.1.1 More 
consistent 
teacher 
attendance
75% average teacher 
attendance rate for each 
school and aggregated by 
district
Teacher 
attendance 
registers.
No (End of 
Program)
CRS Will be verified 
through Director 
Survey and FGDs 
with teachers, 
students, and 
KIIs with school 
administrators. 
1.1.2 Better 
access to school 
supplies and 
materials
4,230 textbooks and other 
teaching and learning 
materials provided as a 
result of USDA assistance
Project records No CRS
1.1.3 Improved 
literacy 
instructional 
materials
144 teachers using the 
national literacy curriculum 
and the related instructional 
materials
Project records No CRS / 
Evaluation 
team
Will be verified 
through Teacher 
Survey.
1.1.4 Increased 
skills and 
144 
teachers/educators/teaching 
assistants trained or 
Project training 
records and 
reports 
No CRS Perceived quality 
of training will 
be assessed 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK (CONTINUED) 
139
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
knowledge of 
teachers
certified as a result of USDA 
assistance
through FGDs 
with teachers.
144 teachers in target 
schools who demonstrate 
use of new and quality 
teaching techniques or tools
Program 
observation, 
interviews, sites 
visits
Yes, but 
should also 
be validated 
through 
means in 
PMP
CRS / 
Evaluation 
team
Will be verified 
through Teacher 
Survey. Self 
reflection of 
teachers, FGDs 
with Teachers 
and classroom 
observation.
1.1.5 Increased 
skills and 
knowledge of 
school 
administrators
144 school administrators or 
officials trained or certified 
as a result of USDA 
assistance
Project training 
records and 
reports
No CRS Will be verified 
through Director 
Survey.
Perceived quality 
of training will 
be verified by 
KIIs with school 
directors. 
106 school administrators in 
targeted schools who 
demonstrate use of new 
techniques or tools
Program 
observation, 
interviews sites 
visits.
Yes, but 
should also 
be validated 
through 
means in 
PMP)
CRS Will be verified 
through Director 
Survey. Self 
reflection of 
directors, and in 
KIIs with school 
directors
1.2 IMPROVED 
ATTENTIVENESS
75% of students in target schools who are 
identified as attentive during class/instruction
Attentiveness 
Observation 
Tool
Yes Evaluation 
team
1.2.1 Reduced 
short-term 
hunger
Less than 20% of parents in 
target schools indicate that 
their children were “hungry” 
during the school day
Parent Survey 
(Chart in the 
survey)
Yes Evaluation 
team
Will be verified 
through FGDs 
with parents
1.2.1.1 
Increased 
access to food 
(school feeding)
95% of students in target 
schools consuming daily 
meals at school
School 
registers; 
reported 
biannually
No CRS Will be verified 
through FGDs 
with students 
and Parents
22,268 boy students 
receiving daily school meals 
(breakfast, snack, lunch) as 
a result of USDA assistance
School 
registers; 
reported 
biannually
No CRS Will be verified 
through FGDs 
with students.
21,536 girl students 
receiving daily school meals 
(breakfast, snack, lunch) as a 
result of USDA assistance
School 
registers; 
reported 
biannually
No CRS Will be verified 
through FGDs 
with students.
18,756,998 daily school 
meals (breakfast, snack, 
lunch) provided to school￾School 
registers; 
No CRS
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK (CONTINUED) 
140
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
age children as a result of 
USDA assistance
reported 
biannually
80,703 take-home rations 
provided as a result of USDA 
assistance
School 
registers; 
reported 
biannually
No CRS
13,884 girls receiving take￾home rations as a result of 
USDA assistance
School 
registers; 
reported 
biannually
No CRS
13,017 boys receiving take￾home rations as a result of 
USDA assistance
School 
registers; 
reported 
biannually
No CRS
1.3 IMPROVED 
STUDENT 
ATTENDANCE
19,607 boy students regularly (80%) attending 
USDA-supported classrooms/schools 
80% attendance during normal school 
operating hours during the school year
Individual 
student data 
from 
school/teacher 
attendance 
records
No (End of 
Program)
CRS
19,069 female students regularly (80%) 
attending USDA-supported classrooms/schools
80% attendance during normal school 
operating hours during the school year
Individual 
student data 
from 
school/teacher 
attendance 
records
No (End of 
Program)
CRS
1.3.2 Reduced 
health-related 
absences
2% of students who report a 
decrease in health-related 
absences
Teacher Survey Yes Evaluation 
team
Will be verified 
through FGDs 
with Teachers 
and Parents, as 
well as Parent 
Survey.
1.3.3 Improved 
school 
infrastructure
144 school kitchens 
rehabilitated/constructed as 
a result of USDA assistance
Project Records No CRS School 
observations
252 cabins 
rehabilitated/constructed as 
a result of USDA assistance
Project Records No CRS Will be verified 
through 
interviews with 
School Director 
and students
144 storerooms 
rehabilitated/constructed as 
a result of USDA assistance
Project Records No CRS Will be verified 
through FGDs 
with Networks 
(PTAs, SILCs, and 
Canteen 
Management 
Committees)
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK (CONTINUED) 
141
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
1.3.4 Increased 
student 
enrollment
21,536 girls enrolled in 
schools with USDA 
assistance
School 
enrollment 
registers
No CRS
22,268 boys enrolled in 
schools with USDA 
assistance
School 
enrollment 
registers
No CRS
1.3.5 Increased 
community 
understanding 
of benefits of 
education
60% of parents in target 
communities who can name 
at least three benefits of 
primary education
Parent Survey Yes Evaluation 
team
Will be verified 
through FGDs 
with Parents.
1.4 
FOUNDATIONAL
RESULTS
1.4.1 Increased 
capacity of 
government 
institutions
Students in 144 schools 
assessed using Early Grade 
Reading Assessment tool
EGRA Yes Evaluation 
team
Will be verified 
through 
interviews with 
relevant 
Government 
stakeholders and 
project records
1.4.4 Increased 
engagement of 
local 
organizations 
and community 
groups
22,268 male social 
assistance beneficiaries 
participating in productive 
safety nets as a result of 
USDA assistance
Project records No CRS Will be verified 
through FGDs 
with networks
21,536 female social 
assistance beneficiaries 
participating in productive 
safety nets as a result of 
USDA assistance
Project records No CRS Will be verified 
through FGDs 
with networks
92 parent-teacher 
associations (PTAs) or similar 
“school” governance 
structures supported as a 
result of USDA assistance
Project records No CRS Will be verified 
through FDGs 
with Networks 
and KIIs with 
School Directors
100% of schools in target 
communities with active 
PTAs or similar “school” 
governance structures
Project records No CRS Will be verified 
through FGDs 
with Networks
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK (CONTINUED) 
142
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
MGD SO 2: Increased Use of Health and Dietary 
Practices 
• 95% of students in schools receiving USDA assistance receiving 
a minimum acceptable diet
Parent Survey Yes Evaluation 
team
2.1 IMPROVED 
KNOWLEDGE OF 
HEALTH AND 
HYGIENE 
PRACTICES
40% of parents in target 
schools who achieve a 
passing score on a test of 
good health and hygiene 
practices
Pre- and post￾test results. 
Reported 
Biannually.
No CRS Will be verified 
through FGDs 
with Parents
50% of students in target 
schools who achieve a 
passing score on a test of 
good health and hygiene 
practices
Not in PMP
Pre- and post￾test results
No CRS Will be verified 
through FGDs 
with Students 
2.2 INCREASED 
KNOWLEDGE OF 
SAFE FOOD PREP 
AND STORAGE 
PRACTICES
100% of food preparers at 
target schools who achieve a 
passing score on a test of 
safe food preparation and 
storage
Pre- and post￾test results. 
Reported 
Biannually.
No CRS
2.3 INCREASED 
KNOWLEDGE OF 
NUTRITION
4 female teachers trained in 
child health and nutrition as 
a result of USDA assistance 
(school gardens)
Project records No CRS
21 male teachers trained in 
child health and nutrition as 
a result of USDA assistance 
(school gardens)
Project records No CRS
2.4 INCREASED 
ACCESS TO 
CLEAN WATER 
AND 
SANITATION 
SERVICES
100% of schools using an 
improved water source
100% of schools using an 
improved water source
Project records No CRS Will be verified 
through school 
observation
144 schools with improved 
sanitation facilities (latrines)
Project records No CRS Will be verified 
through school 
observation
2.6 INCREASED 
ACCESS TO 
REQUISITE FOOD 
PREP AND 
STORAGE TOOLS 
AND 
EQUIPMENT
144 target schools with 
improved food preparation 
and storage equipment
Way bill; 
delivery note
No CRS Will be validated 
through Director 
Survey.
2.7 
FOUNDATIONAL 
RESULTS
2.7.1 Increased 
capacity of 
government 
institutions
4 government staff in 
relevant ministries/offices 
implicated in 
canteen/commodity 
management training
Project records Yes - KIIs 
with 
Government 
Officers
Evaluation 
team
Input from CRS
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 26 – PERFORMANCE MEASUREMENT FRAMEWORK AND 
INFORMATION AVAILABLE AT THE INCEPTION REPORT STAGE AND BEFORE 
FIELD WORK (CONTINUED) 
143
Result Stream Intermediate 
Result
Expected Result Tool or Data 
Source for 
Information 
Specific 
Endline 
Survey 
Responsibili
ty for 
Information 
Comments 
2.7.4 
Increased 
engagement 
of local 
organizations 
and 
community 
groups
71 nutrition or health 
initiatives or activities 
pursued in partnership 
between government and 
local community groups
Project records No CRS Will be verified 
through FGDs 
with Networks
MGD illustrative: Percent of schools with soap and water at hand￾washing stations commonly used by students11
Observational 
Tool
Yes Evaluation 
team
 
11 To be considered a “yes,” 50% or more of hand-washing stations must have soap and water in each school and 
the enumerator must provide the answer “yes, very much” to whether there is evidence that students are using the 
stations.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
144
APPENDIX 27 – TERMINOLOGY AND DEFINITIONS 
Below are the main terms and concepts the evaluation team used in organizing the data collection and 
analysis, presenting conclusions and framing recommendations. The first set are those used by the 
evaluation community in general—from DAC/OECD, the UN and development agencies—and refer to 
criteria against which to assess implementation and results and to explain those results (for instance, why 
they are/are not happening). The second set specifically relate to/are taken from the FFE project, where 
possible quoted directly from key documents, but otherwise based on the evaluation team’s 
interpretation of references in them.
SET A 
• Appropriateness – Concerns the technical aspects or methods of a project, such as the extent to 
which FFE’s design and activities are consistent with good education, health and hygiene practice 
and likely therefore to result in solutions to the problems it is trying to address or 
objectives/changes it is trying to realize. Appropriateness also takes into account specific 
contextual, capacity and resource factors, such as an approach may be technically appropriate in 
the US, but not in Benin given the surrounding conditions.
• Effectiveness – Concerns the success of a project in realizing its expected results (for instance, have 
the policy leaders, implementers and partners fulfilled their roles, responsibilities and functions in 
ways that have contributed, or are likely to contribute, to the well-being of beneficiaries, 
communities, the government, the development needs of the sector?).
• Efficiency – Concerns the extent to which a project’s resources (inputs) have been well-balanced 
against the cost and quality of activities delivered, and results achieved. It concerns time, effort, 
funds and reputation, and is measured by both value-for-money and also by stakeholders’
perception of whether the activity was “worth the effort” (meaning would they do it again?).
• Indicator – Also called evidence, indicators are the measures that a project and its partners, donors 
and beneficiaries agree will be used to confirm that a result (outcome) is being/has been achieved. 
Outcomes typically have multiple indicators reflecting their different aspects. Indicators are not 
targets; targets define the extent (number, scope, degree) an indicator is expected to reach during 
the project (benchmarks) or by its end (ideally set against baselines). Indicators are set at the 
beginning, but can be changed if found to be inadequate or inappropriate as evidence of results.
• Impact – This is related to positive and negative, primary and secondary long-term effects 
produced by a development intervention, directly or indirectly, intended or unintended.
• Informed and timely action – Also called knowledge management, this concerns the extent to 
which a project has designed and consistently applied from the outset good quality results-based 
monitoring and evaluation regime; and has used these data to guide, adapt and improve resources 
and action in a flexible and responsive way.
• Innovation/creativity – This is related to informed action above in that it concerns the extent to 
which a project has been able to adapt to new information and changing circumstances in ways 
that increase the likelihood of outcomes being realized (for instance, has it been able to 
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 27 – TERMINOLOGY AND DEFINITIONS (CONTINUED) 
145
experiment with variations of the design, been ready to take risks to achieve added value, and/or 
generated and used lessons learned from its ongoing experience?). 
• Outcome – An outcome is a describable or measurable change that is produced through a project’s 
inputs or outputs. Outcomes can be described by degrees of change (as immediate, intermediate 
or ultimate, for instance). Immediate outcomes are the weakest changes realized by project 
beneficiaries, and may not be sustainable. However, they contribute to intermediate outcomes that 
are more substantial and reflect a significant change in beneficiary thinking or capacities. Such 
changes are likely to be sustained because the individuals have consolidated them in some way. 
Intermediate outcomes then contribute to ultimate outcomes (usually one per project) that are 
changes in the wider environment. Ultimate outcomes are usually outside the control of the 
project, but concern the change towards which the project aims to contribute. Often an ultimate 
outcome is framed as the rationale for the project.
• Output – Outputs are products of the project itself, in terms of actions taken and completed, such 
as modules developed, teachers trained, materials written. They exist within the boundaries, and 
are due to the capacities and resources, of the project. They do not say anything about the effects 
on beneficiaries—whether they changed beneficiaries or not:
 If CRS staff make the planning and management decisions about the food supply, the fact that 
food is supplied in a timely, sufficient way is an output (produced by the project), whereas if 
planning and management decisions are made by the AMEs or APEs following training, then it is 
an outcome (a change in beneficiaries—their knowledge, skills, commitment, etc.).
 If teachers are trained on the modules by the WEI, this is an output (an activity completed by 
the project), whereas if those teachers later demonstrate their new learning in their teaching 
methods, then it is an outcome (a change in beneficiaries, their knowledge and skills).
 If use of these skills is modest and teachers do not adapt them to different classroom situations, 
it is an immediate outcome, subject to decay if there is no follow-up to support the teachers in 
the classroom. If their use of the skills is continuing to strengthen over time, to be adapted, to 
be talked about with other teachers, it is an intermediate outcome—consolidated and so likely 
to be sustained because teachers have internalized commitment to it. The ultimate outcome in 
this case might be “All Benin children, boys and girls, have a positive, healthy, quality learning 
experience and complete their education to be productive citizens.”
• Partnership and collaboration – Though different in intensity and management, these both concern 
the extent to which, in design and implementation, the project has worked with others to create 
and strengthen joint action, share risk and resources, and ensure a comprehensive approach. 
Collaborative relationships and partnerships should add value and foster local ownership and 
sustainability.
• Relevance – Concerns the extent to which the project goals are consistent with/contribute to the 
expressed priorities, identified needs or interests of the target beneficiaries, the implementing 
agents, and national and/or sector policies. It refers to the importance of a project to stakeholders 
and the wider community, not to its technical aspects. In fact, an activity that is technically poor 
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 27 – TERMINOLOGY AND DEFINITIONS (CONTINUED) 
146
may still succeed where those involved care about what it is trying to do, and are committed to it 
because it addresses their core concerns. The opposite is also true.
• Sufficiency – Concerns the extent to which inputs (activities and resources) of a project have gone 
on long enough, have gone deep enough, and have been comprehensive enough to bring about 
the intended changes. Even where the activity has been technically “right” (for instance, a 
teachers’ training workshop), it may have been too limited in duration, coverage or reach to make 
a difference (for instance, to allow learning to be consolidated).
• Sustainability – This addresses the extent to which the actions/strategies of a project overall, 
and/or its different components, seem likely to have a lasting influence at the individual, local, 
national or regional levels following termination of the assistance. It concerns whether the changes 
achieved are stable enough to be maintained in the face of pressure to return to the status quo. 
Critical questions to be answered by a project from the outset are what it expects to be sustained 
and whether it is ensuring that the conditions/factors needed to so do are in place/being created.
• Theory of change – This explains the underlying assumptions of a project, how and why it expects 
its inputs to produce its results, and what actions will lead to what changes and under what 
conditions. The ToC, then, outlines the mechanisms of change, as well as the assumptions, risks 
and context that support or hinder it. While a ToC is related to the “programme logic” or “logic 
model” of a project, it is more fundamental because it precedes and informs them. ToCs always 
exist, but are often unstated; are not always recognized by the project designers, managers and 
partners; and are not always agreed by all stakeholders when they do put them into the open. It is 
often the failure to make everyone’s ToC explicit, and reconcile differences, that causes a project to 
fail in the end as people work along different tracks.
SET B 
• Attentive/attentiveness – According to the MTE, “[t]he baseline study used the concept of 
attentiveness which includes an element of behavio[u]r, such as paying attention. The concept 
used in this study relies on behavio[u]r only, i.e., what students actually do that is relevant to 
learning to read.”12 The baseline defined attentiveness through two sets of descriptors for 
attentive/inattentive. Because the MTE tool was not provided, and to maintain consistency 
between baseline and endline results, the consultants will use the baseline definition.
• Indicators – The FFE documents were not always clear about what is the expected outcome (result) 
and what are indicators and targets. For example, according to the baseline report: “The two main 
outcomes of interest for parents are the percentage of parents who can name at least three 
benefits of primary education, and the percentage of parents who receive a passing score on a test 
of good health and hygiene. For both indicators, the goal is 60%.” In this case, it makes sense to 
consider naming benefits and passing tests as indicators, but then “of what”? Are the outcomes in 
this case (a) parents more mobilized and sending children to school, and/or more children 
 
12 Mid-Term Evaluation of FFE: Intervening in a Demanding Context. Report prepared by World Education (Maurice 
Garnier, Pierre GBENOU and Serge Marcel Loukpe). April 2017.
Endline Evaluation of CRS FFE in Benin – Final Report September 2018
APPENDIX 27 – TERMINOLOGY AND DEFINITIONS (CONTINUED) 
147
attending school; and (b) children are healthier and/or fewer children are absent due to illness, 
fatigue?
• Reading competence – Is defined in the MTE as obtaining 2 out of 5 points on the reading and 
comprehension parts of EGRA. This indicator specifically states that Grade 2 students should be 
assessed.
• Sustainability – According to the baseline report, the FFE “will ensure sustainability by leveraging 
community-based activities to engage parent-teacher associations (Associations des parents 
d’élèves) and school mothers’ associations (Associations des mères d’élèves) in the management of 
school canteens and other activities.”
13 It is not clear in the documents what “leveraging” means in 
this case, or what indicators are being used to confirm it.
Sustainability is also implied in the FFE intention to “increase government capacity and ownership” 
by supporting the INFRE to “design improved teacher training” and “train pedagogical advisors [...] 
who will in turn directly train and monitor an estimated 212 teachers and 141 officials to assess 
literacy levels and share new and effective teaching methods.”
14 “Ownership” is often used as an 
indicator of sustainable change/outcomes.
• Theory of Change – Based on the MTE, which refers to the FFE concept of “enabling conditions,” 
the project reflects what could be considered a reasonable ToC, meaning that an appropriately 
nourished student practising good hygiene and who engages in healthy behaviour will be “ready to 
learn.” That student will not usually miss school and, when in school, he or she will be attentive, 
participate and learn. A second possible ToC, implied by the baseline, is that “[t]he project will 
improve literacy through provision of school lunches which will increase student attentiveness and 
reduce student absences.”
15
• Reduced short-term hunger – Defined as the percentage of parents whose children (taken from a 
statistically representative sample) state they were “hungry” during the school day.
• Reduced health-related absences – This indicator is to be calculated as a percentage, examining the 
number of students absent due to illness divided by the total number of absent students. This will 
be determined through the attendance records kept by teachers, which are reported to record the 
reason for a student’s absence.
• Increased community understanding of benefits of education – This is defined as the percentage of 
parents who can name at least three benefits of schooling (socio-economic, personal/socialization 
of children, society gains, etc.).
• Minimum acceptable diet – This indicator examines the percentage of students receiving a 
minimum acceptable diet from USDA-supported canteen activities and family meals. Dietary 
diversity is defined as the number of individual food items or food groups consumed over a given 
 
13 Brown, D., and J.C. Guzmán. (March 2015). CRS Food for Education Baseline Study. Notre Dame (Indiana), The 
University of Notre Dame, Notre Dame Initiative for Global Development. p. 11.
14 Ibid.
15 Idem. p. 11.
September 2018 Endline Evaluation of CRS FFE in Benin – Final Report
APPENDIX 27 – TERMINOLOGY AND DEFINITIONS (CONTINUED) 
148
period of time. It is measured by counting the number of food groups rather than food items 
consumed. At the individual level it reflects dietary quality, mainly micronutrient adequacy of the 
diet. According to the baseline report the USDA classifies foods into seven food groups:
 Grains, roots and tubers;
 Legumes and nuts;
 Dairy products (milk, yogurt, cheese);
 Flesh foods (meat, fish, poultry and liver/organ meats);
 Eggs;
 Vitamin-A enriched foods, including vegetable oil, fruits and vegetables;
 Other fruits and vegetables.
A minimum acceptable diet is defined as eating four or more of the seven food groups, and a 
minimum frequency of three times a day.
• Increased capacity of government institutions – This indicator refers to the number of relevant 
government staff with an active role in training canteen committees in proper handling of 
commodities.
• Percent of schools with soap and water at hand washing stations commonly used by students – This 
indicator allows to verify the application of one of the three key hygiene practices, namely:
 The presence of water at hand washing stations;
 The presence of soap or ash; and
 Signs of use.