This publication was produced at the request of the United States Department of Agriculture. It was 
prepared independently by International Solutions Group
Foreign Agricultural 
Service, United States 
Department of Agriculture
Photo: Students in a classroom with materials
Senegal McGovern-Dole Food 
For Education Project
Final Evaluation
August 2018 
DISCLAIMER: 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.
Final Evaluation McGovern Dole International Food for 
Education and Child Nutrition Program Saint Louis Region, 
Senegal
With funding support from the United States Department of Agriculture (USDA), Counterpoint 
International (CPI) implemented the McGovern-Dole Food For Education project in the Saint 
Louis Region of Senegal from 2014 to 2018. The key objectives of the project were to improve
the literacy of school-age children and increase the use of health and dietary practices. 
Agreement Number: FFE-685-2014-029-00
Project Duration: 2014-2018 
Implemented by: Counterpart International
Evaluation Authored by: International Solutions Group
Evaluation Team Leader: Jason Wares
Co-Evaluator: Christopher Root
Technical Specialist: Noreen Mucha
1
Final Evaluation 
McGovern-Dole International 
Food for Education and Child 
Nutrition Program
Saint Louis Region, Senegal
PREPARED BY 
INTERNATIONAL SOLUTIONS 
GROUP
PREPARED FOR
Counterpart International
14 August 2018
2
TABLE OF CONTENTS
TABLE OF ACRONYMS 6
EXECUTIVE SUMMARY 1
FINDINGS SUMMARY 2
STRATEGIC OBJECTIVES 2
ACTIVITY LEVEL RESULTS 3
RECOMMENDATIONS SUMMARY 4
METHODOLOGY 5
DOCUMENT REVIEW 5
QUALITATIVE RESEARCH 5
QUANTITATIVE RESEARCH 6
ANALYSIS 7
IMPROVEMENTS IN PRIMARY SCHOOL STUDENTS’ LITERACY 7
SCHOOL FEEDING SUSTAINABILITY 7
LIMITATIONS OF THE EVALUATION 8
FINDINGS 8
STRATEGIC OBJECTIVE RESULTS 8
SO1: IMPROVED LITERACY OF SCHOOL-AGED CHILDREN 8
IMPROVED LITERACY 9
REDUCED SHORT-TERM HUNGER 14
IMPROVED STUDENT ATTENDANCE 15
INCREASED GOVERNMENT SUPPORT 16
SO2: IMPROVE HEALTH AND DIETARY PRACTICES RESULTS 17
PERCENT OF CHILDREN RECEIVING A MINIMUM ACCEPTABLE DIET AT THE SCHOOL LEVEL 17
INCREASED KNOWLEDGE OF SAFE FOOD PREPARATION AND STORAGE PRACTICES 18
IMPROVED KNOWLEDGE OF HEALTH AND HYGIENE PRACTICES 19
ACTIVITY LEVEL RESULTS 20
SCHOOL SELECTION 20
ACTIVITY 1: CAPACITY BUILDING AT THE LOCAL, REGIONAL, AND NATIONAL LEVEL 22
ACTIVITY 2: CONSTRUCTION OF SCHOOL INFRASTRUCTURE 24
ACTIVITY 3: CONSTRUCTION OF LATRINES AND WATER STATION SYSTEMS 24
ACTIVITY 4: EQUIP SCHOOLS WITH ENERGY SAVING STOVES, CANTEEN EQUIPMENT, AND MATERIALS 26
ACTIVITY 5: GOOD HEALTH AND NUTRITION PRACTICES 27
ACTIVITY 6: TRAINING IN FOOD PREPARATION AND STORAGE PRACTICES 32
ACTIVITY 7: PROVIDE ACCESS TO HEALTH INTERVENTIONS 35
ACTIVITY 8: THE SCHOOL FEEDING PROGRAM 36
ACTIVITY 9: ESTABLISHING COMMUNITY FARMS 42
ACTIVITY 10: IMPROVED TEACHER ATTENDANCE 48
ACTIVITY 11: PARENT ASSOCIATION TRAINING 49
MONITORING AND EVALUATION OF PROGRAM ACTIVITIES 49
3
EVALUATION QUESTIONS 50
IMPLEMENTATION AND RELEVANCE 50
EFFECTIVENESS AND PERFORMANCE 52
EFFICIENCY 53
SUSTAINABILITY 54
LESSONS LEARNED 57
MGD PROVIDES AN IMPORTANT OPPORTUNITY FOR TESTING NEW IDEAS. 57
MGD’S COMPLETE DESIGN APPROACH WAS CRUCIAL FOR ITS SUCCESS. 57
GENDER ISSUES REQUIRE CLOSER EXAMINATION 57
THE MGD PROGRAM DOESN’T HAVE A CLEAR UNDERSTANDING OF THE SUFFICIENCY OF STUDENTS’ DIETS 57
STUDENT ATTENDANCE IMPROVES BECAUSE THERE IS FOOD AT SCHOOL 58
TEACHER ATTENDANCE MAY NOT BE AN ISSUE 58
SCHOOL NEEDS VARY 58
SUSTAINABILITY INITIATIVES SHOULD START EARLY IN THE PROGRAM 58
RECOMMENDATIONS 59
ANNEX 1: DOCUMENT REVIEW LIST 62
ANNEX 2: DATA COLLECTION TOOLS 63
ANNEX 3: SELECTED SAMPLES BY TYPE AND LOCATION 87
ANNEX 4: SURVEY RESPONDENT CHARACTERISTICS 89
ANNEX 5: SCHOOL SAMPLE 90
ANNEX 6: EVALUATION TEAM BIOS 92
ANNEX 7: INDICATOR FRAMEWORK – BASELINE VS MID-TERM DATA 94
ANNEX 8: EVALUATION TERMS OF REFERENCE 96
4
TABLE OF FIGURES
Figure 1 Reading Score by Gender (Average) ..............................................................................................................11
Figure 2 Proportion of Students Reading At or Above Level By Grade ........................................................................12
Figure 3 CM1 Reading Level Distribution.....................................................................................................................13
Figure 4 CE2 Reading Level Distribution ......................................................................................................................13
Figure 5 CE1 Reading Level Distribution ......................................................................................................................14
Figure 6 Frequency with which students are hungry during school.............................................................................15
Figure 7 Minimum Acceptable Diet .............................................................................................................................18
Figure 8 Percentage of Beneficiaries Using Appropriate Hand Washing Practices .....................................................19
Figure 9 Student knowledge of ways to prevent intestinal worms..............................................................................19
Figure 10 Situations when Teachers Washed Their Hands (%) (Multiple Responses Possible)....................................30
Figure 11 The Amount of MGD Food that Is Currently and Potentially Covered by Granaries and School Farms.......45
Figure 12 Contributions to Primary School Canteen Sustainability..............................................................................46
5
LIST OF TABLES
Table 1 Total Interviews Conducted 5
Table 2 Improvement in reading level by grade 10
Table 3 Change in Cohort Reading Level 10
Table 4 Expected Reading Level by Grade 11
Table 5 Maximum Possible Student Attendance 2018 16
Table 6 Minimum Acceptable Diet Food Groups 18
Table 7 School Categorization 21
Table 8 Training Dates, Themes, and Participants 22
Table 9 Classroom Construction Targets and Progress 24
Table 10 Progress toward School Latrine 24
Table 11 Schools with latrine and water 25
Table 12 Cleanliness of School Latrines (n = 159) 25
Table 13: Latrine's condition (mid-term n = 96, Final Evaluation n = 159) 26
Table 14 Health and Nutrition Trainings 28
Table 15 The degree to which teachers have received and use training (Mid-Term n =312, Final = 356) 29
Table 16 Degree to which teachers instruct on nutrition and hygiene (n=312, N=356) 30
Table 17 Student Situtions for washing hands (Mid-Term n = 1028, Final n = 787) 31
Table 18 Access to Hand Watching Facilities (mid-term n = 96, final n = 159) 32
Table 19 Condition of Food Storage Facilities % (Mid-Term n =111, Final N = 159) 33
Table 20 Conditions for storage in warehouse 34
Table 21 Director knowledge of food storage facility maintenance (mid-term n = 108, final n = 159) 34
Table 22 Directors knowledge of inventory control (mid-term n = 108, Final = 160) 34
Table 23 Directors knowledge of serving safe food 35
Table 24 McGovern-Dole Meal Schedule 37
Table 25 Food required to sustain the School Feeding Activity 38
Table 26 School Granary Inventory - Oct 2017 to Feb 2018 (KGs unless noted) 39
Table 27 Criteria for School Categorization 40
Table 28 Days providing meals without USDA assistance by school category 41
Table 29 Number of meals provided without USDA assistance by School Category 42
Table 30 School farm 2017 performance with one and two hectare projections 44
Table 31 Contribution to School Canteen Sustainability 47
6
Table of Acronyms
Acronym Definition
CE1 3rd Grade
CE2 4th Grade
CFA West African Frank
CM1 5th Grade
CPI Counterpart International
DCMS-MEN School Medical Control Division
ISG International Solutions Group
IYCF infant and young child feeding
JICA Japan International Cooperation Agency
LOP Life of Project
MAD Minimum Acceptable Diet
MDD Minimum Dietary Diversity
MGD McGovern-Dole
MMF Minimum Meal Frequency
MoA Ministry of Agriculture
MoE Ministry of Education
SMC School Management Committee
SPI Social Progress Index
TOR Terms of Reference
USD United States Dollar
USDA United States Department of Agriculture
WHO World Health Organization
1
Executive Summary
In Senegal, only 72.3 percent of primary school-aged children are enrolled in school.
1 Of the children that 
enroll, 58.9 percent continue to complete primary school.
2 The 2017 Social Progress Index (SPI) ranks 
Senegal 116th out of 128 countries in access to primary education3. In comparison, the SPI ranks Namibia 
91st in the world and Ghana 19th.
4 More generally, Senegal is substantially below the level of countries at a 
similar level of economic development as defined by the SPI.
5 The low enrollment and completion in primary 
education translates to low literacy rates for adults (42.8% for adults, 52.8% for men, 33.6% for women).6
Approximately 40 percent of adult women in Senegal are literate, and 66 percent of Senegalese adult men are 
literate.
While these statistics describe Senegal as a whole, the problem of insufficient school access may be especially 
acute in the Saint Louis region. One obstacle to school access in Saint Louis is the region’s dependence on 
agriculture. In Saint Louis, 60 percent of the population depends on agriculture for their income.
7 The 
agriculture sector in Saint Louis has experienced increased food insecurity due to irregular rainfalls and 
shortfalls in investment.
8 In Senegal, 85 percent of children that work are employed in the agriculture sector.
When food insecurity increases primary school-aged children work more and attend school less.
9
The McGovern–Dole International Food for Education and Child Nutrition Program (MGD) works around 
the world to address the types of access to education and literacy challenges that Senegal faces. In September 
of 2014, USDA FAS awarded a grant to Counterpart International to implement a program that would 
“improve food security, reduce the incidence of hunger, and improve literacy and primary education” in 
Senegal’s Saint Louis region. After the completion of the baseline report, the project was launched in 2016. 
Counterpart International hired International Solutions Group (ISG) to conduct a final evaluation of the 
McGovern-Dole program as a follow up to the mid-term evaluation ISG conducted in May 2017. The 
purpose of the final evaluation was to “gain an understanding of what the project has achieved, whether the 
project has achieved its targets and objectives and what can be learned from it and provide recommendations 
for future programming to improve implementation and contribute to the MGD learning agenda.” The 
evaluation sought to answer questions related to the project’s implementation, effectiveness and performance, 
efficiency, and sustainability. The evaluation questions are listed on page 10 of the terms of reference, 
included as Annex 8. 
This document presents the results of the final evaluation. Results were gathered through a rigorous mixed￾methods data collection methodology. To conduct the evaluation, the evaluation team surveyed 787 students, 
355 teachers, and 159 school directors. The team also recorded observations of facilities at 159 of the 270 
schools that participated in the program. These observations were of school infrastructure, latrine access, 
 
1 World Bank’s Databank “World Development Indicators” database. 2015 Data series
2 Ibid.
3 https://www.socialprogressindex.com/?tab=2&code=SEN
4 Ghana and Namibia chosen for points of comparison because the SPI includes them with Senegal in the “Lower 
Middle Social Progress” category, they are sub-Saharan African countries, and have similar scores in the 
“Foundations of Wellbeing” dimension of the index. 
5 https://www.socialprogressindex.com/?tab=2&code=SEN
6 2013 data; World Development Indicators Databank; http://databank.worldbank.org/data/source/world￾development-indicators.
7 http://www.fao.org/emergencies/countries/detail/en/c/161500
8 FAO Emergencies – Senegal brief
9 World Bank’s Databank “World Development Indicators” database. 2015 Data series
2
hand-washing facilities, food storage facilities, and school canteens. The student, director, and school 
observation sample sizes correspond to a 95 percent level of confidence and a margin of error of 5 percent. 
The statistically significance of the teacher survey corresponds to a 95 percent level of confidence and a 6 
percent margin of error.
The team also conducted qualitative information gathering to add context and narrative around the 
quantitative data collection results; conducting key informant interviews and informal focus groups that with 
93 stakeholders. These stakeholders were students, teachers, parents, school cooks, school farm presidents, 
school directors, project staff, and government officials at the local, regional, and national levels. 
Findings Summary
The final evaluation presents findings at the strategic objective and activity indicator levels. It also answers 
specific questions included in the terms of reference (TOR) regarding program implementation and relevance, 
effectiveness and performance, efficiency, and sustainability. 
Strategic Objectives
McGovern-Dole in Senegal is focused on achieving two strategic objectives (SO). These are:
• SO1: Improved Literacy for School Aged Children
• SO2: Increased Use of Health and Dietary Practices
Under SO1, McGovern-Dole seeks to improve the literacy of children by achieving four results:
1. Short-term hunger and literacy gains: The evaluation found that students who said that they 
were not hungry during the school day were 61 percent more likely to read at or above grade level
than students who said they were hungry more often than from time-to-time. 87 percent of students 
said that they were not hungry at school except “from time to time.”
2. Improved literacy of school-aged children: Final evaluation ASER test results demonstrate that 
the McGovern-Dole program has improved student literacy. The ASER test scores a student’s 
reading ability on a scale of twelve reading levels. Students who score at level 0 or 1 cannot do more 
than identify letters, while students who score at level 11 have advanced reading ability. Students in 
the CE1 grade are expected to read at level 5, CE2 students are expected to read at level 7, and 
students in CM1 should read at level 9. On an absolute basis, the average reading level for CE1 
students improved from 3.31 to 4.88 between baseline and final evaluation, an improvement of 1.57 
levels. Students in CE2 improved on average from 4.89 to 6.39, an improvement of 1.5 levels. 
Finally, CM1 students improved on average from 6.19 to 7.69, an improvement of 1.5 levels. In the 
full report, the evaluation team adjusts the baseline score because the final evaluation was delivered 
later in the school year than the baseline report. Even with the penalty on baseline scores that the 
adjustment applies, a clear improvement in reading is evident. 
Importantly, the proportion of students that read at their grade level or higher increased 
dramatically between baseline and final evaluation. The percentage of CE1 students that read at 
their grade level or higher increased from 33 percent to 53 percent between baseline and final 
evaluation. CE2 students reading at their grade level or higher increased from 27 percent to 48 
percent, and for CM1 students the increase was from 28 percent to 43 percent. 
3. Improved student enrollment and attendance: Student attendance has increased since the 
baseline study. Teachers and directors that the evaluation team interviewed said that enrollment had 
increased and that students stayed at school longer during the school day as a result of the school 
feeding program. Student attendance records support an increase in school attendance. However, 
the evaluation team suspects that official school records overstate attendance, which show a 99.5 
3
percent attendance rate. This issue is further discussed in the findings and recommendations 
sections. 
4. Increased government support: Increased government support, defined as the percent of 
teachers who had received pedagogical training in the three months prior to the survey, had declined 
between baseline and mid-term. The indicator improved significantly between mid-term and final 
evaluation. At final evaluation, 32 percent of teachers reported that they had received pedagogical 
training in the three months previous to the survey compared to 13 percent at mid-term. The 
evaluation team suspects that the 32 percent average understates actual training as teachers may have 
not counted monthly peer to peer trainings that many teachers attend. 
Under SO2, McGovern-Dole seeks to improve health and dietary practices through achieving three results:
1. Increased percentage of children receiving a minimum acceptable diet: The final evaluation 
found that 58 percent of students received a minimum acceptable diet over school. This is an 
increase over the 43 percent at midterm and 2 percent at baseline who receive a minimum acceptable 
diet at school.
2. Increased knowledge of safe preparation and food storage practices: The strategic objective￾level measure for this indicator is the percentage of beneficiaries who use appropriate hand washing 
practices before meals, before food prep, after using the latrine, and after diaper changing. The final 
evaluation team calculated that 71 percent of boys and 75 percent of girls use appropriate hand 
washing practices, a clear improvement over the baseline finding of 50 percent for boys and 55 
percent for girls
3. Improved knowledge among students of health and hygiene practices: The final evaluation 
estimates that 53 percent of boys and 51 percent of girls can identify at least two ways to prevent 
intestinal worms (the central measure for progress under this objective). This result is a significant 
improvement over both the 8 percent of boys and 10 percent of girls recorded at baseline, and the 45 
percent of boys and 41 percent of girls could do the same at mid-term. 
Activity Level Results
The McGovern-Dole program aims to produce outputs in 11 activity categories and track achievement 
through 52 indicators. The mid-term evaluation found that the program was on track to achieve nearly all of 
the required results under these indicators. Some highlights of the program’s achievement include: 
• 15,151,085 meals served in school over the course of the program against a life of project (LOP)
target of 13,750,000. 10
• 573 Parent Association Members trained against a LOP project target of 540.
• 808 Cooks trained against a LOP target of 540.
• 90,576 doses of de-worming medication distributed compared to an LOP target of 99,998.
• 20 community farms established against an LOP target of 20.
• 270 durable, culturally appropriate energy efficient stoves produced and distributed against an LOP 
target of 270. 
 
10 Targets reflect figures included in USDA’s amendment III to the McGovern-Dole agreement signed by Counterpart 
on June 6, 2017 and USDA on June 1, 2017. 
4
Overall, the evaluation team found that Counterpart’s strategy for program implementation, which creates 
locally sustainable assets around the complete school feeding and literacy program, is an important 
component of the program’s success. The full report describes the quality of these outputs, reasons for 
successes, and areas where improvements could be made.
Recommendations Summary
The report concludes with a discussion of recommendations for future McGovern-Dole programs. We 
include a summary of the recommended actions here. 
1. Continue the program: USDA should continue the program in Saint Louis for another 2-3 years to 
provide time for schools to implement sustainability programs. Ending the program now would be 
pulling food from thousands of kids who have just started on an educational path equipped with the 
energy that USDA commodities provide. 
2. Overcome the diminishing marginal utility of the program: The program showed significantly 
larger gains between baseline and mid-term than between mid-term and final evaluation. To 
overcome the diminishing utility of the program, USDA could initiate second phase initiatives that 
may have a large impact on primary school literacy and improved health practices. 
3. Focus school selection on high and medium food insecurity schools: McGovern-Dole’s school 
selection process and mid-term reorientation toward need-based food provision worked well. In 
future McGovern-Dole programs, McGovern-Dole should use the needs identification process that 
Counterpart developed to categorize schools during the school selection process. The process should 
incorporate means of identifying school needs that include level of food security as well as other 
criteria. 
4. Work with school directors to plan for maximizing each school’s resources: McGovern-Dole
should work with directors to develop school feeding sustainability strategies based on different 
levels of food availability. 
5. Create a School Feeding Board: School feeding programs need to be run more professionally if 
they are to become more sustainable. McGovern-Dole Senegal should consider forming a school 
feeding program board of directors to guide decision making, resource allocation, and school feeding 
coordination. 
6. Identify ways to overcome school farms’ major constraints: School farms potential is limited by 
three major constraints; insufficient access to finance, labor shortages, and inappropriate pump 
provision. School farms would also benefit through forming model farms in partnership with the 
MoA. 
7. Use random spot checks as well as school data to track increased attendance: To truly 
understand the way the program influences school attendance, McGovern-Dole should have an 
independent research plan for studying the topic from the beginning of the program.
8. Research whether infrastructure construction activities provide sufficient value: Interviews 
and surveys revealed that the school feeding and health initiatives were primarily responsible for 
improvements in reading. Infrastructure projects were less successful and created unfulfillable 
expectations among some project beneficiaries. School infrastructure projects were also the single 
biggest direct cost line item in the program's budget. The funds allocated to school infrastructure 
projects could have funded school feeding programs for over 100 new schools and over 17,000 
additional students. Schools certainly require repair and upgrading. USDA could use the data it has 
on hand from previous McGovern-Dole programs to see if there is a correlation between schools 
that benefit from infrastructure projects and improved reading scores. It could also allocate funds for 
school maintenance and upgrading rather than new construction, which would cost less, and use the 
savings to serve more schools. 
5
Methodology
The evaluation team carried out a final evaluation in April and May of 2018 with fieldwork and data collection 
in Senegal taking place between April 20 and May 5. The team used mixed methods including document 
review, qualitative research including key informant interviews (KIIs), focus group discussions (FGDs), and 
quantitative data collection through surveys. The mixed methods approach allows the evaluation team to 
strengthen qualitative findings through statistically representative survey data, and to provide depth and 
context to survey findings. 
Document review
Document review informed all stages of the evaluation. The team used the document review initially to learn 
about the project and inform data collection tools. Later the review was useful in explaining different features 
of program design and tracking the evolution of the program over time. McGovern-Dole’s monitoring and 
evaluation team provided data and documents that served as source data for the evaluation team used to 
compare performance results against LOP targets, assess the frequency of trainings, and changes in student 
and teacher attendance. Finally, project records were used to analyze school farm and granary performance 
and potential. Annex I includes a full list of documents the team reviewed. 
Qualitative research
The evaluation team conducted interviews with a total of 93 respondents to provide context and depth to
survey results. The team asked questions intended to expand its understanding of the McGovern-Dole
project, answer evaluation questions, and confirm information gleaned from other sources. Question guides 
were used to lead semi-structured interviews. The guides ensured that the evaluation team asked the same key 
questions across similar respondent types while still allowing the freedom for open-ended, conversational￾style interviews. These guides are included in Annex 2.
Qualitative field research was conducted by the evaluation team in the three departments (Dagana, Podor and 
Saint Louis) that make up the project area. In total, the evaluators visited 18 schools. The schools chosen for 
the evaluators to visit were all part of the random sample of schools chosen for the survey. One evaluator 
focused primarily on schools with school farms to allow for evaluation of this pilot program (Activity 9 in the 
findings section). Nine schools with school farms were visited. Other interviews happened at schools where 
enumerators were conducting surveys. Within schools, the evaluators conducted interviews with the 
directors, teachers, school farm presidents, cooks and parents, based on availability. The evaluation team 
conducted informal focus groups with students at primary schools.
In addition to interviews at schools, the evaluation team conducted KIIs with education administrators and 
other government officials at the national, regional and departmental levels. Finally, the team interviewed 
McGovern-Dole project staff. A summary of KIIs and FGDs conducted is presented in table 1 below. 
Table 1 Total Interviews Conducted
CATEGORY NUMBER
Directors 17
Teachers 15
Students 24
Parents 8
Cooks 9
Government 9
Project Staff 7
Total 93
6
Quantitative Research
The final evaluation utilized student, teacher, and school director surveys. The questionnaires largely followed 
those used for the mid-term evaluation. Questionnaires are available in Annex 2. In addition, enumerators 
completed a school observation checklist at each school in the sample. Sampling for the surveys followed a 
two-stage cluster design, with the schools comprising the first cluster schools and the subject of the survey 
(students, teachers, or directors) comprising the second. In order to achieve necessary statistical strength, the 
first cluster required 159 randomly selected schools of the 270 project affiliated schools. 
Student Sample
For the student sample, we randomly selected 8 
students from each of the 99 primary schools included 
in the first cluster. This created a total student sample 
of 787 students, corresponding to a confidence level 
of 95% and a margin of error of 5%. In conducting 
the survey, this sample size target was reached. 
Teacher Sample
To reach the same confidence level and margin of 
error, the required teacher sample was required 404 
teachers across 135 of the 159 schools chosen in the first cluster. As a teacher strike was underway and there 
was a lack of certainty around which teachers would be present at any school, enumerators were instructed to 
interview 3 available teachers per school. As it turned out, fewer than three teachers on average were actually 
available to serve as survey respondents in each school, and 355 teacher surveys were completed.
Accordingly, results are statistically significant at 95 percent level of confidence and 6 percent margin 
of error. 
Director and Checklist samples
Director interviews and checklists were completed at all 159 schools in the sample to achieve a confidence 
level of 95% and a margin of error of 5%. 
Note on Comparability
Please note that at mid-term, information collected on school facilities through the mid-term evaluation 
team’s checklist instrument, teacher surveys, and school director surveys had wider margins of error and 
lower levels of confidence than the final evaluation instruments. The section indicates the margin of error and 
level of confidence where necessary for appropriate comparison.
EXHIBIT 1 MAP OF 159 SURVEYED SCHOOLS
7
Analysis
Improvements in Primary School Students’ Literacy
The evaluation team measured the reading skills of primary school students. This measurement was based on 
the same ASER reading test conducted in support of the baseline report.11 The evaluation team compared 
reading scores at baseline, midterm, and final evaluation by grade to determine if the program caused a 
change in student’s reading skills. The reading scores for CE1 (3rd Grade) students from before the 
McGovern-Dole program started are compared with CE1 students at mid-term evaluation and at the time of 
final evaluation, as were scores for CE2 (4th Grade) and CM1 (5th Grade) students. These differences describe 
the change in primary school students’ reading skills for each of the three grades.
Note that the literacy analysis includes a correction to account for the difference in timing of the baseline and 
midterm surveys. The baseline survey was conducted in January and February of 2016 whereas the midterm 
survey in May and June of 2017 and the Final Evaluation in May of 2018. The difference in timing means that 
students at midterm and final evaluation had nearly half a school year longer to improve their reading in 
comparison with their baseline cohorts. Naturally, the extra half year of school should cause the reading 
results from the midterm to be better than those at baseline, independent of the project’s impact. The 
evaluation team corrects for the different timing of the reading tests by adding half of the difference in 
reading scores between grades that are observed to the baseline. For example, the difference between the 
average reading score for CE1 and CE2 at baseline is 1.58. Thus, 0.79 is added to the baseline score for CE1 
to estimate what that average score might be if it were collected at the end of the school year in May or June 
instead of January or February. Likewise, the difference between CE2 and CM1 is 1.3, so 0.65 is added to 
grades for CE2. Because no data is available for CM2, for the CM1 baseline, the adjustment is made by 
adding 0.65 – (0.79-0.65) = 0.51 to baseline data for CM1. This attempts to project the decreasing trend in 
intra-grade reading improvements shown in the difference between reading improvements between CE1 and 
CE2 and then between CE2 and CM1. 
The report also looks at gender differences in literacy outcomes by comparing ASER scores for boys and girls 
in each grade. This comparison is made at baseline and at final evaluation to assess if gender differences in 
reading may have changed with the FFE program’s implementation. 
School feeding sustainability
The evaluation team identified the quantity of food and resources required to conduct a school feeding as 
defined by McGovern-Dole’s operational plan. The team valued products using market prices to estimate the 
resources a community would need to sustain the program after McGovern-Dole concludes. Then the team
identified the food and cash contribution to these requirements made by the school granaries, school farms 
and by the government. These results are used to identify how much of FFE’s meal provision can be 
expected to be replaced after the program ends. The evaluation team also made estimates based on a more 
optimistic scenario by extrapolating granary contribution by a factor of two and increasing school farm size to 
two hectares.
 
11 http://www.asercentre.org/
8
Limitations of the Evaluation
Statistical strength of the teacher survey: The teacher survey had 355 respondents, short of the target of 
404 required to be statistically representative at the 95 and 5 percent levels. The shortcoming may be 
explained by the occurrence of teacher strikes in some areas during the survey. 
School granary data incomplete: Complete school granary data was not available, which may affect the 
accuracy of school feeding program sustainability projections. 
Increased government support - This indicator is difficult to measure because teachers appear to receive 
pedagogical training through many channels, and it is unclear which qualify as “government support.”
Mid-term Survey Statistical Strength: The mid-term surveys for teachers, school directors, and school 
checklists were not statistically representative, which affects the validity of comparisons between those mid￾term and final evaluation survey results. 
Minimum acceptable diet measurements – Unlike the baseline, neither the midterm nor end of project 
evaluations included a survey of mothers. As a result, minimum acceptable diet measurements are only for 
food consumed at school and not at home. 
Findings
In this section, we describe a) the progress that the McGovern-Dole program has made toward strategic 
objectives; b) the program’s activity level results; c) the process of school selection; and d) McGovern-Dole’s 
monitoring and evaluation system. 
Strategic Objective Results
McGovern-Dole in Senegal is focused on achieving two strategic objectives. These are:
• SO1: Improved Literacy for School Aged Children
• SO2: Increased Use of Health and Dietary Practices
SO1: Improved Literacy of School-Aged Children
Under SO1, McGovern-Dole seeks to improve the literacy of children through achieving three results:
1. Improved student attentiveness through reduced short-term hunger 
2. Improved student attendance 
3. Increased government support
9
Improved Literacy
The final evaluation supported the 
mid-term evaluation’s finding that 
the McGovern-Dole program has 
improved student literacy. Following 
the methodology used in the 
baseline study, the evaluation team 
utilized the ASER literary 
assessment to compare students’ 
reading ability to baseline and mid￾term evaluation results. The 
assessment found that on average, 
students had improved in relative 
reading ability and proportion of 
students reading at grade level. 
The evaluation team attributes reading improvement to the program because the team could not come up 
with another reason, other than the program’s effectiveness, that school children across Saint Louis would
significantly improve at reading between the baseline study and final evaluation, holding grade level 
constant.12 Therefore, we believe it is fair to attribute the literacy improvement to the project .
Table 2 shows the average reading level that students achieved at baseline (using the adjusted baseline score),
the reading level measured at midterm, and the reading level measured at final evaluation for each grade. The 
table also shows the difference between reading levels at baseline and final evaluation, which describes 
students’ overall improvement. Reading scores improved for Grades 3, 4, and 5 students by 1.58, 1.5.and 1.5
levels respectively. 
The program appears to have experienced diminishing marginal utility between mid-term and final evaluation
as students’ reading levels did not improve during that period as they did during the first part of the program. 
The final evaluation showed a slightly lower score for grade 3 (CE1) students and slightly higher scores for 
grades 4 (CE2) and 5 (CM1) students. However, the final evaluation scores were not statistically different 
from the mid-term evaluation scores.
13
 
12 As a point of comparison, the St. Louis regional department of education measured a 4% increase in the proportion of 
CE1 students who read at grade level between 2016 and 2017. The MGD program saw over 82% increase between 
2015 and 2016. (Rapport Annuel De Performance (2017); Inspection d’Academie de Saint-Louis; March 2018. Pg 8. 
13 Two-sample t tests showed p values of 0.46 for CE 1, 0.53 for CE 2, and 0.15 for CM 1. 
EXHIBIT 2 BLACKBOARD SHOWING LESSONS IN FRENCH AND 
ARABIC
10
Table 2 Improvement in reading level by grade
GRADE ADJ. 
BASELINE**
MIDTERM FINAL 
EVALUATION
BASELINE/FINAL 
EVALUATION
DIFFERENCE
GRADE 3 (CE1) 3.31 5.04 4.88 1.57
N 356 366 326
GRADE 4 (CE2) 4.89 6.25 6.39 1.5
N 341 379 278
GRADE 5 (CM1) 6.19 7.34 7.69 1.5
N 339 321 194
* statistically significant at p<0.05
The evaluation team also looked at cohort groups that participated in the program. The Baseline CE 1 group 
are students that were in CE 1 at the beginning of the program and now are in CM 1. The Baseline CE 2 
group was in CE 2 at baseline and CM 1 at midterm. The Baseline CM 1 group was in CM 1 at the time of 
the baseline study. 
Table 3 shows how cohort reading scores changed as they progressed through the McGovern-Dole program. 
Students that were in CE 1 at the beginning of the program had a mean reading score of 7.69 by the end of 
CM1. This achievement represents an entire level improvement over students who were in CM 1 at the 
beginning of the program. Students who were in CE 2 at the beginning of the project had a mean reading 
score of 7.34 at the end of the program or 0.6 levels above students in CM 1 at baseline. 
Table 3 Change in Cohort Reading Level
GRADE ADJUSTED 
BASELINE
MIDTERM FINAL 
EVALUATION
BASELINE CE1 4.1 6.25 7.69
BASELINE CE2 5.54 7.34
BASELINE CM1 6.7
Reading level results differ for boys and girls by grade. At baseline and mid-term, boys scored higher than 
girls on the reading assessment. This trend held at Final evaluation for CE1 students. However, for CE 2 and 
CM1 students, girls scored higher than boys. Most significantly, CM 1 girls scored 0.44 higher than boys. 
11
Figure 1 Reading Score by Gender (Average)
Table 4 Expected Reading Level by Grade
Grade Appropriate Reading Level
Grade 3 (CE1) 5 or higher
Grade 4 (CE2) 7 or higher
Grade 5 (CM1) 9 or higher
At baseline, only 33.13 percent of CE 1 students read at an appropriate level for their grade. At final 
evaluation, that number had increased to 53.67 percent. CE 2 students showed a marked improvement, 
increasing from 27.27 percent reading at or above grade level to 48.47 percent. CM 1 students also improved,
from 28.02 percent reading at or above grade level to 43.3 percent by final evaluation. The graphics below 
demonstrate reading level improvement. The black vertical line represents the appropriate reading level at 
each grade. The graphs illustrate how the proportion of students reading at or above level (to the right of the 
line) has increased. 
CE1 students show a decrease between mid-term and final evaluation in the proportion of students reading at 
or above reading level. CM1 students show an improvement. For CM1 students, the improvement makes 
sense, as those students had participated in the program longest of all cohort groups. 
0
1
2
3
4
5
6
7
8
9
Baseline Midterm Endline Baseline Midterm Endline Baseline Midterm Endline
grade 3 (CE1) grade 4 (CE2) grade 5 (CM1)
Male Female
Reading Level
12
Figure 2 Proportion of Students Reading At or Above Level By Grade
In fact, when looking at the distribution of CM1 reading scores, not only did the proportion of students 
reading at or above level improve, but that proportion improved most at the highest reading levels. Figure 3
below shows the percentage of CM1 students that scored at each ASER test reading level. The vertical line is 
at reading level 9, which is the appropriate reading level for CM1 students. The graph shows that a higher 
proportion of CM1 students reading at or above grade level after the project than at baseline. It also shows 
two important findings: an increase in the number of students reading near grade level, and a large increase in 
students who are reading at the highest levels. The percentage of students reading at level 10 increased from 
13.1 percent at baseline to 17.5 percent at final evaluation, and students reading at level 11 increased from 
8.21 percent to 16.5 percent. The data also shows a general shift to higher reading levels. At baseline, about 
20 percent of CM1 students read at levels 0 – 2, while at final evaluation that proportion was approximately 
11 percent.
0
10
20
30
40
50
60
70
Baseline Mid-term Endline
Percentage of Students
CE 1 CE2 CM1
13
Figure 3 CM1 Reading Level Distribution
CE 2 showed a similar trend, as demonstrated in figure 4. The proportion reading at or above level 7 
increased between baseline and final evaluation. At baseline, 40.48 percent of CE2 students were reading at 
level 4 or below. By final evaluation, that number had decreased to 27.44 percent. 
Figure 4 CE2 Reading Level Distribution
Figure 5 shows that by the end of the project there nearly 13 percent fewer students reading at levels 0 and 1 
compared to baseline. Nearly 19 percent more students were reading at or above level 7 compared to at 
baseline. 
0
5
10
15
20
25
30
0 1 2 3 4 5 6 7 8 9 10 11
Baseline Endline
Percentage of Students
0
5
10
15
20
25
30
35
0 1 2 3 4 5 6 7 8 9 10 11
Baseline Endline
Percentage of Students
14
Figure 5 CE1 Reading Level Distribution
Reduced Short-Term Hunger
The McGovern-Dole program is reducing hunger for its beneficiary students. Senegal is a food insecure 
country with 53 percent of Senegalese families report that there have been times in the past 12 months that 
they did not have enough money for food.
14
The program’s primary measure of short-term hunger is the percentage of students that report feeling hungry 
at school.
15 A section of the students’ survey asked students whether or not they became hungry during the 
school day. If a student responded that they became hungry during the school day, they were asked if they 
became hungry always, often, about half the time, or just from time to time. 
The baseline study found that 66 percent of students reported that they do not feel hungry at school. The 
final evaluation found this proportion of students had not changed significantly, with 65 percent indicating 
that they are not hungry during the school day. 
Though the question was not offered at baseline, the mid-term evaluation allowed students to respond to a 
survey question about short-term hunger that they “are not hungry at school, except for time to time.” The 
mid-term evaluation survey found that 89 percent report that they are only hungry “from time to time (figure 
6). At final evaluation, this figure had not changed significantly, with 87 percent choosing that answer. Boys 
and girls chose that answer in approximately equal proportions. 
At final evaluation, 21 percent of students said that they were hungry about half the time, an increase from 
4.3 percent at mid-term. However, the proportion of students who said that they were hungry “always” or 
“often” dropped from 33.54 percent at mid-term to 14.48 percent at final evaluation. 
 
14 2016 Gallup World Poll
15 See MGD Indictor 1.2.1
0
5
10
15
20
25
30
35
40
0 1 2 3 4 5 6 7 8 9 10 11
Baseline Endline
Percentage of Students
15
Figure 6 Frequency with which students are hungry during school
Eight-five percent of students said that they usually ate breakfast at school compared to 78 percent at mid￾term. Forty-five percent said they eat at least one other time at school, compared to 56 percent at baseline. 
At mid-term, 65 percent of students mentioned eating something before school without prompting from the 
mid-term evaluation interviewer. At final evaluation, 97 percent said that they eat before school. While the 
increase in numbers eating before school is encouraging, the evaluation team believes that the dramatic 
change may have to do with the timing of the survey. At mid-term, the survey was conducted during 
Ramadan, which was not the case at final evaluation. 
Relationship between reduced short-term hunger and literacy improvement
At the core of McGovern-Dole design is the assumption that less hungry children will perform better in 
school. We test this assumption using logistic regression. Controlling for location (commune), school student￾teacher ratio and gender, we find that students who report not being hungry during the school day16 are 61% 
more likely to read at or above their grade level than those who are hungry. This result is statistically 
significant at the P <.05 level. 
Improved Student Attendance
Student attendance has increased since the baseline study. As during the mid-term evaluation, teachers and 
school directors that the evaluation team interviewed indicated that more students were attending school and 
that students were staying at school longer during the day. Teachers said that students stay longer because 
they no longer have to return home to eat or use the latrine. A school director in Podor said that he had to 
hire three new teachers, bringing his teacher total to five, because of increased enrollment and attendance. 
Teachers and school directors say that students rarely miss school since the school feeding activity began, and 
they stay at school longer. Directors report growing class sizes and schools operating at capacity. 
 
16 Or only hungry from time to time. 
0
10
20
30
40
50
60
70
Always Often Half the time From time to time
Mid-term Endline
16
Counterpart maintains records on student attendance. The data that Counterpart collects comes from the 
attendance records of schools participating in the program. These records support the statements that school 
directors and teachers made regarding the increase in student attendance. 
That said, the evaluation team believes that data does not accurately reflect that true rate at which students 
attend school. The team has two primary reasons for this judgment. First, the unit of measurement that the 
school system uses to track enrollment seems difficult to track. For example, in the 2017/2018 school year, 
the maximum possible attendance by month was as follows: 
Table 5 Maximum Possible Student Attendance 2018
MONTH MAXIMUM 
POSSIBLE 
STUDENT 
ATTENDANCE
October 99
November 118
December 87
January 128
February 116
March 99
The unit of measurement is the hour, with Mondays, Wednesdays, and Fridays allocated five hours, and 
Tuesdays and Thursdays allocated seven hours. It seems unlikely that teachers and directors have the capacity 
to track hourly attendance for students, especially given the large class sizes and lack of administrative 
personnel at some schools.
Secondly, the data indicates that students attended school 99.5 percent of the time on average during the 
2017/2018 school year. As a point of comparison, a UNICEF survey conducted on school attendance rates 
worldwide calculated Senegal’s rate at 54 percent overall and 43 percent in rural areas.
17 Another point of 
comparison is the US, where school attendance averages about 92 percent.
18 The World Food Program 
estimates that primary school completion rate is 66.7 percent in Senegal.
19 The evaluation team believes that 
McGovern-Dole collected data on school attendance exaggerates student attendance rates. That said, survey 
results combined with interviews and McGovern-Dole’s food consumption records suggest that student 
attendance rates have increased and are high. 
Increased Government Support
Increased government support, defined as the percent of teachers who had received pedagogical training in 
the three months prior to the survey, had declined between baseline and mid-term. The indicator improved 
significantly between mid-term and final evaluation. At final evaluation, 33 percent of teachers reported that 
they had received pedagogical training in the three months previous to the survey compared to 13 percent at 
 
17 https://data.unicef.org/topic/education/primary-education/
18 https://nces.ed.gov/programs/digest/d11/tables/dt11_043.asp
19 https://docs.wfp.org/api/documents/WFP-0000023416/download/
17
mid-term. The evaluation team suspects that the 32 percent average understates actual training as teachers 
may have not included monthly peer to peer trainings that many attend. 
This indicator is difficult to measure because teachers appear to receive pedagogical training through many 
channels, and it is unclear which qualify as “government support.”
Additionally, the three-month requirement may not be most appropriate. As discussed in the mid-term 
report’s recommendations, McGovern-Dole should work with local government to determine how often 
trainings are required and to make sure delivering quality training at regular intervals is established among 
local education inspectors. 
SO2: Improve Health and Dietary Practices Results
Under SO2, McGovern-Dole seeks to improve health and dietary practices through achieving three results:
1. Increased percentage of children receiving a minimum acceptable diet
2. Increased knowledge of safe preparation and food storage practices
3. Improved knowledge among students of health and hygiene practices
Percent of children receiving a minimum acceptable diet at the school level
Dietary diversity is a proxy for adequate nutrient-density of foods, including micro-nutrients. Commonly, the 
minimum acceptable diet (MAD) measurement is one of eight core indicators for assessing infant and young 
child feeding (IYCF) practices as developed by the World Health Organization. This indicator ordinarily
reflects both the minimum dietary diversity (MDD) and minimum meal frequency (MMF) of a population. In 
the context of FFE Senegal, MAD is also important because undernourished kids have difficulty learning20
.
MGD adopted the MAD indicator to analyze if school-age program participants are eating sufficiently to
ensure appropriate growth and development. Without adequate diversity and meal frequency, school-age 
children are vulnerable to undernutrition, especially micronutrient deficiencies which increase chances of 
increased morbidity and mortality. The indicators itself measures which food groups (table 6) are being 
consumed with a 24-hour dietary recall. The data collection instruments asked children about their breakfast 
and lunch intake since this is what they ate as part of the school program. Minimum dietary diversity means 
child is eating from at least four food groups. The cut-off of four food groups is associated with better-quality 
diets.
21
 
20 In it’s more common usage, MAD is used for infants 6-23 months of age and the mother is usually asked in a 24 hour 
dietary recall.
21 In most studies of MAD, the data differentiates between Vitamin A rich fruits and vegetables and regular fruits and 
vegetables. This differentiation was not made in FFE Senegal data collection
18
Table 6 Minimum Acceptable Diet Food Groups
1. Grains or roots
2. Legumes and Nuts 
3. Dairy Products (milk, yogurt, cheese)
4. Flesh foods (Meat, Fish, and Poultry)
5. Eggs
6. Vegetables
7. Fruits
Other (a lot of bread and coffee) (does not count as a food group)
Source: “Indicators for Assessing Infant and Young Child Feeding Practices: Part I 
Definitions.” Washington, D.C. USA: World Health Organization (WHO), 2008. 
http://apps.who.int/iris/bitstream/10665/43895/1/9789241596664_eng.pdf.
The McGovern-Dole final evaluation indicates that 58 percent of the targeted Senegalese school-age children 
(both sexes) usually eat at least 4 of the seven food groups at school (figure 7) for the minimum acceptable 
diet. it is a significant increase over the mid-term evaluation’s result. Figure 7 shows the baseline results for
children’s diet at home at baseline versus at school for the midterm and at the end of the project. By 
comparison, the baseline result for children’s minimum acceptable diet at school was two percent.22
We calculate that 58 percent of students receive a minimum acceptable diet in school at the final evaluation, a 
substantial improvement over the mid-term evaluation’s 43 percent estimate. The final evaluation percentage 
is slightly higher for females (58.2%) compared to males (55.3%). 
Figure 7 Minimum Acceptable Diet
Increased Knowledge of Safe Food Preparation and Storage Practices
The strategic objective-level measure for this indicator is the percentage of beneficiaries who use appropriate 
hand washing practices before meals, before food prep, after using the latrine, and after diaper changing. The 
final evaluation calculated that 74 percent of beneficiaries use appropriate hand washing practices, a clear 
improvement over the baseline finding of 50 percent for males and 55 percent for females. The final 
 
22 Note that the midterms and endline evaluations did not include a mothers’ survey necessary to assess children’s diet at 
home
0
10
20
30
40
50
60
70
80
Baseline result Midterm result Endline result
19
evaluation average dropped somewhat from the mid-term overall average (77%). However, this result is 
within the mid-term’s margin of error. 
Figure 8 Percentage of Beneficiaries Using Appropriate Hand Washing Practices
Improved Knowledge of Health and Hygiene Practices
The measure of this indicator is the percentage of students that can name at least two ways to prevent 
intestinal worms (MGD Indicator Framework, Annex 7). The final evaluation showed a significant
improvement in this indicator over the baseline and mid-term. At baseline, 8 percent of boys and 10 percent 
of girls could identify at least two ways to prevent intestinal worms. At mid-term, 45 percent of boys and 41 
percent of girls could do the same. By final evaluation, students ability to identify two ways to prevent 
intestinal worms had improved to over 51 percent, with boys and girls scoring about equally (figure 9). 
Figure 9 Student knowledge of ways to prevent intestinal worms
0
10
20
30
40
50
60
70
80
90
Baseline Mid-Term Final
0
10
20
30
40
50
60
All Boys Girls
Baseline Mid-Term Final
20
Activity Level Results
The following section reports on final evaluation results for each McGovern-Dole program activity. This 
section compares final evaluation results to mid-term results for information that was collected during mid￾term and final evaluation but not at baseline, for example in reporting on the cleanliness of latrines that were 
not yet constructed at baseline. Please note that at mid-term, information collected on school facilities 
through the mid-term evaluation team’s checklist instrument, teacher surveys, and school director surveys had 
wider margins of error and lower levels of confidence than the final evaluation instruments. The section 
indicates the margin of error and level of confidence where necessary for appropriate comparison. 
School Selection
The McGovern-Dole program’s school selection process was transparently and efficiently managed. It 
worked to secure the participation of the school communities. While participation was slow to develop for 
many schools at the beginning of the program, McGovern-Dole staff took steps to deepen school community 
participation between mid-term and final evaluation. 
McGovern-Dole’s selection process focused on selecting schools that were the most likely to have success in 
the program as well as those with great need. Many very needy schools were included in the program. 
However, practical considerations led to a bias toward including schools that were more feasible to serve over 
schools that had the most needs. 
The McGovern-Dole program selected eligible schools based on six criteria: 
1. A minimum of 30 students.
2. A sufficient number of teachers given the number of students, though no specific ratio was given.
3. Low enrollment statistics, which were provided by the inspector. 
4. Accessibility, meaning within a reasonable distance to major towns and accessible by trucks. Schools 
that were only reachable by crossing a river were ineligible. 
5. A kitchen and food storage building. 
6. Schools with a community that has demonstrated a commitment and willingness to contribute to 
previous projects, such as the Development Activity Proposal project23; including contributions in 
cash or in-kind support. 
McGovern-Dole asked the school inspector’s office for a list of schools in Saint Louis that met these criteria. 
Then, Counterpart and School Inspector’s Office staff visited the schools to make sure the project could 
serve them. After visiting the schools, the Inspector’s Office and Counterpart realized that a number of the 
schools were not a good match for the program, so Counterpart asked the Inspector’s office to identify 
replacement schools. At the end of school selection, the inspector’s office initially identified 180 primary 
schools and 44 preschools that met the program’s criteria. The program chose to serve all 44 preschools and 
163 of the primary schools from this first round of school selection. Currently, the program supports 204 
primary schools and 66 pre-schools.24
As reported in the mid-term evaluation, the program changed two of its initial school selection criteria. The 
first was the requirement that schools have a canteen, as these were relatively rare in primary schools and 
prohibitively expensive for many schools to build. The second was the community commitment criterion. 
 
23 The Development Activity Proposal project was a school feeding project that came before MGD-Food for Education. 
24 Information on school selection derived from Counterpart staff interviews and school election documentation. 
21
Counterpart realized many schools needed more help 
rallying commitment than the program originally anticipated. 
In the mid-term evaluation, the evaluation team 
recommended that the McGovern-Dole program consider 
allocating resources to schools according to need, rather 
than following a one size fits all approach. In response to 
this recommendation, Counterpart segmented the schools 
selected for the project by each school’s level of food 
insecurity. All of the schools participating in the project were 
grouped into one of three categories as shown below.
Table 7 School Categorization
Category Percent
Low food insecurity 15%
Medium food insecurity 45%
High food insecurity 40%
This categorization of schools led to realization about the difference is the needs of high food insecurity 
schools versus the other categories, and the difficulties in maximizing benefit to the neediest schools. The 
most food insecure schools also had the most basic infrastructure needs. At these schools, temporary 
structures may serve as classrooms. They have no auxiliary structures, such as a barrier fence/wall, library or 
school director’s office. 
When asked “What else could the McGovern-Dole program provide your school to improve results?” school 
directors and teachers at low food insecurity schools replied that they needed “computers.” Low food 
insecurity schools also appeared to have greater community contributions. At the Mbilor primary school in 
Dagana, for example, the community had built classrooms with community contributions independent of the 
McGovern-Dole program. Two fathers that the evaluation team spoke with said that they had attended the 
school in 1978 and alumni had contributed to the school’s growth. 
One parent at a low food insecurity school said that she liked the program because “I used to have to give 
money (for food), and now I save money.” This comment expresses a much different type of need than is 
seen in the high food insecurity schools where parents and teachers commented that they liked the program 
because students were no longer hungry. 
The Regional School Inspector Office in Saint Louis mentioned that another way to categorize schools was 
to distinguish those in the diéri and those in the wallo, a factor that was considered in Counterparts 
classification system. The diéri is an area in Saint Louis that is outside of the rivers flood zone and is thus 
drier and more dependent on rainfall. Schools in the diéri are unlikely to qualify for a school farm and are 
generally in poorer communities. Many are also harder to reach. 
While school and student needs differed across McGovern-Dole’s school categories, students in all of the 
categories benefited. Students at schools in each category improved reading scores, nutrition, and hygiene. In 
interviews, school directors across the three categories said that attendance was up, hygiene had improved, 
and the students seemed healthier. 
Clearly, the McGovern-Dole program did not focus only “on selecting the schools that had the most needs,” 
although certainly a great many of the program’s schools fell into that category. The program focused on 
schools that had serious needs, that showed potential for achieving McGovern-Dole objectives and potential 
EXHIBIT 3 CONTRASTING SCHOOL BUILDINGS
22
to graduate and that were accessible for commodity delivery. The program also chose a portfolio of low and 
high food security schools that mitigated the risk of serving only the neediest schools. 
Activity 1: Capacity building at the local, regional, and national level
Counterpart International delivered trainings to 1,956 participants since 2016 (table 8). This number includes 
1,211 teachers, 447 school directors, 262 school management committee (SMC) and community members,
and 37 school inspectors and government officials.
Table 8 Training Dates, Themes, and Participants25
Date Title Purpose Teachers Directors Gov. Community 
members/
Parents
28 April – 2 May 
2016
Curriculum for Basic 
Education and 
innovative reading 
instruction.
Training of trainers in literacy 
curriculum. 
18
7-10 May 2016
10-13 May 2016
Curriculum for Basic 
Education and 
innovative reading 
instruction.
Teacher training basic 
education curriculum and 
innovative reading 
instruction.
827
24 – 25 Nov 2016
8-9 Dec 2016
10 – 13 Jan 2017
17 – 20 Jan 2017
CGE members 
training on 
governance and 
management.
270 256
14-15 March 2017
13-14 April 2017
Promote the 
presence of teachers
Improvement of literacy of 
school-aged children.
Promote teacher attendance.
4526
22-23 May 2017 Advocacy workshop 
on the regular 
holding of teaching 
trainings.
To identify the real obstacles 
to monthly teaching training 
and to find practical, 
sustainable solutions adapted 
to the realities of the field.
2 17 6
24– 25 January 2018 Training of trainers 
in innovative reading 
instruction
Strengthen the capacity of 
school inspectors in reading 
instruction.
14
7-9, 13-15, and 20-
23 February 2018
Teacher training in 
innovative reading 
instruction methods
Build the capacity of teachers 
in innovative reading 
instruction.
337 143
28,29 March 2018 Management of 
multi-grade classes
1727 5
 
25 Data provided by Counterpart International
26 Attendees included teachers, but school directors, education inspectors and local administrative authorities and 
representatives from the regional Pas’ Association. Participants include 5 teachers, 7 inspectors, 11 PAs members and 
local administrative authorities. Categories were not broken down in the training report. 
27 It is unclear from Counterpart’s report if attendees were school directors or teachers. 
23
Teacher Training
McGovern-Dole’s most common training focused on improving reading instruction. McGovern-Dole and 
the national and regional government education ministries designed training that conformed to the current 
National education instruction methodologies. The training also includes assessment tools and lesson plans 
that assist teachers in designing classes. McGovern-Dole tested the effectiveness of these trainings with pre￾and post-test surveys. Learners scored in the 40 percent range on the pre-test and in the low 90 percent range 
on posttests. 
New literacy teaching skills are formulated at the central government level and then disseminated throughout 
the educational system through a training-of-trainers model. McGovern-Dole worked with the Ministry of 
National Education’s Directorate of Elementary Teaching to revise Senegal’s recently established curriculum 
and design training tools around the curriculum. The Directorate then trained local school inspectors, who 
trained the teachers of targeted schools. The trainings covered the new literacy curriculum, the importance
of reading skills, reading assessment tools, curriculum and lesson plans.
During the final evaluation, the evaluation team asked teachers whether they had received training. Teachers 
indicated that they had received training in health and nutrition and improved techniques for teaching 
reading. Teachers either directly attended trainings or were trained by the school director or another teacher 
that had attended a training. Teachers easily discussed new tools they had learned for teaching. Particularly, 
they discussed methods for teaching the new government curriculum, which focused on teaching the five 
components of reading. One teacher said that this method made the learning process clearer for students. 
One new technique that teachers learned was a method that teaches reading by emphasizing the sound of the 
word rather than piecing letters together. Teachers also said that they now start teaching reading to younger 
children than before McGovern-Dole. 
In interviews, several teachers and directors reported report that they were very pleased with the new, more 
interactive teaching approach they had learned and felt it was more effective. The survey reveals that of the 
teachers receiving pedogeological training, 60.4% applied what the learned “always”, 23.8% “often”, 15.8% 
“sometimes” and 0% “never”. 
In addition to training in literacy instruction, McGovern-Dole held advocacy workshops and assisted school 
teachers in organizing monthly performance meetings which are mandatory in Senegal but were not held with 
regularity in Saint Louis. The monthly meetings allow teachers to discuss work issues, and also to deliver 
peer-to-peer trainings on improved teaching techniques. Counterpart worked with the Saint Louis Academic 
Inspector to form teacher units at schools across the region. The purpose of the units is to address classroom 
instruction issues and promote student performance. 
Director Training
School directors received training on managing the school feeding program. These trainings included proper 
food storage and granary maintenance, courses in training teachers about new reading instruction methods, 
training in the School Health and Nutrition Guide (SH&NG), school farm management, and canteen 
management. School directors that the evaluation team interviewed found the trainings useful. The trainings’ 
effectiveness is revealed in the high survey scores on knowledge of food storage practices and hygiene 
practices discussed in the Activity 6: Training in food preparation and storage practices section. 
24
Activity 2: Construction of school infrastructure
Table 9 Classroom Construction Targets and Progress
Construction of School Classrooms # of classrooms
LOP Target 43
Completed 30
In Progress 13
The McGovern-Dole Program aims to build classrooms for schools that have the greatest need. As indicated 
in the mid-term evaluation, schools in Saint Louis have a wide range of infrastructure needs. Some school 
buildings, usually those further from urban areas, suffer from lack of infrastructure maintenance. Many 
schools in more remote locations need complete construction of classrooms as they currently consist of a 
temporary structure built from scrap wood and metal with a thatched roof. 
As of the final evaluation, McGovern-Dole had not met its LOP target under this activity, having completed 
30 of the 43 classrooms indicated as the LOP target. Delays with a current contractor and the rebidding of 
the work caused delays in the first half of 2018, but McGovern-Dole staff indicate that the remaining 13 
classrooms will be complete before the program ends in September 2018. 
School directors whose schools had received new facilities were generally satisfied with their new facilities. 
One director said that the quality of the classroom and latrine were very good compared to the old ones that 
the new buildings replaced. There were however some problems with new construction. A school in Saint 
Louis mentioned that they had issues with their latrine not draining properly. Others who had not received 
new facilities either said that they were promised new facilities and they hadn’t been built yet or had questions 
about how schools were chosen for construction. One school director said that he thought it wasn’t “equal 
how classrooms and toilets are assigned.” The communication and construction issues that the program has 
faced were likely caused by the need to replace the program’s construction contractor. Counterpart 
International is planning to address these issues in the program’s final quarter. 
The mid-term evaluation team recommended that McGovern-Dole reallocate resources in this area to ensure 
that budget allocated for infrastructure was appropriate for wealthier and poorer school needs. However, by 
the mid-term evaluation, the program had already allocated its budget for this activity. In its response to the 
mid-term evaluation report’s recommendations, McGovern-Dole acknowledged the recommendation and 
indicated that a new infrastructure allocation policy would be put in place for future programs. 
Activity 3: Construction of latrines and water station systems
As part of developing the school feeding system, Counterpart builds latrines and water station systems. As of 
the final evaluation, Counterpart completed construction of latrines and water stations at 24 schools.
According to Counterpart staff, construction has started on the remaining 19 sites, and they expect to meet 
the activities goal before the project ends in September. 
Table 10 Progress toward School Latrine
and Water Station Construction Targets
Construction of School Latrines #
LOP Target 43
Completed 24
In Progress 19
25
One hundred and forty two of the 159 schools (89%) that the evaluation team visited have some kind of 
facility where students can relieve relieve themselves. This is about the same percentage as at mid-term 
(86.5%). Only 53 of them are connected to water. If a school has water, it is likely it is connected to the 
latrine: 42 of the schools that had a water connection also had a water connection for their toilet. One school 
director in Dagana that the evaluation team interviewed did complain that the lack of water for the latrine 
made the latrine “not very good.” A director at a school in Dagana with 53 enrolled students that did not 
have a latrine said that created a hygiene problem for the school.
Table 11 Schools with latrine and water 
infrastructure 
Status of school #
Schools with latrines 142
Connected to water 53
Water connected to latrine 42
The evaluation team graded school latrines as very clean, mainly clean, a little clean, or not clean. As shown in 
table 12, a majority of those latrines are at least mainly clean. Still, 40 percent of latrines were either “a little 
clean” or “not clean.” This percentage is excessively high. Students that find latrines unfit will still walk to 
unsafe outdoor areas to relieve themselves and possibly avoid coming to school or will leave school early.
Table 12 Cleanliness of School Latrines (n = 159)
How clean are the latrines? %
Very clean 21
Mainly clean 39
A little clean 30
Not clean 10
Since the issue was identified at the mid-term 
evaluation, Counterpart has taken steps to 
improve the cleanliness of latrines. Counterpart 
developed an infrastructure maintenance guide 
for use at the school level. The guide covered 
maintenance of classrooms, latrines, and other 
infrastructure. In 2017 they also carried out 
trainings in use of the guide. However, follow up 
visits revealed that latrines were still not up to 
standard. Counterpart has instructed its 
facilitators to continue to raise awareness at 
schools and in communities about the 
importance of maintaining latrines. 
The final evaluation found that 83 percent of 
schools that had latrines had separate facilities for girls and boys, which is an improvement over mid-term 
(75%). The improvement is the likely result of Counterpart’s efforts since the mid-term to establish proper 
signs on latrine doors. Counterpart worked with the Ministry of National Education to procure Female and 
Male signs for restroom doors and had them installed in early 2018. 
EXHIBIT 4 UNFINISHED LATRINE FOR BOYS AND 
GIRLS
26
Table 13 below shows structural conditions of latrines at mid-term and final evaluation. Both mid-term and 
final evaluation asked about community contribution to latrine maintenance. While we found that 44 percent 
of schools report community contribution to this activity, we also found through interviewing teachers and 
school directors that schools organize students to participate in this activity as well. School directors report 
that they divide the students up, generally into boys groups and girls groups, and the students clean the 
schools on a regular schedule. The evaluation team asked school directors how they decided which tasks the 
boys would do, and which tasks the girls would do. At one school, the school director said that some jobs 
were for boys and others for girls, and the girls cleaned the latrines. However, at every other school we 
interviewed, the director said that the boys and girls switched tasks during each session and that tasks were 
not assigned by gender. 
Table 13: Latrine's condition (mid-term n = 96, Final Evaluation n = 159)
Mid-term % Final %
Very well maintained 22 16
Well maintained 44 51
Somewhat maintained 29 30
Not maintained 4 4
Latrines available to boys and 
girls 75 83
Parents and community 
contribute to latrine 
maintenance 
40 44
School Directors that the evaluation team interviewed said that they were generally pleased with the new 
latrines. In interviews, beneficiaries highlighted the importance of access to latrines. However, there were 
issues at some locations. One school director in Dagana said the lack of a water connection to the latrine 
created a problem. Students said that before the program they relieved themselves outside, but now they use 
the latrine which is much better. Students also mentioned that having separate boy and girl latrines did make 
female students more comfortable. 
Activity 4: Equip schools with energy saving stoves, canteen equipment, and 
materials 
McGovern-Dole equips schools so that they have tools they need to prepare and serve meals to students. 
These tools include bowls and utensils to prepare and serve food, pots to cook the food, and an energy 
efficient stove. Eighty-four percent of schools that the evaluation team visited have energy efficient stoves 
compared to an estimated 58 percent at mid-term. The final evaluation also found that 84 percent of schools 
had sufficient utensils for cooking and eating. 
The development and distribution of the energy efficient stove was the most important and complicated 
element of this activity. Making an energy efficient stove available to the schools was important for two 
reasons. First, energy efficient stoves reduce the environmental impact of the school feeding program. 
Secondly, energy efficient stoves reduce the work volunteers, primarily women must perform to gather 
firewood for cooking the school lunches. 
27
Initially, the project intended to manufacture the stoves in cooperation with community members and Peace 
Corps volunteers. Counterpart brought in experts to research how to make a stove using local materials. After 
considerable effort, Counterpart concluded that the kind of stove that the Program could construct under its 
initial plan was not feasible. These types of stoves, those usable with large pots, broke down after a few 
months. As a next step, Counterpart researched firms that specialized in making energy efficient stoves. They 
found businesses with this specialty outside of the Saint Louis region. They met with three of the specialists 
and asked them to manufacture sample stoves. Once Counterpart staff verified that the samples worked, the 
team considered having communities make the stoves, but the communities’ manufacturing costs were too 
high. Counterpart then researched firms that could construct the stoves within the program’s budget. 
Following that step, Counterpart had volunteer cooks use the stoves for three weeks. After the test period, 
Counterpart collected their feedback. The cooks made several suggestions to improve the base models. For 
example, they suggested that a door with a latch be placed where they inserted the wood. The cooks also 
made recommendations for the placement of holes to speed up cooking and create less smoke. Using cooks’ 
recommendations, the samples were re-tested and 
cooks clearly preferred a stove manufactured by a
Senegalese firm. This stove incorporated all of their 
suggested adjustments. Counterpart distributed this 
final model to schools. At the final evaluation, 
enumerators observed stoves at 135 of the 159
schools that the evaluation team visited.
As at mid-term, the final evaluation team interviewed 
cooks about the energy efficient stoves. As at mid￾term, cooks say that its primary benefits are that it 
requires less wood, which means that they spend less 
time gathering the wood. It also shields the fire from 
the wind, and it is very heavy, making it difficult to 
steal or misuse. 
The mid-term evaluation identified three challenges 
with the stoves: 1) one stove per school was often not 
sufficient; 2) the stoves could not accommodate the various types of pots and pans that school cooks used to 
prepare meals; and 3) preschools did not use the stoves because of limited access to charcoal or wood. Some 
schools using the stove commented during final evaluation interviews that “one is not enough,” and that they 
“would like another stove,” even though it was not the perfect size for other cooking implements. 
Following recommendations made at mid-term, Counterpart procured an additional 204 stoves, 102 of which 
had been distributed to elementary schools, which allowed some schools to have two stoves. The second 
batch was due for distribution in late May, 2018.
Activity 5: Good health and Nutrition Practices
The good health and nutrition practices activity promotes the use of the School Health and Nutrition Guide 
(SH&NG). The guide aims to improve the health and hygiene practices of the school community including 
students, teachers, and directors. Counterpart promotes the guide by training school directors on SH&NG. 
The directors then train teachers on how to use the guides, and the teachers build lessons into the curriculum 
to improve students’ practices. 
Counterpart conducted trainings for the 204 McGovern-Dole primary school directors, leading to the training 
of 1,045 teachers. Table 14 below lists the trainings. Counterpart also trained 270 cooks who went on to train 
other volunteer cooks in their communities. 
FIGURE 5 AN FFE STOVE AND TRADITIONAL POT USED TO 
COOK SCHOOL LUNCH
28
Table 14 Health and Nutrition Trainings28
Date Title Directors Cooks Government
25-26 March Guidance for IEFs on 
training managers in the use 
of the Health / Nutrition / 
Environment Guide
7
15-20, 23-26 
April 2017
Training elementary school 
principals in using the 
Health / Nutrition / 
Environment guide
204
(Directors 
trained
1,045 teachers)
6 March 2018 Retraining of Cooks in 270 
School Canteens 
270
Counterpart also printed 3,500 copies of the SH&NG. At mid-term, teachers and directors acknowledged 
that they had received the guide, and most had read it. During teacher interviews, teachers could pull out the 
guide and show sections of it to evaluators upon request. The teachers and directors that evaluators asked 
about the guide found it useful and practical as an educational resource. 
Counterpart reports that the program will distribute the last 500 guides between June and September of 2018. 
The guide was also distributed to the Regional Training Center for Education Personnel in Saint Louis, and 
200 copies went to schools in the region not covered by the McGovern-Dole program. Counterpart is 
currently training new teachers and directors on the guide, and they will receive a copy for their classrooms. 
School directors trained teachers at their schools in hygiene and nutrition using the SH&NG. One director at 
a large school reported that he trained 13 teachers. 
Counterpart also conducted a “Hygiene Week” at each of the McGovern-Dole schools. Hygiene weeks were 
popular with parents and teachers. At the mid-term evaluation, the director of the Ehmane Racine Sy primary 
school said that hygiene weeks worked well. She had noticed that students hand washing had increased before 
meals and after using the bathroom as a result of hygiene week activities. 
Teacher Knowledge and Practices
The mid-term evaluation attempted to measure the degree to which teachers were themselves trained in 
nutrition and hygiene, and the degree to which they instruct students in nutrition and hygiene. The teacher 
results are shown in table 15.
 
28 Data taken from Counterpart training reports. 
29
Table 15 The degree to which teachers have received and use training (Mid-Term n =312, Final = 
356)
Mid-term % Final %
Teachers that have received 
nutrition training in the past 
3 months
33.87 13.88
Teachers who learned new 
tools and techniques to teach 
nutrition 
96.2 97.96
Teachers who use new 
nutrition tools and techniques 
"always" or "often"
73.53 62.50
Teachers who have received 
training on hygiene in the 
past 3 months
38.46 17.56
Teachers who learned new 
tools and techniques to teach 
hygiene
95.83 98.39
Teachers who use new tools 
and techniques always or 
often
81.74 89.16
Fewer teachers that responded to the final evaluation survey said that they had received training in nutrition 
or hygiene in the past three months. Teachers that had received training found it useful, saying that they 
frequently used the new tools and techniques that they learned. 
One reason for the low percentage of teachers that say they received pedagogy training in the last three 
months may be a lack of clarity in what officially constitutes “training.” Teachers may not have received 
training from a school inspector, but may have received peer to peer training through regular teacher 
meetings, or received in-school training from the school director. It is unclear whether teachers responding to 
this survey question counted these events as training. Another reason may be that, while training in the 
SH&NG was delivered to all teachers last year, this year it was only delivered to new teachers, and was not 
scheduled until after the final evaluation, resulting in one fewer training for most teachers scheduled within 
three months of the evaluation. 
Table 16 shows how often the teachers provide instruction to their students. Teachers more often teach 
about hygiene in their classes, with 99 percent saying that they teach it “usually.” Seventy-seven percent said 
they will teach about hygiene in the current week, indicating that it is a frequent topic of discussion. Nutrition 
is taught less frequently. Slightly over 83 percent of teachers say that they teach nutrition with 54 percent 
saying they will teach in the coming week. 
30
Table 16 Degree to which teachers instruct on nutrition and hygiene (n=312, N=356)
Mid￾Term
Final
% usually teach hygiene in their class 99.04 99.43
% usually teach about nutrition 88.14 83.29
% have or will teach hygiene in 
current week
76.28 77.33
% Have or will teach nutrition in 
current week 
54.49 54.11
When enumerators asked teachers if they had washed their hands the day of the survey, 94 percent said they 
had. Of the 94 percent that said they had washed their hands on the day of the survey, figure 10 shows the 
situations in which they had washed their hands that day. 
The most common answer at final evaluation “Before eating as at mid-term. The second most common 
response at mid-term and final evaluation was “after using the latrine.” However, while 44 percent of 
teachers gave this response at mid-term, only 29 percent gave the response at final evaluation. These 
percentages are surprisingly low given that the surveys were given in the middle of a school day. 
Figure 10 Situations when Teachers Washed Their Hands (%) (Multiple Responses Possible)
Student Knowledge and Practices
The final evaluation sought to measure the degree to which students engaged in good hygiene practices. The 
final evaluation used the mid-term team restructuring of the baseline survey evaluation questions to overcome 
some of the social desirability bias found in the baseline report. Rather than asking questions such as “did you 
wash your hands today before eating?” enumerators first asked students, “did you wash your hands today?” 
Of those that said yes, enumerators asked: “why did you wash your hands?” The intention of rewording the 
question was to remove prompts for when hands should be washed, and also lessen social desirability bias 
that comes when prompting specific instances of hand washing. 
0 10 20 30 40 50 60
After touching an animal
After changing a diaper
Before giving food to someone else
When you have earth on your hands
Before touching or preparing food
After touching something dirty
After using the latrine
Before eating
Mid-Term Final
31
When asked if they had washed their hands today, 94 percent still said that they had, not significantly fewer 
than at baseline (95.3 percent). As students who are regularly taught lessons on hygiene, this result probably 
still reflects bias as all students knew that they should have washed their hands at least once. Table 17 shows 
student explanations of why they had washed their hands. A majority of students said that they washed their 
hands “before eating” (67%). As in the mid-term, the second most common answer was “after using the 
latrine” (30%). 
Table 17 Student Situtions for washing hands (Mid-Term n = 1028, Final n = 787)
When did you wash your hands today? (multiple 
answers possible)
Mid￾term %
Final29
%
Before eating 75.58 67.33
After using the latrine 36.48 30.00
After touching something dirty 19.84 12.00
When you have earth on your hands 6.03 5.07
Before touching or preparing food 8.17 1.60
After touching an animal 5.16 1.07
Before giving food to someone else 13.33 0.13
After changing a diaper 2.92 0.07
Hygiene Infrastructure
One hundred and fifty-seven of the 159 schools of the schools that 
enumerators visited had a place to wash hands before eating. At mid￾term, many schools used buckets supplied by the program, and water 
brought from outside the school as hand washing stations. Between 
mid-term and final evaluation, Counterpart launched an initiative to 
place portable hand washing stations at each school. Counterpart 
collaborated with the School Medical Control Division (DCMS￾MEN) to design and procure one hand washing station for each 
school. The evaluation team saw these hand washing stations in 
action at the schools the team visited. 
Eighty-five percent of school’s hand washing areas were at least 
“mainly clean” at the final evaluation, which is a large improvement 
over the 46 percent judged clean at mid-term. Sixty-one percent had 
soap at final evaluation, another large improvement over the 40 
percent with soap at mid-term. Fifteen percent had a hand towel for 
drying. The gross numbers are provided in table 18 below. These 
improvements reflect the distribution of portable hand washing 
facilities that Counterpart distributed to schools (exhibit 14), and the new trainings provided to school 
directors and Counterpart’s facilitators. 
 
29 This excludes “Other” which accounts for 26.5 percent of responses. 
FIGURE 6 A MCGOVERN-DOLE 
PROVIDED HAND WASHING 
STATION WITH SOAP
32
Table 18 Access to Hand Watching Facilities (mid-term n = 96, final n = 159)
30
Hand washing 
facilities
Mid-term Final 
w/ place to wash 
hands before eating
93 158
w/sink 33 100
Very clean 13 37
Mainly clean 31 100
A little clean 21 19
Not clean 31 6
% w/soap 39 99
% w/hand towel 5 24
About the same percentage of schools at mid-term and final evaluation 
had a place to wash hands before eating (99% at final). However, the 
percentage of schools with a sink tripled between mid-term and final 
evaluation.
The final evaluation team witnessed school hand washing before lunch. Teacher’s took turn supervising the 
hand washing line. Hand washing was orderly, and students were diligent in doing it correctly.
Activity 6: Training in Food Preparation and Storage Practices
As a key part of the school feeding program, Counterpart works with schools to form school feeding 
management committees. McGovern-Dole’s facilitators work with these committees; comprised of teachers, 
parents, school administrators and other community members; to oversee the acquisition, storage, 
preparation, and delivery of food for the school feeding program. McGovern-Dole provides training to 
committee members. The two main types of training are the management of the food storage system, and 
skills in the sanitary preparation of nutritious school meals. 
Training Volunteer Cooks
McGovern-Dole provided training to 808 school cooks from 270 schools over the course of the program, 
about three cooks per school. The cooks then returned to their communities and train other cook volunteers. 
Through interviews, the evaluation team learned that school cooks conducted the trainings, and the 
volunteers that were trained valued the instruction. In interviews, cooks revealed that they learned how to 
combine foods to make meals more nutritious. They most enjoyed learning new recipes and about proper 
nutrition. They also learned to tone down the spiciness so that the food would be more palatable to children. 
Cooks that the evaluation team interviewed were proud of the cleanliness of the cooking area and of washing 
their hands before cooking. 
Some cooks expressed a need for more and better equipment. Cooks that did not have structures to cook in 
said that they needed a kitchen. One factor that makes a kitchen necessary is the high wind in many school 
 
30 Results statistically significant at final but not mid-term.
EXHIBIT 7 STUDENTS WASHING THEIR 
HANDS BEFORE LUNCH
33
locations. Cooks working in the open or in temporary shelters face safety issues as open flames that are used 
when cooking with traditional implements flare and lurch unpredictably. 
Another issue that cooks face is the additional burden of unpaid care work 
faced when they volunteer for the McGovern-Dole program. The cooks that 
participate in the program are all women who cook for the school in addition 
to the care work they do at home. In some communities, where there were few 
volunteers, the few women who volunteer end up cooking every day at the 
school. The evaluation team interviewed one volunteer cook who said: “it’s 
difficult to do this and my other work.” 
Storage and Inventory Management
School granaries were established in the second year of the program to store 
community food contributions for school canteens. Granaries are buildings in 
the community that are appropriate and available to store grain and other food 
for schools. Food contributions come from the community and may also come 
from other sources, including in at least one case a local military base. As of 
March 2018, 189 schools had served nearly 646 thousand meals to students 
from school granaries. The section on Activity 9: School farms, discusses the 
potential of contributions from granaries and school farms to sustain the 
McGovern-Dole program. 
School inventory management is handled efficiently. Table 19 shows the condition of food storage facilities at 
the schools we visited. 96.84 percent of the storage facilities had a secure roof, and 95.57 percent were clean 
and well maintained. Food storage security was at 91.77 percent a large improvement over mid-term, and 
13.29 percent had some pest damage, which is also a considerable improvement. 
Table 19 Condition of Food Storage Facilities % (Mid-Term n =111, Final N = 159)
Food storage Mid-term Final
% food storage areas 
clean
92.79 95.57
% food storage w/ 
sealed roof
97.3 96.84
% food storage secure 80.18 91.77
% food with signs of 
pest damage
20.72 13.29
When asked how food should be stored, directors knew that the area should be well aerated (83%) and clean 
(86%). Many also knew that the food should be stored a specific distance from the walls. Fewer mentioned 
specifics about the materials the storage facility should be made of or other factors, but given the other issues 
they manage, not having these specifics on the tip of their tongue is not surprising.
FIGURE 8 USING THE 
STOVE IN A 
TEMPORARY SHELTER
34
Table 20 Conditions for storage in warehouse 
How food should be stored in the 
warehouse? (Multiple answers possible)
Mid-term
%
Final %
Aerated 83.33 82.91
Adequate size 21.3 30.38
Clean 78.7 86.08
Safe 25.93 27.22
Accessible 31.48 20.89
Wall and cement floor 32.41 13.92
Distance between walls and food 57.4 27.21
Table 21 Director knowledge of food storage facility maintenance (mid-term n = 108, final n = 159)
How should the storage space be 
maintained? (Multiple answers possible) 
Mid￾Term %
Final %
Clean interior and exterior 72.22 84.18
Sweep the soil weekly 50.00 73.52
Clean the roof, walls, and pallets once a week 53.70 59.63
Clean the outside of the warehouse once a week 40.74 59.63
Remove weeds and trash that attract rats and insects 34.26 45.57
Other 25.93 25.95
A higher percentage of directors had a proper understanding of taking inventory at final than mid-term (table
22). They knew about the inventory stock card and forms and to compare the quantity present against the 
amount listed on the card and theoretical stock (79%). Most also knew about completing the inventory form 
(73%).
Table 22 Directors knowledge of inventory control (mid-term n = 108, Final = 160)
How to keep inventory (Multiple answers 
possible)
Mid-term Final
Compare the quantity inventoried with the quantity on 
the stock card
77.78 86.71
Record the date of the inventory and sign your initials 20.37 30.38
Compare the physical inventory to the theoretical stock 76.85 78.48
Report reasons for loss in case of discrepancies 17.59 6.33
Completing the inventory form 69.44 72.78
Make a Loss allowance 7.41 3.16
35
Similar to the mid-term measurement, directors had a good understanding of how to ensure food was sanitary 
before serving it to students, as shown in table 23.
Table 23 Directors knowledge of serving safe food
How to serve safe food (Multiple answers 
possible)
Mid-term Final
Wash hands with clean water and soap before preparing 
food or before eating
98.15 98.73
Protect food from flies, cockroaches, dust 69.44 69.62
Respect the shelf-life dates 23.15 24.05
Avoid food with mold 32.41 20.25
Avoid breaking the cold chain 13.89 3.16
All raw consumed food (fruits and vegetables) must be 
cleanly washed before consumption
45.37 57.59
School feeding management committees’ work with McGovern-Dole facilitators to track inventory, order 
more food when required, and solicit community contributions. Food generally arrives on time, except for a 
delivery due in January 2018, which was delayed due to customs clearance issues. This delay and 
Counterpart’s handling of it is discussed further in the section on Activity 8: School Feeding. Incoming and 
outgoing food is tracked adequately. Food committee members could easily demonstrate inventory tracking 
forms and knew the schedule. 
Activity 7: Provide Access to Health Interventions
Access to health interventions aimed to reduce absenteeism and increase student attentiveness through three 
activities; provision of vitamins, provision of de-worming medication, and the establishment and use of 
moringa plantations. 
Deworming Medication and Vitamin A Supplements
McGovern-Dole received and processed two shipments of de-worming medication and Vitamin A 
supplements. The first shipment was distributed to students in January 2017, when 44,095 students in all of 
the program’s 270 schools received deworming medication. The second shipment of the medication was 
divided into thirds: 1/3rd for distribution to McGovern-Dole schools in Saint Louis now, 1/3rd for 
distribution to other schools in Senegal now, and 1/3rd set aside for distribution in December 2017. Over 
the course of the program, 90,726 doses of deworming medication were distributed by the McGovern-Dole
program. 3,097 doses for teachers and 9,039 doses for volunteer cooks were also distributed. 10,931 doses of 
vitamin A supplements were distributed to preschool students. 
After Counterpart’s last deworming campaign, Counterpart interviewed school cooks to see if they had 
noticed any changes in student behavior. Cooks mentioned that students were eating more food after the 
campaign. A parent at a primary school said that his son had parasites before the program and no longer 
does. He also has a daughter who had no appetite before the program, and her appetite has grown. He 
attributed the improvement to the vitamin supplements but also, and more likely that they’ve reduced the 
spiciness of the food they cook at home, which they learned from the McGovern-Dole program. 
Counterpart’s Health officer mentioned that the biggest struggle in the deworming program was spreading 
correct information about how parasites are acquired. While students and parents had many correct ideas 
about how parasites were acquired and could be prevented, some myths persist. The most common myth that 
the evaluation team heard was that kids get parasites from eating too much sugar. 
36
Healthy Message Murals
The McGovern-Dole program promoted healthy behavior through mural 
competitions. Students submitted ideas for murals that promoted healthy 
messages, and the best one was selected to be painted on a wall at the 
student’s school, which was both an honor for the student and an 
effective means of communicating messages about healthy behavior to 
the community. Examples of mural topics include the importance of 
avoiding unclean water and the reasons it is necessary to keep common 
areas swept and clean. 
Moringa Plantations
The McGovern-Dole program piloted moringa plantations at schools as a 
way of promoting student health. Moringa trees grow nutritious leaves 
that can be ground into a powder and added to food. Counterpart
purchased 12,114 moringa plants over the course of the program. School 
teachers and directors demonstrated the plants to the evaluation team and 
could explain the benefits of the plant as well as the way it is processed 
and used in food. Counterpart reports that 43 of the 270 McGovern-Dole
schools regularly use moringa powder in school meals. 
The moringa plantation program produced some of the most 
questionable results of all of the program. Moringa trees are difficult to 
maintain in the Saint Louis region. They require rainfall to be productive where they could be of the most 
use. In years like FY2018, an unusually dry year, the trees are likely to die. Additionally, many schools lack 
fences or barrier walls, which exposes the trees to goats and other animals that feed on them. Lastly, in some 
areas where schools are located, the soil isn’t appropriate for moringa cultivation. 
Activity 8: The school feeding program 
School feeding is the core of McGovern-Dole. As reported at mid-term, McGovern-Dole Senegal’s school 
feeding program was highly successful. The program provided over 15 million meals over the course of the 
program, greatly exceeding its LOP target. From October 2017 to March 2018, the program motivated its 
beneficiary schools to provide an additional 645,841 meals without the assistance of USDA provided 
commodities. Stakeholders attribute the achievement of the program’s higher-level objectives, namely 
improved literacy and increased student attendance, to the school feeding activity.
The McGovern-Dole program’s primary activity is to provide food to schools in food insecure regions. 
Distributing the food through schools provides incentives for students to attend school and to make sure 
they have the energy required to be attentive and focus on their studies. Many of McGovern-Dole’s other 
activities are implemented in support of the school feeding program.
School farms and school granaries have been implemented through McGovern-Dole to replace McGovern￾Dole’s food donations to schools. The section on Activity 9, School Farms, attempts to address the question 
of feeding program sustainability and replacement of McGovern-Dole donations through farms and 
granaries.
FIGURE 9 TWO MURALS ON MCGOVERN￾DOLE SCHOOLS
37
Table 24 McGovern-Dole Meal Schedule
School Meal Portions
Preschool Breakfast 5 days a 
week
100 grams rice or maize meal; 20 grams 
lentils or peas; 15 grams vegetable oil
Primary 
school
Breakfast 5 days a 
week
100 grams rice or maize 
meal; 20 grams lentils or 
peas; 15 grams vegetable 
oil
Lunch 2 days a 
week
150 grams rice or maize 
meal; 25 grams lentils or 
peas; 20 grams vegetable 
oil
The program requires successful coordination of the following activities, some of which are discussed in 
more detail under their own activity heading below: 
• Coordination with government entities including local and regional inspectors and the Ministry of 
National Education
• The establishment of successful food storage facilities at each beneficiary school
• Training of school directors in food storage requirements
• Training of school committees in inventory management
• Secure and sanitary storage of food used in the program
• Oversight by McGovern-Dole facilitators
• The transfer of USDA commodities to school facilities
• The organization of volunteers to oversee and maintain food in storage
• The organization of local volunteers into groups of cooks that prepare food for the students
• Raising funds and food contributions from parents to complement the commodities provided 
through the program. 
Food Acquisition and Community Contribution
The school feeding program relies on USDA-provided commodities (Milled Rice, soy-fortified cornmeal, 
Peas, lentils and vegetable oil). Communities also contribute cash and locally produced food to compliment 
the commodities. As discussed under Activity 6 in this document, USDA’s commodities are well managed. 
The schools are consistently adequately supplied. Food loss is rigorously accounted for. School management 
volunteers demonstrated in interviews that they are well versed in proper storage and distribution protocol. 
School feeding requirements
Table 25 below shows the volume of food and cost of food required to provide meals for the school feeding 
activity. The table presents requirements per day and per year. It also shows the volume and cost of food per 
student and per school. Note that throughout we exclude food that is served with meals but not provided by 
MGD, including vegetables, salt, milk and animal protein. These supplementary foods are sometimes 
provided through contributions from parents, or in the case of vegetables, through school farms. 
Costs are estimated using the market price information (see footnotes to table 25 for details). The cost of 
food to provide preschool students (91 students per school on average) one meal a day for the whole school 
year is 950 thousand CFAs or just under 1,679 USD. The annual cost of providing 7 meals a week for a 
primary school is nearly 3 million CFA or 5,125 USD. 
38
Table 25 Food required to sustain the School Feeding Activity
Day Year31
Student School32 Student School
Preschool – Volume (KG)
Grain 0.1 9.1 18 1,638
Oil 0.02 1.8 3.6 328
Pulses 0.02 1.8 3.6 328
Primary school - Volume (KG)
Grain 0.25 50.6 28.8 5,846
Oil 0.04 7.1 4.1 840
Pulses 0.05 9.1 5.4 1096
Preschool – Cost (CFA)
Grain33 22 2,002 3,960 360,360
Oil34 28 2,548 5,040 458,640
Pulses35 8 728 1,440 131,040
Total 58 5,278 10,440 950,040
Primary school – Cost (CFA)
Grain 55 11,165 6,336 1,286,208
Oil 49 9.947 5,796 1,76,588
Pulses 18 3.654 2,160 438,480
Total 122 24,766 14,292 2,901,276
Granaries
From October 2017 to February 2018, 140 primary schools with granaries had used or still had in stock, on 
average, 673 kilograms of grain (mostly rice), 35 kilograms of oil, and 30 kilograms of pulses. In addition, they 
had spent over 51,000 CFAs (Approximately $90). These figures are significantly lower for the 49 preschools 
with granaries (see table 26 below). Note that these figures are not representative of the total stock 
accumulated or spent throughout the year as the school year runs from October to July, and food may come 
into the granary throughout the year. Also note that the program’s data showed opening stocks for October 
2017, but it is not clear when this stock was received and over how long a period it was accumulated. 
 
31 Based on 180 school days per year.
32 School estimates are based on the 91 average students per preschool and 203 per primary school found for FFE 
schools. 
33 The price used for maize is 220 CFA/KG. This is the midway point between the school sale price of 164 and the 
March 2018 retail price of 274 CFA/KG (Bulletin Mensuel d’Information sur les, Marchés Agricoles N°360 – Mars 
2018). This is an estimate of a wholesale price, due to schools buying in bulk.
34 The price for vegetable oil used is 1400/KG. (http://apanews.net/index.php/news/senegal-legere-baisse-du-prix￾du-litre-dhuile-vegetale-en-fevrier-2018).
35 Cowpea prices are used to cost pulses. The price used is 400 CFA/KG. This is slightly less than the reported February 
retail price of 410 to account for buying in bulk. 
39
Table 26 School Granary Inventory - Oct 2017 to Feb 2018 (KGs unless noted)
USE STOCK36 TOTAL
PRESCHOOL (49)
TOTAL GRAIN (RICE
IN PARENTHESIS)
94 
(76)
162 (157) 256 
(233)
OIL 5 9 13
PEAS 10 1 11
CASH (CFA) 8,772 7,013 15,786
PRIMARY SCHOOL (140)
TOTAL GRAIN (RICE) 352 
(287)
322 (314) 673 
(601)
OIL 20 15 35
PEAS 25 6 30
CASH (CFA) 29,517 21,886 51,403
Meal Preparation and Delivery 
McGovern-Dole’s facilitators work with schools to organize parents into school feeding management 
committees. These committees recruit mothers in the community to serve as volunteer cooks. The volunteer 
cooks serve in groups of two or more. Each group prepares lunch for the school for a week at a time on a 
rotating schedule. For example, for a school that has six groups of volunteer cooks, each group will prepare 
lunch for one week out of every six. 
Through interviews and focus groups with students, the final evaluation and mid-term evaluation teams
found that the food was popular with students and parents. Students often said that they prefer the food at 
school to what they ate at home. Lentils are particularly popular. Directors report that the popularity of the 
food has increased attendance considerably. Examples include the Ndiakhaye school, where enrollment has 
increased by 30 students, the Ndiawgone preschool, who said that enrollment has approximately doubled 
since before the program’s implementation, and the EE de Thiago school, where the director mentioned that 
his CE level classes went from 20 students to 47 students per class. One school director reported that he even 
had one student from Mauritania, which he attributed to the school feeding program. 
In addition to increased enrollment, the school feeding program has improved the quality of attendance. This 
improvement takes three forms. First, students from poor families would often eat very simple breakfast or 
no breakfast. These students now receive sufficient food at school, which improves their attentiveness. 
Parents, directors, and teachers reported that the students were healthier and now have more energy for 
learning. Additionally, students would often go home for lunch. Teachers report that few of these students 
would return for afternoon classes. Now, students have no reason to leave, so they are more likely to stay for 
afternoon classes. 
 
36 As of the end of February, 2018.
40
Building School Feeding Sustainability
The mid-term evaluation recommended that Counterpart change its formula for calculating how much 
donated food to distribute to each school. Counterpart distributed a specific quantity of food per student to 
each school. The mid-term evaluation pointed out that McGovern-Dole schools varied widely in the wealth 
of their communities and school needs. The mid-term evaluation recommended distributing food, and other 
resources, based on school need rather than taking a one size fits all approach. 
In response, Counterpart developed a system for making sure that schools with the greatest need got the 
most food, while schools with lesser need were encouraged to rely more on community resources (table 8, 
school selection section). Schools were categorized based on the score their assigned facilitator gave the 
school. Scores were calculated by indicating how the school ranked on 16 indicators grouped into 4 categories 
(table 27). The higher the score, the higher the level of the school’s food security. Schools scoring higher than 
8 were deemed as having “low food insecurity,” schools scoring between 4 and 8 were categorized “medium 
food insecurity, and schools lower than 4 had “high food insecurity. 
Table 27 Criteria for School Categorization
CRITERIA POINTS
Criterion Agricultural Perimeter POINTS
Community Fields Developed By Project 3
Community Field Developed By The Population 2
School Garden 1
No Agricultural Perimeter 0
Criteria Location And Economic Opportunities
POINTS
School Located In The Chief Town Of Region And 
Department (Strong Economic Opportunity) 3
School Located In The Chief Places Of Communes 
(Average Economic Opportunity) 2
School Located In Rural Areas And Easy To Access 
(Low Economic Opportunity) 1
School Located In Rural Areas And Difficult To 
Access (Little Economic Opportunity) 0
Criteria Contribution to the GRANARY POINTS
Strong Contribution In Loft With More Than 15 Bags 
Of 50kg 3
Average Contribution In Loft Between 10 And 15 
Bags Of 50kg 2
Low Contribution Attic With Less Than 10 Bags Of 
50kg 1
No Contribution 0
Criteria For Access To Irrigation Points
Located Less Than 1 Km From The River / Lake 3
Located 1 To 3 Km From The River / Lake 2
Located More Than 3 Km From The River 1
Located In The Diéry 0
41
In McGovern-Dole’s original work plan, the program was scheduled in its last year to reduce food distributed 
to school’s equivalent to one meal per week per student every quarter. As the final year approached, the 
program had two challenges to address. The first was how to change this original plan so that it worked 
toward sustainability for schools based on each school’s need rather than implementing an across the board 
reduction in food distribution. The second was that Counterpart’s commodities shipment for the last quarter 
of 2017 was delayed, so the program was restricted to distributing the food it had on hand from October 
through November. 
Counterpart addressed these challenges by asking schools to provide food without McGovern-Dole
assistance for a specific number of days during the school year based on the school’s categorization. High 
food insecurity schools were asked to provide 5 days of meals to students, medium insecurity schools were 
asked to provide 10 days of food to their students, and low food insecurity schools were asked to provide 15 
days of meals for their students. 
All categories of schools exceeded the targets that Counterpart set for them, demonstrating their ability to 
work toward program sustainability. Table 28 shows the achievement for each school group when 
McGovern-Dole served them based on school need and sustainability goals. Achievement occurred between 
January and the final evaluation. 
Table 28 Days providing meals without USDA assistance by school category
Category
Target 
number 
of days
Average 
number 
of days
Median
Minimum 
Number of 
days
Maximum 
Number of 
days
1
st
Quartile
3rd
Quartile
Low Food 
Insecurity 15 7.07 2 037 51 0 12
Medium 
Food 
Insecurity
10 9.59 5 038 74 0 12
High Food 
Insecurity 5 10.56 5 039 73 0 10.25
The results show a high level of variability within categories. For low food insecurity schools, half of the 
schools were unable to provide food for more than two days without using USDA commodities. For medium 
food insecurity schools, half of the schools could provide meals for five days, or half the target set by the 
program. 56% of the high food insecurity schools provided meals for their students for five days or less, less 
than a third of the target. 
However, schools that were motivated to provide meals for their students greatly exceeded the target that 
Counterpart set for them. Twenty-seven high food insecurity schools (more than doubled the target number 
of days set to provide meals without USDA assistance, and at the highest level, those schools performed as 
well or better than the other two categories. For the low food insecurity category, only 8 schools met or 
exceeded the target 15 days. Two of those schools more than doubled the target, with one feeding students 
 
37 11 low food insecurity schools did not provide meals without using USDA commodities. 
38 49 medium food insecurity schools did not provide meals without using USDA commodities
39 37 high food insecurity schools were unable to provide meals without using USDA commodities
42
for 34 days and the other for 51 days. The remaining 6 schools fed students for 15 – 22 days. Thirty-eight
medium food insecurity schools met the target, with 16 of those more than doubling the target number of 
days. 
Table 29 Number of meals provided without USDA assistance by School Category
Category
Target 
number 
of meal 
days
Average 
number of 
meals
Median
Minimum 
Number of 
meals
Maximum 
Number of 
meals
1
st
Quartile 3rd Quartile
Low Food 
Insecurity
5 1,218 175 040 10,578 0 1,558
Medium 
Food 
Insecurity
10 2,250 495 041 36,206 0 2,800
High Food 
Insecurity
15 1,642 349.5 042 24,371 0 1531.5
The number of days that a school can provide food for its students is related to the number of students 
enrolled in the school. Clearly, larger schools have to provide more food than smaller schools. When looking 
at the number of meals that schools in each category provided the sustainability picture becomes more 
surprising. Though medium and high food insecurity schools provided meals for a similar number of days on 
average, medium food insecurity schools provided many more meals per school, indicating that medium 
insecurity schools are larger than high food insecurity schools and low food insecurity schools. Medium 
schools provided an average of 235 meals per day while high insecurity schools provided 156 and low 
insecurity schools provided 173. In each category 50 percent of the schools provided a fraction of the average 
for the category, indicating that when the schools in the top 50 percent became motivated, they highly 
overperformed. For high food insecurity schools, a couple of highly motivated schools brought the average 
number of meals very high. However, high insecurity schools still have a median number of meals that 
outperforms low insecurity schools and is comparable to the apparently larger medium insecurity schools. 
Determining next steps toward sustainability should factor in the motivating forces that led some schools to 
perform well and others unable to perform at all. 
Activity 9: Establishing Community Farms
School farms were a pilot program of MGD. The pilot’s intention was to test how effectively farms managed 
on behalf of schools could generate food and revenue to support school feeding programs. McGovern-Dole
established twenty school farms near schools. To qualify for a school farm, schools had to procure at least 
half a hectare of arable land in their community, have access to water for irrigation, and demonstrate 
community enthusiasm to support the project through the provision of volunteer labor. 
 
40 11 low food insecurity schools did not provide meals without using USDA commodities. 
41 49 medium food insecurity schools did not provide meals without using USDA commodities
42 37 high food insecurity schools were unable to provide meals without using USDA commodities
43
In the first year of the program, MGD gave the 20 selected school farms pumps and other inputs, such as 
seeds and fertilizer. MGD along with the Ministry of Agriculture (MoA) also provided technical and 
managerial training to school farm presidents and helped to mobilize labor in communities to work on the 
farms. School farms were instructed to save part of their harvest each year to buy inputs for the following 
season. MGD and the MoA provided follow-up technical support and monitoring in subsequent seasons. 
Some school farms started production in 2016. The bulk, however, began producing in 2017. Table 30 below 
shows that in 2017, eleven farms produced rice on an average of half a hectare of land. Paddy (unmilled rice) 
yields were over 5 metric tons per hectare, exceeding the national average of 4.14. This national average, 
however, is deflated by low-yielding rainfed rice in the southern part of the country that accounts for nearly a 
third of national production.43
 According to the MoA staff, the yields observed on the MGD school farms 
were lower than on other farms in the same area. 
Profit from rice cultivation - rice kept or sold for use by the school canteen - was 32 percent. This figure is 
lower than the 39 percent that the Japanese International Cooperation Agency (JICA) calculated in 2013 for 
their study of rice production in Podor and Dagana.
44 This difference may result from the JICA report not 
having included the cost of “free” family labor provided for family farms. Labor appears to be a significant 
issue for school farms and is discussed further below. 
Nine school farms produced vegetables and fruit, including tomato, onion, okra, and watermelon. Vegetables 
and fruit were produced on an average of 0.39 hectares in 2017. In 2018, some school farms are also 
cultivating groundnuts. Profit on vegetable cultivation is also 32 percent with only 7 percent consumed by 
school canteens and the remaining 25 percent sold to benefit the canteen coffers. 
This year, CPI is encouraging school farms to produce multiple crops per year. The main rice season is from 
March to July while vegetables are mostly cultivated from September to February, which allows for the 
cultivation of both crops on the same land in the same year. Cultivating both rice and vegetables is a way to 
maximize the productivity of the school farm and provide the school canteen with fresh vegetables for a 
good portion of the school year. 
School farms make an important contribution to the schools that manage them and show potential to make a 
substantial contribute to the school feeding program’s sustainability. In order to achieve true sustainability
however, schools would have to expand the size of their farms. The last two columns of table 30 show 
projections for average school farm sizes of one and two hectares, each cultivating both rice and vegetables. 
Increasing average farm size to one hectare, cultivated with both rice and vegetables, would increase the total 
school farm profit by a factor of 4, from 95,480 to 403,184 CFA. Increasing to two hectares with both rice 
and vegetables would double this profit to over eight hundred thousand CFA. Through discussions with 
school farm presidents, the evaluation team found that in many cases land was available for farm expansion, 
and in fact, school farms had access to more land than they could cultivate. 
 
43 CGIAR. Ricepedia: Senegal. Accessed May 2018.
44 JICA. 2014. Project on Improvement of Rice Productivity for Irrigation Schemes in the Valley of Senegal in Republic 
of Senegal: Final Report
44
Table 30 School farm 2017 performance with one and two hectare projections45
2017
RICE 
OR 
VEG
2017
20 
FARM 
AVG.46
1 HECTARE
PROJECTION
2 HECTARES
PROJECTION
RICE
Farms Producing Rice (%) 100 58 100 100
Average Hectares Rice 0.51 0.29 1 2
Yield (Kg/Ha) 5,248 5,248 5,248 5,248
Production 2,650 1,534 5,248 10,496
Price (Cfa/Kg) 164 164 164 164
Profit (%) 32 32 32 32
Consumed (%) 29 29 29 29
Consumed - Paddy(Kg) 769 445 1,522 3,044
Consumed – Milled (Kg)47 515 298 1,522 2,039
Sold (%) 3 3 3 3
Value Of Sales (CFA) 13,039 7,549 25,820 51,640
Total Value Of Profit (CFA) 97,490 56,441 193,049 386,097
VEGETABLES48
Farms Producing Vegetables (%) 100 47 100 100
Average Hectares Vegetables 0.39 0.19 1 1
Yield (Kg/Ha) 3,998 3,998 3,998 3,998
Production 1,568 743 3,998 7,996
Price (Cfa/Kg) 164 164 164 164
Profit (%) 32 32 32 32
Consumed (%) 7 7 7 7
Consumed (Kg) 110 52 280 560
Sold (%) 25 25 25 25
Value Of Sales (CFA) 64,386 30,499 164,168 328,336
Total Value Of Profit (CFA) 82,414 39,039 210,135 420,270
TOTAL CASH INCOME (CFA) 77,425 38,048 189,988 379,976
TOTAL VALUE OF PROFIT 
(CFA)
179,904 95,480 403,184 806,367
However, expanding farms and producing higher yields faces two important constraints: lack of capital for 
inputs and labor constraints. For example, rice yields were only 3,292 KG/HA in 2016. Low yields may have 
 
45 Note that there is data for 11 rice producing farms and 9 vegetable producing farms. Only one farm, Themeye Thiago, 
produced both. 
46 This is the weighted averaged across the 20 school farms.
47 Conversion rate of 67% (CGIAR. Ricepedia: Senegal. Accessed May 2018.)
48 Tomato, onion, okra and watermelon. This year some farms report planting groundnuts too.
45
been the result of pumps that were too small and relied on costly fuel. Several school farm presidents noted 
the low output and high cost of fuel for pumps. One president said it was the reason he was not producing 
vegetables this season. Another noted that while they had more land available for cultivation, the pump 
constrained their ability to expand. CPI is aware of the pump problem and is in the process of replacing 
them. 
Access to capital is another constraint to school farms’ ability to cultivate multiple crops per year or expand 
their area under cultivation. In the program’s first year, school farms received free inputs including seeds and 
fertilizers from CPI. In the subsequent years, school farms were required to pay for inputs through savings 
from the previous year’s harvest. Because of pump issues, discussed above, the first year’s yields were low,
and therefore savings for 2017 inputs was also low. As a consequence, the area under rice cultivation per farm 
declined, and only one farm produced both rice and vegetables in 2017. 
Labor availability is another a constraint to farm production and expansion. Many schools have a difficult 
time mobilizing volunteer, productive labor. If school farms expand, the burden of mobilizing volunteer 
labor will increase. Some schools that have difficulty mobilizing sufficient volunteer labor either pay for labor
with rice at the end of the season or use a mix of free and remunerated labor. In interviews, school farm 
presidents said they had their first priority is to manage their own family farms, which likely represents the 
sentiment of many potential volunteers.
A Summary of Contributions to School Canteen Sustainability 
Table 31 below shows the potential
of school granaries and farms to 
replace the McGovern-Dole
donated food for school feeding. 
For primary schools, the table also 
includes the 375,000 CFA the 
government provides to schools 
with a canteen. Total 1 shows the 
actual total contribution based on 
2017/2018 data of school granary
and farm production. Totally 1
describes schools with both 
granaries and school farms. 
Total 2 shows the total 
contribution of granaries and farms 
if granary contribution were to 
double and school farms were to 
expand to two hectares on average. 
The total also includes the 
contribution of two-hectare farms producing both rice and vegetables. Like total 1, the second total is only 
for schools with both granaries and farms. 
Figure 11 The Amount of MGD Food that Is Currently and 
Potentially Covered by Granaries and School Farms
Preschool 2017/2018 Preschool 2 HA farm
 Granary, School Farm, Deficit
10.0%
10.9%
79.1%
19.9
%
87.2%
46
For preschools, current school 
farms and granaries could 
replace 20.9 percent of donated 
food, or one meal per week. The 
contribution of granaries and 
school farms is almost equal. If 
preschools could cultivate 2-
hectare farms, and the 
contribution from granaries was
to double (Total 2), all of the 
preschools’ food requirements 
would be met (see figure 11). 
Primary schools with granaries 
and school farms that receive 
government funds could
currently replace 25.5 percent of 
McGovern-Dole donated food. 
This contribution comes from
granaries (9%), school farms (3.6 %) and the government (12.9 %). If granary contribution is doubled and 
school farms size increased to 2 hectares (rice and vegetables), 59.4 of the food provided by FFE would be 
replaced (figure 12). 
Figure 12 Contributions to Primary School Canteen 
Sustainability
Primary school 2017/2018 Primary school 2 HA farm
Granary, School Farm, Deficit, Government
9.0%
9.0%
12.9%
69.1%
18.0%
28.6%
12.9%
40.6%
47
Table 31 Contribution to School Canteen Sustainability
 
49 The value of vegetables consumed from the school farm is excluded. 
Required 
for FFE
Granary Farm (2017)49 Farm (2 HA) Government Total 1
2017/18
Total 2
2 HA
Total % Total % Total % Total % % %
Preschool
Grain (KG) 1,638 256 15.6 298 18.2 2,039 124.5 33.8 155.8
Oil (KG) 328 13 4 4 7.9
Pulses (KG) 328 11 3.4 3.4 6.7
Cash (FCA) 15,786 1.7 38,048 4 379,976 40 5.7 44.3
Total value (FCA) 950,040 94,706 10 103,635 10.9 828,638 87.22 20.9 107.2
Primary school
Grain (KG) 5,846 673 11.5 298 5.1 3,044 52.1 16.6 57.9
Oil (KG) 840 35 4.2 4.2 8.3
Pulses (KG) 1,096 30 2.7 2.7 5.5
Cash (FCA) 51,403 1.8 38,048 1.3 379,976 13.1 375,000 12.9 3.1 29.6
Total value (FCA) 2,901,276 260,463 9 103,635 3.6 828,638 28.6 375,000 12.9 25.5 59.4
48
The evaluation team’s analysis shows that in a best-case scenario, preschools could meet their feeding 
requirements, but primary schools’ could only meet 60 percent of their feeding requirements. Most likely, not 
all schools will have both a granary and a well-functioning two-hectare farm. Therefore, many schools would 
have to find an alternative solution if McGovern-Dole in Saint Louis ends in September. 
The school farm pilot program and the establishment of granaries have the potential to make major 
contributions to a sustainable school canteen program. This opinion was also expressed in MoA expressed in 
interviews. Also, the Saint Louis regional school inspectors office stated that “generalizing granaries to the 
whole region” was among its hopes for building on the program’s success. 
However, with farms’ current limited size and planting schedule, which results in one crop sewn per year, 
school farms make only a minimal contribution to school canteen sustainability (<4% for primary schools). 
Access to three types of capital be addressed before the significant expansion of the school farm pilot project 
is considered: access to finance, labor, and management. 
Activity 10: Improved Teacher Attendance
Under this activity McGovern-Dole was tasked with promoting teacher attendance through the reduction of 
administrative duties. The activity requires organizing advocacy workshops that would result in the reduction 
in time and resources teachers required to collect their salaries. 
The McGovern-Dole Program conducted two advocacy workshops, both taking place before the mid-term 
evaluation. One workshop that hosted 25 participants from IEFs, the Regional Agency for Early Childhood 
Education, and one that hosted 45 teachers from Dagana and Podor. 
The evaluation team calculates 92.5 percent teacher attendance rate for 2018 through the end of April, based 
on Counterpart’s teacher attendance data. This figure seems reasonable considering that teachers often live 
close to the school’s they serve. As a point of comparison, teachers in the US are present for 94 percent of 
the school year.
50
 Also, in interviews, directors said that teachers arrive on schedule and on time. Another 
school director said that she had no problem with teacher attendance and “except for the strike, they are 
here.” Teachers said that they had no obstacles to attending work. 
Teacher attendance is a complicated issue in Senegal. During the final evaluation, teachers across the country 
were on strike to demand better working conditions and to protest what the teacher’s union saw as an overly 
strong response by the government to a previous teacher strike. Teacher strikes complicate education in 
Senegal. The Saint Louis School Inspectors office said that a primary school teachers’ strike in 2016 lasted 
three months. The next year, CM 2 (6th grade) students were behind, and it was difficult for the school system 
to provide resources for students to catch up to level. 
The evaluation team asked teachers whether they felt their administrative duties were onerous or whether 
they had any difficulty collecting their salary. None of the teachers that we interviewed indicated that they had 
any problem in these areas. 
The evaluation team also asked parents, students, teachers, and directors whether teacher attendance was a 
problem at their school. While teachers are absent from time to time, no interviewee indicated a concern 
about teacher attendance.
There are other issues that McGovern-Dole could focus on that might have a greater impact on teacher 
performance. For example, while teachers praised the school feeding activity, several mentioned that they 
 
50 https://www.nctq.org/dmsView/RollCall_TeacherAttendance - data from the 2012 – 2013 school year. 
49
would also like lunches provided for teachers. They also said they would prefer their own latrines rather than 
using the same latrines as the students. 
Activity 11: Parent Association Training
McGovern-Dole relies heavily on community volunteers. The program supports government created parent 
associations called School Management Committees, to build community ownership of the program, keep the 
program aware of community needs, and ensure program transparency. The Associations also serve as a 
means through which schools set parent contribution policy and communicate that policy to community 
members. 
PAs were formed in 2016. Between November 2016 and January 2017, 526 PA members received training in 
governance, leadership, and management. In later periods, SMC members were also trained in infrastructure 
maintenance
SMCs also serve as the contact point for directors to organize the community. At some wealthier schools, the 
SMCs raise funds to build classrooms, provide electricity, and assist the school director in making decisions. 
Monitoring and Evaluation of Program Activities
As reported in the mid-term report, McGovern-Dole appears to have an excellent and efficiently functioning 
monitoring and evaluation system. Facilitators work with community members, school directors, and teachers 
to collect data important to the program. This data includes all output level data reported in semi-annual 
progress reports. 
As at mid-term, the evaluation team met with the program’s monitoring and evaluation officer to test their 
system. The monitoring and evaluation system could quickly bring up information about specific students in 
randomly chosen schools. When the evaluation team visited schools, the information brought up in the 
monitoring and evaluation system proved accurate. The system had adequate data entry staff which allowed 
quick reporting on program activities. 
At mid-term, the evaluation team reported on the monitoring and evaluation system’s important role in fine￾tuning the project’s energy efficient stoves and serving the program’s management in monitoring the 
performance of staff. Since mid-term the monitoring and evaluation system proved its worth in responding to 
the mid-term evaluation’s recommendations. 
At mid-term, the evaluation team recommended that the McGovern-Dole program provide resources based 
on need rather than follow a one size fits all approach. Counterpart’s first step towards a needs-based 
approach was to use its monitoring and evaluation system to identify differences in school status, such as 
relative level of wealth, the stability of the local population, economic potential of the community, and needs. 
The program’s monitoring and evaluation staff developed a data analysis tool that took in information 
provided by school facilitators and created a school ranking based on level of food security. The development 
and implementation of the tool demonstrated the strength of the program’s monitoring and evaluation 
system. The tool was developed quickly and rolled out to facilitators who could swiftly understand its purpose 
and use it, demonstrating Counterpart’s commitment to learning and adapting to new lessons. 
At mid-term, the evaluation team recommended that three indicators merited reconsideration: 
1. 100 percent of school-age children receive a minimum acceptable diet at the school level.
2. 90 percent of schools have energy-saving stoves.
3. 20 percent increase in teacher attendance rates compared to baseline.
We note that USDA dropped the minimum acceptable diet indicator and the teacher attendance indicator 
from its current standard indicator definitions. 
50
The final evaluation brings forward three other recommendations regarding indicators:
Improve Student Attendance
The LOP target for this indicator is the number of students that regularly (80%) attend USDA supported 
classrooms, disaggregated by gender. McGovern-Dole Senegal’s monitoring and evaluation system adequately 
collects school attendance records for more than 45,000 students and enters the information into a central 
database where the monitoring and evaluation staff can analyze the data and uncover trends. However, the 
data collected through school records may not accurately reflect attendance rates. The high reported rates of 
attendance, including very high rates reported at baseline, and the insufficient administrative staff at schools 
makes the data seem suspect. Also, though records show an increase in student attendance increased of 2%, 
from 97% at baseline to over 99% by final evaluation, school teachers and directors report a much bigger 
increase over the course of the program. Much of this increase is due to kids attending school more regularly 
and staying at school longer during the day because they do not need to return home to eat. To truly 
understand the way the program influences school attendance, McGovern-Dole should have an independent 
research plan for studying the topic from the beginning of the program. The plan could include random spot 
checks once per quarter at a sample of schools, which would generate a clearer picture of the influence of 
McGovern-Dole on student attendance. 
Understanding children who do not attend school
McGovern-Dole works to overcome barriers to school attendance and measures the benefits of the program 
for children who regularly attend McGovern-Dole schools. McGovern-Dole should also consider researching 
reasons that some kids do not attend school despite the school feeding incentive. Collecting data on this 
population would lead to innovative activities that reduce barriers to attending school. 
Understanding the status of children who are not in school would also provide a clearer picture of 
McGovern-Dole’s impact. For example, the final evaluation team believes it is possible that boy students who 
become frustrated with learning to read are more likely to drop out of school than girls. Using McGovern￾Dole’s current data collection methodology, only boys in school take the reading test conducted by the 
external evaluator, so boy’s average reading scores may be inflated by the non-participation of boys who are 
poor at reading and have dropped out of school. 
Improve the collection of the Minimum Acceptable Diet (MAD) indicator. 
The program’s measurement of the percent of school-age children receiving a minimum acceptable diet at the 
school level differs from the definition used commonly by international health organizations. The indicator 
could be improved by using 24-hour recall consistently and ensuring that the survey uses common definitions 
of food groups: 1. Grains, roots and tubers, 2. Legumes and nuts, 3. Dairy products (milk, yogurt, and 
cheese), 4. Flesh foods (meat, fish, and poultry), 5. Eggs, 6. Vitamin A rich fruits and vegetables and 7. Other 
fruits and vegetables.
Evaluation Questions
Implementation and Relevance
Implementation compared to design: The McGovern-Dole program’s design is its strongest characteristic. 
The best source of evidence of the program’s success in design and implementation is the results it has 
achieved in student attendance, student literacy, and stakeholder’s improved knowledge of nutrition and 
51
hygiene. As discussed in the mid-term evaluation report, the program creates a holistic system using improved
health and nutrition to increase literacy. Rather than simply delivering food to schools and making demands 
on teachers, the program creates structures to manage various aspects of the program. McGovern-Dole sees 
that schools are prepared to receive, store, and prepare food and that they contribute to the program to 
ensure local ownership over achievement and outcomes. The program also realizes that if people are going to 
eat, they will also need a latrine to relieve themselves including the sanitation infrastructure that goes along 
with latrines. Program interventions are designed to minimize the impact of new activities that fuel demand 
on the local environment. Teachers and school directors praised the high quality, practical, and professional 
training that they received. Parents recognized positive changes in their children. 
While Counterpart has implemented the program as they have designed it, the organization’s management has 
also responded flexibly when activities have required adjustment. Since the mid-term evaluation, Counterpart 
has experimented with a needs-based approach to resource delivery where possible, including commodity 
distribution and re-allocation of energy efficient stoves. The program has also assessed how it can more 
effectively engage with communities to improve hygiene practices and school maintenance. The development 
of the energy efficient stove presented in the report and at mid-term is another example of the Counterpart’s 
flexibility and willingness to learn.
As discussed at mid-term, Counterpart’s systematic vision includes exercising restraint in activity 
implementation where there are no options for sustaining the activity after the program ends. Thus, 
McGovern-Dole does not provide soap to schools for handwashing because schools could not provide the 
soap themselves if the program were to end. Similarly, toilets do not include water connections where 
communities would be unable to maintain them. Counterpart’s judgment in these program components is 
that it is not beneficial to the community to set up unsustainable assets even if the assets might produce 
short-term benefits. The evaluation team believes this judgment is correct, with the caveat that Counterpart 
should be careful not to use “unsustainability” as an excuse to not looking for creative solutions to difficult 
problems. 
The evaluation team also noted that Counterpart’s inclusion of innovative pilot projects in the overall 
program was effective. Pilots include the establishment of school farms, the provision of energy efficient 
stoves and the planting moringa plantations. Though the pilot projects achieved varying levels of success, 
McGovern-Dole learned a great deal from them that will assist future program. Importantly, pilot projects
appropriately balanced the potential payoff of success with the opportunity cost of administering a project 
that failed. 
Design compared to needs of beneficiaries: As reported in the mid-term evaluation, the program design 
reflected the needs of beneficiaries in most important respects. As the projects results at mid-term prove, 
delivery of food matched with appropriate training does deliver the program’s intended results. The 
Senegalese government, which is used to responding to requests for assistance, has benefited from the 
demonstration of an organization taking the lead to organize communities and achieve goals. McGovern￾Dole has taught teachers and directors how to organize and leverage resources in their communities, and
many have adopted those practices. Additionally, McGovern-Dole has advanced GOS reading objectives by 
providing trainings so teachers could more effectively teach the Government’s new reading curriculum. 
As reported at mid-term, there were a few activities that were not as relevant to stakeholders as others. These 
included:
• Improvements in teacher attendance –It is unclear that teacher attendance in McGovern-Dole
schools is a significant problem or that the programs identification of administrative burdens that the 
program identified are truly an obstacle to teacher attendance or effectiveness. 
52
• School Construction – As noted in the mid-term evaluation, the school construction element of the 
program should be re-oriented to the specific needs of schools, ranging from patching up roofs to 
complete classroom construction. The evaluation team notes that the program’s construction budget 
was allocated by the mid-term evaluation, so no changes could be made in response to this 
recommendation. 
Interestingly, Counterpart’s experiment with allocating food based on school need revealed that some school 
needs are not a factor of community wealth or local food security. As discussed under Activity 8: School 
Feeding, many high food insecurity schools outperformed many low food-insecurity schools in providing 
student meals without the assistance of USDA. Some medium food insecurity schools provided school meals 
self-sufficiently despite the large student populations that those schools were feeding. Clearly, there is 
something that motivates some schools to perform at a high level when challenged to do so, while many 
others sink into complete inactivity. This factor appears to be independent of the relative wealth of the 
school’s community, or possibly has an inverse relationship to local wealth. 
Program relevance given the economic, cultural, gender, and political context: The program provides 
food and education in a region that is short on both assets. The program is in line with the national 
government’s health, nutrition, and education plans. Counterpart’s leadership has coordinated with 
government and political entities to ensure support for the project. In fact, the GOS has followed up on the 
success of McGovern-Dole school feeding with funding allocated for school feeding for schools with existing 
canteens. However as noted above in table 31, this is likely to provide about 13 percent of school feeding 
needs. Going forward the balance will have to come from community contributions including through school 
granaries and farms. Communities are in many cases willing to contribute labor, time and money for the 
school feeding program. However, school feeding self-sufficiency could have been accelerated with more 
emphasis from inception of the three-year program.
The mid-term evaluation noted that one area where the project’s relevance requires further research is in its 
handling of gender issues. Counterpart has taken steps by initiating a study into gender issues that is still in 
draft form. At final evaluation, the evaluation team found that the way gender interacts with education for 
McGovern-Dole students is still unclear. As noted, the program’s reliance on volunteers exacerbates the 
unpaid care work burden for women in McGovern-Dole communities. The final evaluation supported the 
mid-term’s estimate that more girls (60%) than boys (40%) attend school. However, the imbalance is not true 
for all schools. Where the imbalance existed, explanations included that boys were required at home to help 
breed animals, that a national campaign to promote education for girls meant that boys felt it wasn’t as 
important for them, and that boys were more likely to get frustrated and drop out of school when they 
struggled with reading or math. 
Effectiveness and Performance
Changes in student enrolment, attendance and drop-out rates since the start of the project: As 
mentioned in the SO1 section of this document, enrollment, and attendance have increased. 
Increasing student literacy outcomes: As noted in the SO1 section of this document, student literacy has 
significantly improved since baseline, and that students’ improvement between baseline and mid-term was 
much greater than between mid-term and final evaluation. 
53
Improved student knowledge and behavior about nutrition, health, and sanitation: Please see SO2: 
Improve health and dietary practices results and Activity 5: Good health and Nutrition Practices. 
Teacher and mothers’ knowledge and behavior about nutrition, health, and sanitation: This document 
addresses teacher knowledge and behavior under Activity 5: Good health and Nutrition Practices. To the 
extent that the McGovern-Dole program addresses mother’s behavior, it does so in mother’s capacity as 
volunteer cooks or as part of SMC which this document addresses under Activity 6: Training in Food 
Preparation and Storage. Otherwise, the FFE program does not serve mothers as a direct beneficiary group. 
Impact of new infrastructure on school enrollment and attendance: The evaluation found limited 
information that supports the idea that school enrollment or attendance is influenced by new infrastructure. 
The evaluation team encountered one parent who said that she had stopped sending her child to the school 
because she thought the structure was unsafe, and teachers at two schools that said students were more likely 
to attend in inclement weather because of new facilities. 
Beneficiaries appreciated new facilities, and particularly connections to water, new classrooms, and useable 
latrines. However, at mid-term and final evaluation, when the evaluation team asked teachers and school 
directors if enrollment or attendance had changed, nearly all replied that it had increased and attributed the 
increase to the school feeding program. Beneficiaries attribute increased enrollment, increased attendance, 
and students spending more hours per day at school to the school feeding program. 
Improvement in reading fluency and comprehension skills due to new/renovated school 
infrastructure: Too few schools benefited completed or renovated construction at the time of the final 
evaluation to know whether infrastructure influenced reading achievement. Also, unless the construction was 
allocated randomly, too many other factors could have influenced student’s improvement in reading skills. 
Efficiency
Could other implementation strategies have achieved more with the same resources? Counterpart’s 
overall implementation strategy is effective and efficient. Counterpart does an excellent job of ensuring that 
stakeholders at all levels are engaged and informed and that the program is implemented with a complete 
school feeding system strategy. 
Counterpart’s adjustment to allocating resources based on need shows that there are other strategies that 
might achieve more than the McGovern-Dole program in its original design. These strategies include:
• Phase in sustainability measures earlier in the project: When Counterpart challenged schools to 
provide food without the assistance of USDA, many schools stepped up to the challenge and 
exceeded targets. If initiated earlier in the program, these schools may have served as models for 
other schools that had difficulty mobilizing. 
Additionally, working with schools and communities to establish and work toward canteen 
sustainability should start in year one. A crucial part of this is to work with schools to calculate the 
quantity of food they need to achieve sustainability and setting specific targets to achieve that 
sustainability within a reasonable time frame. 
• Taking a needs-based approach: In response to the mid-term evaluation, Counterpart began 
allocating resources according to need rather than quantity per student. The evaluation team believes 
this is a more effective means of achieving McGovern-Dole results. As Counterpart’s foray into 
needs-based activities demonstrated, not all needs are a function of relative wealth. Some low food 
insecurity schools would have benefited from more assistance motivating community engagement for 
example. Counterpart made important strides in the redirection of the program’s approach, and the 
evaluation team believes it should be carried into future phases of the program. 
54
• An Alternative: Implementing fewer activities: The evaluation team believes that activities related 
to school feeding and teacher training make the greatest contribution to McGovern-Dole’s strategic 
objectives. An alternative strategy could be to focus on those activities that were related to the 
provision of food, the preparation of food, organizing communities in managing and sustain the 
feeding program, and training teachers in innovative reading, nutrition, and hygiene instructional 
methodologies. In light of the difficulties faced by the construction activities and other activities not 
directly related to school feeding or instruction, focusing resources on activities that appear to be the 
most effective may have allowed more schools to participate and permitted greater achievement in 
current schools. Focusing on the program’s core activities rather than school infrastructure,51
McGovern-Dole Senegal could have served approximately 100 additional schools and an estimated 
17,600 students.
52
Sustainability
Steps/actions required to realize full sustainability of activities beyond the life of project: At the 
highest level, ensuring sustainability requires improved transparency among USDA, the Senegalese 
government, and Counterpart International. Government officials at the ministry of school feeding said that 
their mission is to take over the school feeding program and make it sustainable. They are currently 
formulating policies and developing a monitoring system to make that happen. At the local level, regional 
ministries of education and school inspectors’ offices must take over the duties of Counterpart’s facilitators to 
continue to drive community engagement and continuous improvement among school directors, teachers,
and students.
A second step involves training directors to create a detailed budget of the school feeding program’s items 
and costs, and develop a plan for working with their communities to fund that budget. During final 
evaluation interviews, few directors could articulate the programs costs and describe how they could sustain 
school feeding in a realistic way. School feeding programs are funded currently through USDA donations, 
granaries, school farms (in some cases), and government allocations. In the absence of USDA donations, the 
other categories will have to expand to sustain the program, or new revenue-generating activities will have to 
be developed. 
As mentioned at mid-term, the workload that the program creates for school teachers, directors, and female 
volunteers threatens the program’s sustainability. The program should measure the work burden that the 
program produces for women and school directors, and identify means of reducing the work burden or 
otherwise compensating workers. This could involve encouraging men in communities to devote more time 
to volunteering in addition to making cash or food contributions. One possibility might involve encouraging 
men to provide cooking fuel to schools to reduce women’s burden of collecting firewood. Another might be 
for men to work more closely with school directors to more easily oversee aspects of the school feeding 
program, so it does not interfere with their other school management duties. 
A final important step involves planning for the sustainable management of the volunteer component of the 
program. Currently, volunteers run most aspects of the program. They do so in close coordination with 
Counterpart’s facilitators. Even with the facilitator’s regular oversight, school results in terms of food storage 
and preparation, acquisition of community contributions, and facilities management are less than perfect. 
There is a strong possibility that the Senegalese government entity that assumes ownership of the project will 
 
51 School infrastructure refers to the construction of classrooms. It does not include latrines or water stations. 
52 The estimate divides the school infrastructure budget for 2018 but the per school cost of the school feeding program, 
stoves and equipment for cooking and serving students, training in food preparation and storage, good health and 
nutrition practices initiatives, and warehousing costs. 
55
not have Counterpart’s expertise in volunteer management and underestimates its importance to the program. 
As part of its transition plan, USDA should conduct a cost-benefit analysis for the Senegalese government to 
calculate whether maintaining the program as largely volunteer-run is worthwhile given the losses and quality 
control difficulties it implies, or if the government should offer communities and school directors incentives 
to assume the program’s duties. 
The challenges and successes of the project activities: As discussed in the sections covering SO1 and 
SO2, the program has made great strides toward improving student literacy and improving health and dietary 
practices. Reading scores show that literacy has improved greatly, and interviews with parents and school 
personnel reveal that children are healthier, cleaner, and more energetic. 
The program’s challenge is revealed by the diminishing returns it demonstrated in the period between mid￾term and final evaluation. While students reading has improved, a large portion of students still have not
achieved grade level.
53
Future iterations of the program will have to develop new activities or partner with organizations specializing 
in literacy or hygiene initiatives to create significant beneficiary improvement. These initiatives could include 
maternal literacy, supply more books to schools and communities, or bringing other more advanced literacy 
and hygiene programs to communities. 
The challenges and successes of project activities are addressed throughout the Findings section of the 
document. 
Lessons for future sustainability: The strongest direct lesson is that to achieve sustainability, schools 
require a relatively large source of food or income, like from 1 to 2-hectare farms, increased granary 
inventory, and increased government assistance to achieve school feeding program sustainability. The details 
of this calculation are covered under Activity 9. 
How the program incorporates lessons learned from the Matam McGovern-Dole program exit 
strategy: The evaluation team understands that the Matam MGD program did not have an exit strategy; that 
it ended abruptly. When the program ended in Matam, the school feeding program also ended at many 
schools, though it continued at many more preschools than primary schools. Where the program was 
terminated, it was for several reasons, including54: 
• Lack of will in the communities.
• Directors that preferred to focus on technical issues, like pedagogical training, rather than managing 
the school feeding program.
• Disengagement by the government precipitated by the lack of notice that the program was coming to 
an end.
Schools that had school feeding programs that persisted exhibited the following characteristics:
• An innovative culture among teachers that worked to make the most of resources on hand. 
• Consistent cash contributions from a community of parents eager to continue the program.
• Canteen management not dependent on Counterpart or the McGovern-Dole program. 
 
53 57% of CM1 students, 52% of CE2, and 46% of CE1 students read below grade level. 
54 Evaluation Report McGovern-Dole – Matam, June 2016.
56
The current McGovern-Dole program incorporated these lessons by focusing on stakeholder training and 
engagement, to varying levels of success. 
• The school created official positions for community members, such as SMC members, food 
program managers, school farm presidents, and other positions intended to build ownership and 
permanent engagement with the program. 
• McGovern-Dole built a culture of contributing and supporting the program in communities and 
challenged them to take responsibility for feeding and educating students. 
• The program built the capacity of school directors to engage with communities and raise funds and 
resources. 
• Multiple stakeholders in schools and the community that understand the school feeding program, 
including volunteer scheduling, inventory management, and dietary diversity. 
The extent to which project activities will continue with the absence of support from both USDA and 
Counterpart: Under a best-case scenario, schools could sustain 60 percent of the primary school feeding 
program and 100 percent of the preschool feeding program. This does not take into consideration the effort 
required to maintain engaged volunteers or perform the other functions that McGovern-Dole facilitators 
carry out. 
Other program activities will be more difficult to sustain. Regular pedagogical training, for example, will 
require initiative from school inspectors and a leader to maintain monthly teacher meetings and regular skills 
development. If Counterpart can identify that leader and work with him/her to plan for continued updates to 
the professional skills of teachers in the region, the activity has a reasonable chance of sustaining itself. This 
outcome requires clear communication about the programs progress and end date. 
The government’s School Health Office is interested in assuming ownership of the program’s health-related 
activities. McGovern-Dole deworming and hand washing campaigns are similar to campaigns that the office 
already runs. However, as in the case of the central government, time and clear transition plan are required to 
successfully hand the program over. 
57
Lessons Learned
This section details particular lessons learned during the project. Recommendations highlighting lessons 
learned appear in the next section. 
MGD provides an important opportunity for testing new ideas. 
Over the course of MGD Senegal, Counterpart tested several new and valuable ideas, including new energy 
efficient stove designs, new means of establishing school farms, and the planting of moringa trees to improve 
student health. Counterpart’s approach to managing these new ideas emphasized using them as a learning 
opportunity without exaggerating their successes or failures. Counterpart also kept the cost of trying new 
approaches as a reasonable part of the program’s budget, allowing them to promote ideas that succeeded 
while dropping ideas that failed. 
MGD’s complete design approach was crucial for its success. 
The MDG program’s central aim was to increase literacy and health outcomes by providing nutritious food in 
sufficient quantities to students in schools. The positive results that the program achieved were amplified by 
Counterpart’s strategy for program implementation, which creates locally sustainable assets around the 
complete school feeding and literacy program. This approach ensured that community structures were in 
place to take ownership of the food and manage its storage, safe handling and preparation. Trainings
prevented new bouts of illness in student communities that can come with a new communal feeding system.
Latrines were available so that students had a safe, clean place to relieve themselves after eating. This total 
systems approach was crucial to the program’s success. 
Gender issues require closer examination
Among the goals of the evaluation was to “pay close attention to gender and social inclusion dynamics to 
ensure that considerations related to gender and social inclusion are factored into future programming.” The 
way the McGovern-Dole program’s initiatives interacts with gender and social inclusion is still unclear. There 
are two areas where further investigation is needed. First, with respect to students, different populations in 
the Saint Louis region of Senegal appeared to have different traditional norms with respect to boys and girls’
education. Across the region, more girls (60%) than boys (40%) attend school. However, this imbalance 
certainly didn’t represent all schools. As noted in this report, reasons given for why girls or boys may or may 
not attend school seemed to vary from community to community, with no obvious, single factor tipping the 
balance one way or the other. To understand student attendance and performance as a function of gender, a 
gender specific study is required. 
Secondly, the program’s reliance on volunteer labor creates an unpaid work burden for women that the 
program should address. Interviews with female volunteers indicated that women took responsibility for most 
of the new labor requirements that MGD puts on communities, and that very often this burden fell on a 
small number of volunteers. 
The MGD Program doesn’t have a clear understanding of the sufficiency of 
students’ diets
While the program made an effort to measure the sufficiency of students’ diets, the measurement did not
include a clear picture of what students were eating outside of school and did not emphasize the consumption 
of foods rich in micronutrients and other components of more traditional Minimum Acceptable Diet 
definitions. It is possible that taking a comprehensive survey of students’ diets is beyond the scope of the 
program, in which case a different estimate of SO2: Improved Health, should be adopted. 
58
Student attendance improves because there is food at school 
Student enrollment and attendance at school improved. Interviews with stakeholders indicates that the single 
biggest reason for the improvement was the school feeding program. 
Teacher attendance may not be an issue
The evaluations ToR asked that the evaluation team examine improvement in teacher attendance due to 
reductions in teachers administrative duties. The evaluation team did not find that school administrators, 
students, or parents found teacher attendance to be lacking, and teachers did not report that administrative 
duties were burdensome. 
School needs vary
Understandably, the original design of the MGD program sought to maximize efficiency by applying a one￾size-fits all approach to schools. Accordingly, the amount of food provided to schools was the same per 
student, infrastructure projects focused on new construction, and capacity building projects focused on 
building the same community structures. 
The MGD program changed its approach to provide interventions for schools based on each school’s need 
rather than continue with the previous approach. MGD found that schools’ needs were not always obvious 
and that they were not always a function of relative wealth. Some relatively poorer schools provided more 
food to their school canteens, for example, because they had more developed community structures. Some 
relatively wealthier schools required more help than expected motivating community engagement. 
Sustainability initiatives should start early in the program
Achieving sustainability requires that a number of complicated objectives are achieved. For example: 
• The government requires a budget allocation and policy in place for assuming ownership of the 
program.
• School inspectors and other government staff have to take over roles previously performed by 
project staff, including the management of the volunteer program 
• School directors must have capacity to create a detailed budget of the school feeding program’s items 
and costs, and develop a plan for working with their communities to fund that budget.
• To achieve school feeding sustainability, schools require a relatively large and diverse source of food 
or income, including from 1 to 2-hectare farms, increased granary inventory, and increased 
government assistance.
• Communities have to assume responsibility for monitoring school feeding programs and providing 
food and resources when other sources fall short.
Achieving the objectives takes time. The program would benefit from drawing stakeholders’ attention to 
them as early in the program as possible. 
59
Recommendations
As the prior sections of this document have shown, McGovern-Dole is a successful project. It has improved 
the literacy of primary school students, improved knowledge and practices of hygiene and nutrition among its 
stakeholders, built the capacity of schools and communities to sustain school feeding programs, and 
brightened the future prospects for tens of thousands of children. It has also built the capacity of dozens of 
school communities to support and manage school feeding programs. 
That said, every program, no matter how successful, can be improved. In this section, the report recommends 
the following 8 priority actions that the program should incorporate into its next phase. 
Continue the program: USDA should continue the program in Saint Louis for another 2-3 years. 
Ending the program now would be pulling food from thousands of kids who have just started on an 
educational path equipped with the energy that USDA commodities provide. The next phase of the 
program should focus on building school, community and governmental capacity to sustain and scale 
up the school feeding program. Steps to achieving this include: 
• Convening a workshop of existing school directors and community leaders to identify and 
consolidate school feeding program best practices with regard to sustainability. 
• Focusing on self-sufficiency from the outset of the program. This includes investing in school farm 
and granaries and formation and capacity building for school feeding boards (see below).
• Have a steeper annual phase out of food provided by the program, forcing schools to learn lessons 
about school feeding sustainability with plenty of time left in the program. For example, a new 
program could provide new schools with 100, 67 and 33 percent of school feeding requirements in 
years 1, 2 and 3 respectively. For existing schools, the phaseout could look more like 75, 50 and 25 
percent respectively.
• Not only would a more rapid phasing out of food accelerate sustainability, it would reduce costs per 
school, allowing for expanded outreach. 
Focus school selection on high and medium food insecurity schools: The school selection 
process and reorientation of need-based food provision worked well. However, allowing the regional 
Ministry of Education to have final say over school selection may have led to a higher number of 
relatively wealthy schools participating in the program that otherwise would have been the case. In 
future McGovern-Dole programs, McGovern-Dole should use the process that Counterpart 
developed to categorize schools in the school selection process. School selection should minimize 
the number of low food insecurity schools that the program serves. McGovern-Dole in these schools 
is more likely to replace meals rather than provide additional meals and benefit less from the program 
than the larger medium insecurity schools or, the needier high insecurity schools. 
 
Work with school directors to plan for maximizing each school’s resources: McGovern-Dole
should work with directors to develop school feeding strategies and schedules for different levels of 
food availability. These strategies should be aimed at maximizing the nutritional and educational 
benefits of school feeding within the resources that schools can acquire. For example, preschools 
with only 60 percent of the recommended school feeding budget may serve breakfasts on Monday, 
Wednesday, and Friday. Or a primary school only able to meet 75 percent of the MGD feeding 
recommendations may choose to serve lunch but not breakfast on Tuesdays and Thursdays. More 
difficult or important parts of the curriculum may be scheduled on these days when school 
attendance and student concentration may be highest.
60
Support establishment of School Feeding boards at school and regional levels: School feeding 
programs need to be run more professionally if they are to become more sustainable. At the school 
level, McGovern-Dole Senegal should consider forming a school feeding board to guide decision 
making, resource allocation, and school coordination. This will also help to reduce the burden on
directors that will likely increase with more self-sufficient school feeding programs. For example, the 
board of directors may decide that for the school farm to expand, they need to hire the farm 
president as a full-time manager. They may also help set school feeding schedules or assist schools in 
sharing assets and resources. Members of the school level feeding board should include the director, 
the school farm president, the lead cook, parents and a community leader. 
A regional school feeding board should also be formed for St Louis. The purpose of this would be to 
identify and disseminate best practices in building self-reliant school feeding programs and to 
advocate for policy and resource support for school feeding within the government and private 
sector. Members would include the MoE and MoA at departmental and regional levels, food and 
agriculture companies, as well select members of school level feeding boards. 
Identify ways to overcome school farms’ major constraints: School farms potential is limited by 
four major constraints:
• Insufficient access to finance
• Labor shortages
• Lack of sufficient equipment and inputs
• Lack of a model farm to learn from
School directors and community members are already contributing a good deal by taking on extra work 
and volunteering to support extracurricular activities. The next phase or iteration of the program should 
work to help school farms to overcome these constraints so that farms can reach their potential in 
contributing to a sustainable school feeding program. Part of this activity could involve partnering with 
the MoA to create school farm demonstration plots. These plots would introduce improved technologies 
and approaches to the community and increase the productivity of school farms. In the evaluation 
team’s interview, The Ministry of Agriculture expressed interest in pursuing this idea.
Overcome the diminishing marginal utility of the program: To overcome the diminishing utility 
of the program, USDA could look at second phase initiatives that may have a large impact on 
primary school literacy and improved health practices. Potential second phase initiatives to improve 
literacy include improving the reading skills of mothers with preschool children, developing reading 
events or competitions, establishing libraries or otherwise increasing access to a wide variety of 
books, developing summer learning and reading initiatives, and generally providing more resources 
for reading and raising the profile of its importance. For health and hygiene practices, the program 
could promote improved household food processing, storage and preservation to retain nutrient 
value, shelf life, and food safety; conduct community-level cooking training and demonstrations and 
advocating for improved household level hand-washing practices.
Use random spot checks as well as school data to track increased attendance: The system 
currently does a good job of collecting school attendance records and entering the information into a 
central database, where the team can analyze the data and uncover trends. However, particularly in 
the case of student attendance, the high reported rates of attendance and the insufficient 
administrative staff at schools makes the data seem. suspect. 
61
To truly understand the way the program influences school attendance, McGovern-Dole should have 
an independent research plan for studying the topic from the beginning of the program. One 
approach could be use random spot checks to estimate a difference between reported attendance 
rates and actual attendance rates. To conduct spot checks, Counterpart’s Program Manager would 
collect student enrollment data for each school at the beginning of the school year. Then, the 
Program Manager would randomly choose days for school project officers to take a student and 
headcount at each school. The Program Manager would choose at least one day each quarter of the 
school year, and would take into consideration fluctuations due to seasonal attendance and the time 
of day. On spot check days, school project officers would take a head count of students for each 
grade in the school. Comparing reported attendance against attendance taken during spot checks 
would show whether or not official school attendance records were accurate. If they are not accurate, 
the spot check head counts divided by total class enrollment would give a more accurate estimate of 
attendance. Over the course of a three to five-year program, USDA could analyze the spot check 
data to see if the MGD project produced any attendance trends. 
Research whether infrastructure construction activities provide sufficient value: Interviews 
and surveys revealed that improvements in nutrition and health initiatives were primarily responsible 
for improvements in reading. Infrastructure projects were less successful and created unfulfillable 
expectations among some project beneficiaries. School infrastructure projects were also the single 
biggest direct cost line item in the programs budget. The funds allocated to school infrastructure 
projects could have funded school feeding programs for over 100 new schools and over 17,000 
additional students. Schools certainly require repair and upgrading. USDA could use the data it has 
on hand from previous McGovern-Dole programs to see if there is a correlation between schools 
that benefit from infrastructure projects and improved reading scores. It could also allocate funds for 
school maintenance and upgrade rather than new construction. These savings could be used to serve 
more schools through other project activies. 
62
Annex 1: Document Review List
1. CPI Performance Indicators Attachment E
2. CPI Performance ME Plan Revised 12/09/15
3. CPI Plan of Operation Attachment A
4. CPI MGD Midterm Report - Updates to USDA April 2018
5. CPI MGD Program Spending Report
6. CPI Reports on health and nutrition activities
7. CPI Reports on literacy training and capacity building activities
8. CPI School Categorization document
9. CPI Students, teachers, schools, and other monitoring and evaluation system databases
10. FFE Sustainability Evaluation Report Sept 2016- Matam
11. Food For Progress and McGovern-Dole Indicators and Definitions
12. FFE Semiannual Performance Report April 2017 - September 2017
13. FFE Semiannual Performance Report April 2016 - September 2016
14. FFE Semiannual Performance Report April 2015 - September 2015
15. FFE Semiannual Performance Report Oct 2014 - March 2015
16. FFE Semiannual Performance Report Oct 2015 - March 2016
17. FFE Semiannual Performance Report Oct 2016 - March 2017
18. FFE Semiannual Performance Report Oct 2017 - March 2018
19. Food For Education Activity Management guides and protocols.
20. Food for Education Annual Workplans
21. Fuchs, Marek. "The Reliability of Children’s Survey Responses: The Impact of Cognitive Functioning 
on Respondent Behavior." Symposium 2008: Data Collection: Challenges, Achievements and New 
Directions (2008): 1-9. Web.
22. Kampman, Haile et. al. “How Senegal created an enabling environment for nutrition: A story of 
change,” Global Food Security, pgs 57 – 65, 2017.
23. Memos of Understanding between CPI and Communities
24. MGD Indicator Framework
25. Monitoring and Evaluation Databases and Data Collection Tools
26. Performance ME Plan Revised 12/09/15
27. prnewswire.com/news-releases/global-handwashing-day-focuses-on-need-for-universal-hand￾hygiene-300159521.html
28. Raw Baseline Data
29. School Canteen Guide
30. School Hygiene and Nutrition Guide
31. School Selection Criteria Document
32. Senegal School Curriculum Guides
33. Social Progress Index Senegal Report
34. Staff Field Visit and Training Reports
35. World Bank’s Databank “World Development Indicators” database.
36. World Food Program “Senegal Transitional Interim Country Strategic Plan 2018.”
63
Annex 2: Data Collection Tools
Student Survey
1. Start time
2. End time
3. Date of interview
4. Device ID
5. subscriber id
6. username
7. Telephone number
8. SIM serial number
9. Introduction: "You have been selected to participate in a survey on health, nutrition and education as 
part of the School Canteen project. 
Your participation in this study is entirely voluntary. You are under no obligation to participate. You have the 
right to refuse to answer questions and to withdraw from the study at any time. 
If you agree, please answer all questions as honestly as possible. If you are unable to answer one of the 
questions, you can ignore the question. 
All your answers are strictly confidential."
10. Please select the interviewer name from the list bellow
11. Commune
12. Name of school
13. Select preschool or elementary level
14. Type de structure
15. "Ask for the respondent's consent”
If the person agrees to participate, go with the interview."
16. What is your first name
17. What is your last name?
18. How old are you?
19. Gender?
20. Grade?
21. On school days, do you usually eat breakfast? 
a. Yes
b. No
22. Why do not usually eat at breakfast time?
a. I do not have enough food at home
b. I do not have enough time to go home
c. I do not like food in the canteen
d. I do not have money to buy food
e. other (Specify)
23. What do you usually eat for breakfast?
a. Tubers, or grains (such as millet, rice, maize, sorghum), or roots (such as cassava)
64
b. Nuts or legumes (such as beans like niebe)
c. Milk, yogurt or cheese
d. Meat, fish, poultry or liver/organ meats
e. Eggs
f. Vegetable red palm oil, or fruits and vegetables including pumpkin, carrot, squash, sweet 
potato, dark green leafy vegetables, ripe mango, cantaloupe melon, apricot, ripe papaya, 
peach, red peppers, moringa leaves, bean leaves
g. Fruits and vegetables, such as onion, eggplant, watermelon, oranges, green peppers, cabbage, 
tomoatoes, dates
h. Other
24. Where do you usually get your food from at breakfast time?
a. Home 
b. School canteen
c. Snack vender
d. Other (Specify)
25. Do you usually eat something at lunch time while you're in school?
a. Yes
b. No
26. Why don't you usually eat something at lunch time?
a. I don't have enough time to eat
b. I don't have enough time to go home
c. I don't like the food in the canteen
d. I do not have food
e. Other
27. What do you usually eat at lunch time? 
a. tubers, or grains (such as millet, rice, maize, sorghum), or roots (such as cassava)
b. nuts or legumes (such as beans like niebe)
c. milk, yogurt or cheese
d. Did you eat meat, fish, poultry or liver/organ meats
e. Eggs
f. vegetable red palm oil, or fruits and vegetables including pumpkin, carrot, squash, sweet 
potato, dark green leafy vegetables, ripe mango, cantaloupe melon, apricot, ripe papaya, 
peach, red peppers, moringa leaves, bean leaves
g. fruits and vegetables, such as onion, eggplant, watermelon, oranges, green peppers, cabbage, 
tomoatoes, dates
h. Other
28. Where do you usually eat at lunch time? 
a. Home 
b. School canteen
c. Other (Specify)
29. Do you usually eat any other time at school besides breakfast and lunch?
a. Yes
b. No
30. What time of day do you eat besides breakfast and lunch when you're at school?
a. after school starts but before lunch
b. after lunch but before leaving school
c. Other
31. What do you eat at this time normally?
a. tubers, or grains (such as millet, rice, maize, sorghum), or roots (such as cassava)
65
b. nuts or legumes (such as beans like niebe)
c. milk, yogurt or cheese
d. Did you eat meat, fish, poultry or liver/organ meats
e. Eggs
f. vegetable red palm oil, or fruits and vegetables including pumpkin, carrot, squash, sweet 
potato, dark green leafy vegetables, ripe mango, cantaloupe melon, apricot, ripe papaya, 
peach, red peppers, moringa leaves, bean leaves
g. fruits and vegetables, such as onion, eggplant, watermelon, oranges, green peppers, cabbage, 
tomoatoes, dates
h. Other
32. What is your opinion of the food served at the school canteen?
a. Big Smiley Face
b. Smiley Face
c. Frown Face
d. Big Frown face
33. Some pupils are hungry during the school day, other pupils are not hungry during the school day. Are 
you ever hungry during the school day?
a. Yes
b. No
34. How often are you hungry during the school day?
a. I'm always hungry during the school day
b. I am often hungry during the school day
c. I'm hungry about half the time
d. I'm hungry from time to time
e. do not know
Now, I'm going to ask you some questions about health and cleanliness 
35. Intestinal worms are parasites that live in people's bodies and make them sick. Can you think of any 
ways to prevent intestinal worms?
a. Avoid walking barefoot (wear shoes)
b. Don't bathe or swim in stagnant water
c. Eat meat that's cooked well done (especially beef and pork)
d. Avoid direct contact with contaminated water, but if necessary wear boots and gloves
e. Wash hands with water that's safe to drink and soap before preparing food, serving food or 
eating
f. Wash hands with water that's safe to drink and soap after using the toilet
g. Protect food from flies, cockroaches, and dust
h. Keep food in a pantry, refrigerator, or place that's clean and well-ventilated
i. Other
36. Some people wash their hands a lot, others almost never wash their hands Thinking about your day 
so far today, have you washed your hands today?
a. Yes
b. No
37. When did you wash your hands today?
a. Before eating
b. Before touching or preparing food
c. Before giving food to someone else
66
d. When you have earth on your hands
e. After touching something dirty
f. After touching an animal
g. After using the latrine
h. After changing a layer
i. Other (Specify)
38. What did you use to wash your hands?
a. Water
b. Ashes
c. Sand
d. Soap
e. Other
39. In your opinion, when do you think a person should wash their hands?
a. Before eating
b. Before touching or preparing food
c. Before giving food to someone else
d. When you have earth on your hands
e. After touching something dirty
f. After touching an animal
g. After using the latrine
h. After changing a layer
i. Other (Specify)
40. Now I’m going to ask you a few questions about healthy food. In your opinion, what food that can 
help you grow?
a. Meat
b. Fish
c. Eggs 
d. Cheese
e. Milk
f. Nuts
g. Other
41. What food can give you energy?
a. Butter 
b. Oil
c. Sugar
d. Bread
e. Rice
f. Millet
g. Other
42. What foods can protect you from diseases?
a. Oranges 
b. Apples
c. Mangoes
67
d. Tomatoes
e. Carrots
f. Watermelon
g. Melon
h. Cabbage
i. Leaves of bissap
j. Beans
k. Moringa
l. Other
43. What do you take when you are thirsty?
a. Water
b. Other
44. What is the reading level of the learner?
Thank you so much for answering my questions. 
Please take the GPS coordinates of the interview location.
Teacher Survey
1. Start time
2. End time
3. Date of interview
4. Device ID
5. subscriber id
6. username
7. Telephone number
8. SIM serial number
You have been selected to participate in a survey on health, nutrition and education as part of the School 
Canteen project. Your participation in this study is entirely voluntary. You are under no obligation to 
participate. You have the right to refuse to answer questions and to withdraw from the study at any time. If 
you agree, please answer all questions as honestly as possible. If you are unable to answer one of the 
questions, you can ignore the question. All your answers are strictly confidential.
Please select the interviewer name from the list below
INTERVIEWER: PLEASE TYPE IN YOUR NAME
9. Do you agree to participate in this survey?
a. Yes
b. No
10. Department
11. Preschool or Primary school?
12. What is your name?
13. What is your last name?
14. What is your age?
68
15. Gender
16. What is the highest level of education you have achieved?
17. Have you completed a teaching certification or formal training in teaching children?
a. Yes
b. No
c. Don’t Know
18. Did you receive literacy training this school year?
a. Yes
b. No
c. Don’t Know
19. Were you consulted or considered before the selection of training modules?
a. Yes
b. No
c. Don’t Know
20. After the training was completed, were you given the opportunity to provide feedback?
a. Yes
b. No
c. Don’t Know
21. What grades do you mostly teach?
a. CE1
b. CE2
c. CM1
d. petite section
e. moyenne section
f. grande section
g. Multigrade class
h. other
22. How long have you been working as a teacher?
23. How many pupils are in your main class?
24. What is the size of the CE1
25. What is the size of the class of CE2
26. What is the size of the CM1 class
27. What is your professional status?
a. Full Time
b. Part time (50% to 90% of full hours)
c. Part time (less than 50% of full hours)
28. Do you regularly talk about nutrition in your class?
a. Yes, already have
b. Yes, I’m planning to
c. No
29. Did you teach or plan to teach about nutrition in your classes this week, which means between 
Monday and Saturday?
a. Yes
b. No
30. Have you received training on teaching nutrition in the last three months?
a. Yes
b. No
31. Do you think the training you received provided you with new tools or techniques to teach nutrition?
a. Yes
69
b. No
32. How often do you use the new tools or skills to teach nutrition?
a. Always
b. Often
c. Sometimes
d. Never
33. Do you usually teach hygiene in your class?
a. Yes, already
b. Yes, planning
c. No, does not and does not plan
34. Did you teach or plan to teach about hygiene in your classes this week, which means between 
Monday and Saturday?
a. Yes
b. No
35. Have you received training on hygiene education in the last three months?
a. Yes
b. No
36. In your opinion, did the training you receive provide you with new tools or techniques to teach 
hygiene?
a. Yes
b. No
37. How often do you use the new tools or hygiene skills?
a. Always
b. Often
c. Sometimes
d. Never
38. Q27 Have you attended a course in pedagogy supported by the government in the last three months?
a. Yes
b. No
39. Q28 Do you think the training she has provided you with new tools or techniques to teach?
a. Yes
b. No
40. Q29 How often do you use the new tools and skills in teaching?
a. Always
b. Often
c. Sometimes
d. Never
41. In your opinion, what foods can help children grow?
a. Meat
b. Fish
c. Eggs 
d. Cheese
e. Milk
f. Nuts
g. Other
42. What foods can give you energy?
a. Butter 
b. Oil
c. Sugar
70
d. Bread
e. Rice
f. Millet
g. Other
43. What foods can protect you from diseases?
a. Oranges 
b. Apples
c. Mangoes
d. Tomatoes
e. Carrots
f. Watermelon
g. Melon
h. Cabbage
i. Leaves of bissap
j. Beans
k. Moringa
l. Other
44. Have you washed your hands today?
a. Yes
b. No
45. When did you wash your hands today?
a. Before eating
b. Before touching or preparing food
c. Before giving food to someone else
d. When you have earth on your hands
e. After touching something dirty
f. After touching an animal
g. After using the latrine
h. After changing a layer
i. Other (Specify)
46. What did you use to wash your hands?
a. Water
b. Ashes
c. Sand
d. Soap
e. Other
47. In your opinion, when do you think a person should wash their hands?
a. Before eating
b. Before touching or preparing food
c. Before giving food to someone else
d. When you have earth on your hands
e. After touching something dirty
f. After touching an animal
g. After using the latrine
h. After changing a layer
i. Other (Specify)
71
Thank you very much for answering my questions.
Director Survey
1. Start time
2. End time
3. Date of interview
4. Device ID
5. subscriber id
6. username
7. Telephone number
8. SIM serial number
You have been selected to participate in a survey on health, nutrition and education as part of the School 
Canteen project. Your participation in this study is entirely voluntary. You are under no obligation to 
participate. You have the right to refuse to answer questions and to withdraw from the study at any time. If 
you agree, please answer all questions as honestly as possible. If you are unable to answer one of the 
questions, you can ignore the question. All your answers are strictly confidential.
9. Please select the interviewer name from the list bellow
10. INTERVIEWER: PLEASE TYPE IN YOUR NAME
11. Ask for the respondent's consent. If the person agrees to participate, go with the interview.
12. Select school
13. Select preschool or elementary level
14. What's your first name?
15. What is your last name?
I would like to ask you some questions about your background
16. How old are you?
17. Gender
18. What is the highest level of education you have completed?
19. Have you ever completed a teacher certification program or received formal training in teaching 
children?
20. Have you received training on teaching literacy since the school year started, that is since last 
October?
21. Do you teach at this school ?
22. What grade do you teach?
23. How long have you been the principal of this school?
Since you are a school principal I would like to ask you a few more questions. Let's start with how the school 
stores food and prepares meals…
24. In your opinion, what is a way to ensure that the storage facility at the school is safe to store the 
food?
a. Airy
b. Adequate size
c. Clean
d. afe
e. Accessible
f. Walls and floor cemented
72
g. Distance between walls and food (30cm)
h. Don’t Know
25. What is a way to maintain the storage facility at the school?
a. Clean inside and out
b. Sweep floor every week
c. Clean roof, walls, and piles once per week
d. Completely clean outside warehouse once a month
e. Remove weeds and waste that attracts rats and insects
f. Other
g. Don't Know
h. Other, please specify
26. What is a way to track the stock/food in your storage facility?
a. Compare the inventoried quantity with the quantity listed on the stock sheet
b. Note the date of the inventory and sign your initials
c. Compare the physical stock to the theoretical stock
d. Report in case of discrepancy the reasons for the loss
e. Fill in the inventory sheet
f. Make a loss PV
27. What is a way that food can be safely prepared?
a. Wash hands with clean water and soap
b. Protect food from flies, cockroaches, dust
c. Respect the expiry dates of foods
d. Avoid foods with mold
e. Avoid breaking the cold chain
f. Any food consumed raw (fruits and vegetables) must be washed cleanly before consumption
g. Do not Know
Great! Now, I'd like to ask you a few questions about the community farm in this community…
28. From your knowledge, does the community have a community farm?
a. Yes
b. No
29. Who manages the community farm?
a. Community Farm Management Committee
b. CGE
c. APE
d. Principal
e. Others
f. Don't Know
30. Does the community support the community farm?
73
a. Yes
b. No
31. What are some of the ways the community supports the farm?
32. Has the received any of the following?
a. Agricultural training
b. Small scale farm equipment
c. Other 
d. Don’t Know
33. Has the community farm produced any yields so far?
a. Yes
b. No
34. Are the yields used to make school meals at the school?
a. Yes
b. No
35. Are the yields sold at local markets?
a. Yes
b. No
36. How are the proceeds from the sale used?
a. Support management of school canteens
b. Support school feeding
c. Pay school lunch cooks
d. Pay members <<who ever manages the farm>>
e. Other 
f. Don't Know
37. What else are the yields used for?
Thank you! Now, I would like to ask you a few question about Moringa…
38. Have you heard of moringa?
a. Yes
b. No
39. Can you give me an example of a benefit of moringa?
a. Provides vitamin C or helps with growth and repair of tissues and healing wounds 
b. Provides protein or is needed for growth and adequate nutrition
c. Provides calcium or is needed for bone growth 
d. Provides vitamin A or helps improve immune system function
e. Provides iron to treat "tired blood" (anemia)
f. Other 
g. Don’t Know
40. Does your school grow moringa?
a. Yes
b. No
41. Does your school use moringa to prepare the school meals?
a. Yes
b. No
42. How often do you use moringa for school feeding during the week, that is Monday through Friday?
a. Always
b. Very Often
74
c. Sometimes
d. Rarely
e. Never
f. Don’t Know
43. Q33 Why does your school not use moringa for school feeding?
a. Moringa is too expensive
b. Do not have access to moringa
c. Do not like the taste of moringa
d. Do not think it is beneficial
e. Other
44. Q34 Does your school have a classroom built by Counterpart?
a. Yes
b. No
45. Q35 Is the classroom in good condition and well equipped?
a. Yes
b. No
46. Q36 Is the classroom used regularly?
a. Yes
b. No
47. Q37 How often does the community contribute to the maintenance of the classroom?
a. Always
b. Very Often
c. Sometimes
d. Rarely
e. Never
f. Don’t Know
Thank you so much for answering my questions.
Observation Checklist
1. Start time
2. End time
3. Date of interview
4. Device ID
5. subscriber id
6. username
7. Telephone number
8. SIM serial number
INTERVIEWER: PLEASE TYPE IN YOUR NAM
75
Q1 Select Department
Q2 Select preschool or elementary level
Q3a Dagana primary school
Q3b Pété primary school
Q3c Podor primary school
Q3d Dagana Pre-school
Q3e Pété Pre-school
Q3f Podor Pre-school
Q3g Saint Louis Pre-school
# Latrines and hand washing
Q5 Does the school have a latrine?
Q6 Is the latrine connected to water?
Q7 Is the water connection working?
Q8 How clean is the latrine?
Q9 Is the latrine well maintained in terms its structure?
Q10 Is it available to boys and girls?
Q11 Is it equipped with a sink for washing hands?
Q12 Is the sink area clean?
Q13 Is there soap?
Q14 Is there a clean towel or other material for drying?
Q15 Is there evidence that the school and parents contribute to latrine management? What is the evidence?
# Food storage
Q16 Is the storage area clean?
Q17 Is the roof covered?
Q18 Is the establishment secure?
Q19 Are there signs of damage from rodents or other pests?
Q20 Which of the following is offered in the canteen?
Other, specify
# Cooking area
Q21 Is the cooking area generally clean?
76
Q22 Is there soap or detergent?
Q23 Are there adequate utensils for cooking and eating?
Q24 Is there an energy efficient stove?
Q25 Is there a way to wash your hands before eating?
# observations
Interview Guide
Introduction for the Interviewer
ISG is conducting research on behalf of Counterpart International. The objective of the research is to analyse 
the impact that Counterpart International’s Food For Education program has had on the beneficiaries and 
communities where it has been implemented. ISG will conduct its research via a survey, focus group 
discussions, and key informant interviews. This interview guide serves as a manual for data collectors 
conducting the key informant interview portion of the research. The interviews will cover five specific areas: 
1. The degree to which FFE has met the needs of students, schools, and communities. 
2. The effectiveness of FFE’s activities aimed at improving nutrition, hygiene practices, literacy, school 
infrastructure and community involvement in schools.
3. FFE’s success at increasing student enrolment and attendance as well as teacher attendance. 
4. The relevance of FFE’s interventions to achieving its intended objectives and providing for the needs 
of targeted beneficiaries. 
5. Examples of improvements FFE could make to activities and services to improve community 
satisfaction and improve student achievement.
ISG’s researchers will conduct face to face interviews with FFE beneficiaries. While this guide is intended to 
assist the interviewer in setting key areas for research and orienting the discussions, the interview should 
happen as a conversation, allowing the interviewer to ask follow up questions that may lead to discovering 
new and important information. 
NOTES FOR THE INTERVIEWER
• You should ensure that you cover all the major issues noted in this guide. However, some questions 
may elicit long discussions and others short answers depending on the relevance of individual topics 
for discussion to individual respondents. Exploring the most relevant areas (and others that may be 
important to the respondent, but not necessarily covered in this guide) is more important than 
covering every question in the guide. 
• FFE’s activities operate in a dynamic and complex environment. It is likely that there are key issues 
that the author of this guide did not anticipate. The interviewer should feel free to ask questions that 
the interviewer feels are fruitful whether or not they are included in the guide.
• Please take notes in a separate document.
• Make sure to record your notes at the end of each day. Too much time between the interview and 
recording notes can lead to loss of salient information that will have a negative impact on the rest of 
the research.
77
INTERVIEW QUESTIONS
Introduction
Hello! Thank you for participating in this interview. My name is _________ and I am working on a project 
for International Solutions Group, a research organization based in the United States. ISG is conducting 
research on behalf of Counterpart International. 
The purpose of our research is to assess the effectiveness of Counterpart International’s Food for Education 
program. Counterpart is interested in understanding if the program has achieved its intended goals. I will ask 
you questions about your life and experience working with Counterpart and the FFE program. 
The interview today will last about half an hour. During the interview, I’ll take notes so I can remember what 
you said.
I want to mention a few important points before we start.
• There are no ‘right’ or ‘wrong’ answers to any of the questions. We are interested in hearing your 
opinions, whatever they are.
• Your participation in this interview is voluntary. If you don’t want to answer any question, you don’t 
have to. The responses that you provide are confidential: we won’t mention your name in any report 
we write and we will never connect your name with anything you say. Your responses will be 
combined with responses from other people.
Do you have any questions before we begin?
Now that I’ve told you about the topics we’ll discuss and answered your questions, are you comfortable 
proceeding with the interview?
1. IF NO: ASK IF YOU COULD PROVIDE ANY OTHER INFORMATION THAT WOULD 
MAKE THE RESPONDENT MORE COMFORTABLE. IF RESPONDENT PREFERS NOT 
TO CONTINUE, TERMINATE INTERVIEW.
2. IF YES: CONTINUE TO THE NEXT SECTION.
DIRECTOR INTERVIEWS
Respondent Background 
1. First, I’d like to understand your history working at the school where you currently work. 
2. Did you ever teach? 
FFE Program Responsiveness
3. The FFE program has several components – school feeding, infrastructure, hygiene, community 
farm construction, training, and others – which has been the most successful at this school? Why? 
4. Did the food the program provided arrive on time? Did you have any trouble with logistics? 
5. How does your school use Moringa? 
6. Have the project’s de-worming and vitamin A interventions worked at your school? How do you 
know? 
7. Which has faced the greatest challenges? Why? 
Teacher and Student Attendance
8. Some schools have problems with teach attendance. Has the FFE program had any impact on 
teacher attendance at your school? How did the change happen? 
78
9. Also, some communities see obstacles to students enrolling in or attending school. Have any aspects 
of the FFE program changed student enrolment or attendance? Which and how? 
Community Involvement
10. How involved is the community in this school in terms of contributing to the community garden, 
infrastructure, or other activities? Where does the community contribute the most? How? 
11. What’s the biggest challenge in engaging the community? 
12. Who contributes most from the community? 
Training
13. Have you received any training through the program? What kind of training? 
14. What was the most important thing you took away from the training? 
15. Has it changed the way you work at all? If so, how? 
Sustainability
16. How does the school ensure proper maintenance of the program and/or facilities? 
17. How will the program sustain itself in the future? 
Conclusion
18. Do you have any final comments or any questions?
TEACHER INTERVIEWS
Respondent Background 
1. First, I’d like to understand your history teaching at the school where you currently work. How long 
have you been a teacher? 
2. How long have you worked at your current school?
3. What grade do you teach?
School Enrolment and Attendance
4. One of the purposes of the FFE project is to increase student enrolment and attendance. In your 
opinion, why do some students not enrol or attend school? Why do some students face obstacles to 
attending or enrolling in school? 
5. Are there some obstacles that boys face that girls don’t? 
6. Are there some obstacles that girls face that boys doing? 
7. Has the FFE program helped improve enrolment? Which parts of it, such as school feeding or access 
to latrines has contributed the most? 
8. What could it do to further improve enrolment and attendance? 
9. Sometimes, teachers also face obstacles in attending school. Have you noticed a change in teacher 
attendance? Why did the change occur? 
Hygiene and Nutrition Practices
10. The project has carried out a number of health and hygiene projects, such as building latrines and 
delivering trainings. Have you noticed a change in student’s hygiene or nutrition practices since the 
program began? Why did the changes occur? 
11. Teachers were also trained in nutrition and hygiene practices. Has their behaviour changed? How? 
Why did it occur? 
12. Have you received any training in hygiene or nutrition? What was the most important skill or piece of 
knowledge you took away from the training? Were you satisfied with the training? 
13. Have the project’s de-worming and vitamin A interventions worked at your school? How do you 
know? 
79
14. Does the school utilize the School Nutrition and Health Guide? How?
General Questions
15. What do you like most about teaching at your school? 
16. Are there any difficulties you face in your daily duties that the FFE program could address? 
Conclusion
17. Do you have any final comments or any questions?
80
PTA MEMBERS/Parents
Respondent Background 
1. First, I’d like to understand your history in this community. Do you have kids at this school? 
2. How long have you served on the PTA?
School Enrolment and Attendance
3. One of the purposes of the FFE project is to increase student enrolment and attendance. In your 
opinion, why do some students not enrol or attend school? Why do some students face obstacles to 
attending or enrolling in school? 
4. Are there some obstacles that boys face that girls don’t? 
5. Are there some obstacles that girls face that boys doing? 
6. Has the FFE program helped improve enrolment? Which parts of it, such as school feeding or access 
to latrines has contributed the most? 
7. What could it do to further improve enrolment and attendance? 
8. Sometimes, teachers also face obstacles in attending school. Have you noticed a change in teacher 
attendance? Why did the change occur? 
Community Involvement
9. Does this school have a community farm? Is the farm important to the school? Why? 
10. Does the community participate in the farm? How does it contribute? 
11. Are there any other ways community members contribute to the school? How? Does the community 
contribute to school infrastructure? 
12. What does the community think of the Moringa plantations? 
13. [if the school has a farm] Does the community like the school farm? Is it effective? How will they 
maintain it in the future? 
Training
14. Did you receive any training as part of the program? What was the best part of the training? How did 
the training change your knowledge or behavior? 
15. What other kinds of training would you like to receive? 
General
16. In your opinion, does the FFE program respond to the school and community’s needs? What is the 
most successful thing the project has done? What has been its greatest challenge? 
Conclusion
17. Do you have any final comments or any questions?
COOKS
Background
1. How long have you cooked at this school? 
81
Nutrition and Hygiene
2. The FFE program seeks to improve nutrition and hygiene at schools. Did you receive any training in 
nutrition or hygiene? What did you learn? What was the best part of the training? 
3. Has the FFE program changed the way you work? How? 
4. What is the best thing about the FFE program? 
5. What is the most difficult thing about the program? 
Food delivery and storage
6. How does the process for food delivery work? When does it come? Where does it come from? 
7. How do you store the food when it arrives? Is it easy to store? How do you know proper storage 
practices? 
8. Are there ever any problems with delivery? What are they? 
Stoves
9. Do you use the cookstoves that the program provided? 
10. How have the cookstoves changed the way you provide food for the students? 
Conclusion
11. Do you have any final comments or any questions?
82
COMMUNITY FARM MANAGERS
Respondent Background 
1. First, I’d like to understand your history in this community. Do you have kids at this school? 
2. How long have you managed the farm?
Community Farm Management
3. How does the farm work? Who provides the labor? Who provides seeds and other inputs? 
4. What happens to the food the farm produces? How much goes to the school? Is any sold? Who 
markets the produce? How is the profit distributed? 
5. How does the farm relate to the community? Are there strengths and weaknesses in community 
ownership? 
6. What has been the farm’s biggest success? 
7. What are the farm’s challenges? 
8. What could FFE do to make the program more successful? 
Conclusion
9. Do you have any final comments or any questions?
Project Staff55
Background
1. Name
2. Position
3. How long have you worked on this project? 
Implementation
4. What do you think is the most successful aspect of the project? What contributes to its success? Has 
the management of that aspect changed in any way? 
5. How do you ensure that the project’s activities meet the needs of its beneficiaries? How were 
schools, communities, and beneficiaries selected? 
6. What aspect of the project has struggled the most? Why? 
Efficiency
7. What are the best parts of how the project is managed? 
8. Do you feel like the project has sufficient resources? 
9. Is there anything you would have done differently in retrospect? 
Sustainability
10. How does the project work toward graduating its communities? 
11. Where is the project most sustainable? Least sustainable? 
Conclusion
12. Do you have any other comments or thoughts you’d like to share?
 
55 Questions asked will differ according to the position of the staff member interviewed.
83
Government Officials56
1. Name, position, department.
2. Why is the Food For Education important to [official’s department or scope of work]? 
3. What have been the program’s most important successes in your opinion? 
4. Every project, no matter how successful, has areas where it can improve. In your opinion, what 
would have improved the design, implementation, or other aspect of the Food For Education 
project? 
5. What’s your vision for the future of this project? How likely is it that the vision will be achieved? 
What resources will be required to achieve the vision? 
6. If the project were starting over today, what advice would you give the new management team? 
7. Do you have any other comments or thoughts you’d like to share?
Focus Group Guide
Introduction for the focus group leader
ISG is conducting research on behalf of Counterpart International. The objective of the research is to 
understand the impact of the Food for Education and Child Nutrition program. Particularly, we want to 
know how the program has worked to meet you expectations, where it has succeeded, and where it could 
improve. ISG’s research will be conducted via a survey, focus group discussions, and key informant 
interviews. This focus group guide serves as a manual for the focus group leader and note taker conducting 
focus groups. 
ISG will conduct focus groups with mothers of students, PTA members, teachers, and school directors. 
Focus groups will cover five topics:
1. The degree to which project activities served the needs of beneficiaries. 
2. The attitude of community members toward ownership and maintenance of infrastructure and asset 
building projects
3. The effectiveness of program in increasing student enrolment and attendance.
4. The importance of community farms.
5. The success of school feeding programs.
The purpose of these focus groups is to present broad issues and to let participants generate ideas 
and build on each other’s ideas. The focus group leader’s goal is to encourage participation and ask 
questions that build on participants thoughts. This guide provides follow up questions that the leader 
may use. The leader may ask specific participants to compare their opinion or thoughts to those of 
other participants through prods such as “Participant B, how does your experience compare to the 
experience of participant A that we just heard?” or “Participant C, what would you add to participant 
B’s comments?” The interaction that the focus group leader encourages will provide depth to the 
information we are gathering through the survey and interviews. 
 
56 The interviewer will also ask questions specific to the government officials area of expertise. Ministry of education 
officials will be asked about training and student achievement. Government of health officials will be asked about the 
SH&NG, moringa, and de-worming and vitamin A programs, for example
84
Preparation
• Make sure there is a sign-in sheet, pen, and paper for each participant.
• Make sure the room is set up with flip charts or whatever materials you need.
• Set up refreshments if appropriate. 
• If available, set up small gifts to thank the focus group participants.
Agenda 
Introduction, Ground Rules, Logistics and Sign-in Sheet
[FOCUS GROUP LEADER: Introduce yourself and the note taker.]
Thank you all for participating in this focus group. My name is _________ and I will lead the group, and this 
is ____________ who will take notes. Our discussion will last no more than an hour and a half. Before we 
get started, let me mention that there are drinks and refreshments available (point them out), and the 
bathroom is (indicate location of bathroom). Please feel free to get up and move around or help yourself to 
refreshments during the discussion. 
The purpose of this focus group is to understand the impact of the Food for Education and Child Nutrition 
program. Particularly, we’d like to understand: 
• How project has served your community’s needs. 
• Your opinion of the project’s construction projects, such as buildings, latrines, or community farms.
• Whether the program has increased school attendance and why or why not.
• The project’s food and hygiene programs.
I wanted to mention a few important points before we start.
• The purpose is for us to learn from you, both your positive and negative experiences and opinions. 
• We’re gathering information, not trying to achieve consensus. Please contribute your experience and 
opinion. It’s ok for there to be disagreement or different ideas.
We have a few ground rules that we hope you’ll agree with: 
• Please participate. Your contribution is valuable and will improve the program’s activities in the 
future. 
• Please agree to keep the information provided in the focus group confidential. We won’t discuss any 
sensitive topics. Still, people may share ideas that they’d rather not be discussed privately. 
• Stay with the group and please don’t have side conversations.
• Turn off cell phones if possible.
Are there any questions before we get started? 
FOCUS GROUP LEADER: Go around the table and ask people to introduce themselves. 
85
Questions 
1. Thinking about the students who go to your school and their families, how have students most 
benefited from the Food for Education program? 
Follow Up Questions
a. What has been the biggest improvement in the school since the program began? 
b. Does the school, its students, or its teachers have any needs that the program hasn’t 
addressed? Which? 
2. I’d like to talk about how the program has changed student attendance. Has student attendance at the 
school increased or decreased since the Food for Education program began?
Follow Up Questions
a. What are the main obstacles to attending school for students in your community? 
b. What are the things that the program has done to overcome those obstacles? Have they been 
effective? 
c. Is there anything else the program could do to help students attend school? 
3. The project has worked with the community to build many assets at the school. These might include 
latrines, classrooms, energy efficient stoves, or community farms. Which has been the most 
important? Why? 
Follow Up Questions
a. Thinking about community farms, is there one at your school? How does your community 
contribute to the farm? Why is the farm important? 
b. Some schools have also built canteens. Is there one at your school? Why is it important? 
How does the community contribute to its maintenance? 
c. What other assets has the program brought that you value? Why?
4. One of the program’s central activities is providing improved access to nutritious food and 
improving nutrition education. Does your school have these programs? 
Follow Up Questions
a. What is the most successful part of the school food program? How has it helped students 
and the community? 
b. How has the program changed student or teacher behavior? 
c. What is the school food program’s biggest challenge? 
d. Does the program as a whole meet the community’s needs? How could it improve? 
86
5. Another program activity is improving health, including hygiene and deworming medication
Follow Up Questions
a. What is the most successful part of the hygiene program? How has it helped students and 
the community? 
b. How has the program changed student or teacher behaviour? 
c. What is the school health and hygiene program’s biggest challenge? 
d. Does the program as a whole meet the community’s needs? How could it improve? 
6. One last question: If you were the boss of the program and could change it in one way, how would 
you change the program? 
87
Annex 3: Selected Samples by Type and Location
Student Sample
Dagana Pete Podor Total
CE1 70 113 147 320
CE2 83 76 119 278
CM1 55 56 83 194
Total 208 245 349 802
Dagana Pete Podor Total
Girls 108 166 97 371
Boys 100 79 252 431
Total 208 245 349 802
Sample - Teachers
Dagana Pete Podor St. 
Louis
Total
Primary 68 72 114 255
Preschool 28 16 17 40 101
Total 96 88 101 356
Female Male Total
Dagana 43 53 96
Pete 27 61 88
Podor 35 96 131
Saint Louis 35 5 40
Total 140 215 355
88
Sample - Directors
Dagana Pete Podor St. 
Louis
Total
Primary 30 23 25 78
Preschool 10 2 4 14 30
Total 40 25 29 14 108
Female Male Total
Primary 2 76 78
Preschool 23 7 30
Total 25 83 108
Sample - Checklists
Dagana Pete Podor St. Louis Total
Primary 32 38 49 119
Preschool 13 6 6 16 41
Total 45 44 55 16 160
89
Annex 4: Survey Respondent Characteristics
90
Annex 5: School Sample
Dept School Code Type Dept School Code Type
Podor 12047 Primary Saint Louis 24028 Preschool
Dagana 13066 Primary Saint Louis 24012 Preschool
Dagana 23006 Preschool Saint Louis 24026 Preschool
Dagana 13010 Primary Saint Louis 24027 Preschool
Pete 13073 Primary Saint Louis 24025 Preschool
Dagana 23007 Preschool Saint Louis 24009 Preschool
Dagana 13029 Primary Podor 12062 Primary
Dagana 13009 Primary Podor 12069 Primary
Dagana 23001 Preschool Podor 12026 Primary
Pete 13065 Primary Podor 12048 Primary
Pete 13067 Primary Podor 22014 Primary
Dagana 23010 Preschool Podor 12031 Primary
Dagana 13001 Primary Podor 22004 Preschool
Pete 13049 Primary Podor 22014 Preschool
Pete 13068 Primary Saint Louis 24006 Preschool
Pete 13074 Primary Saint Louis 24010 Preschool
Dagana 13037 Primary Saint Louis 24015 Preschool
Dagana 13015 Primary Dagana 23011 Preschool
Dagana 11040 Primary Pete 13057 Primary
Dagana 13064 Primary Podor 12044 Primary
Dagana 13008 Primary Dagana 13042 Primary
Pete 13054 Primary Dagana 13045 Primary
Pete 13047 Primary Podor 12074 Primary
Pete 13071 Primary Podor 12061 Primary
Dagana 13075 Primary Pete 11003 Primary
Pete 13052 Primary Pete 11037 Primary
Pete 13053 Primary Pete 11007 Primary
Dagana 23016 Preschool Pete 11015 Primary
91
Pete 13069 Primary Podor 11018 Primary
Dagana 13023 Preschool Podor 12066 Primary
Dagana 13028 Primary Podor 12017 Primary
Dagana 13012 Primary Pete 11023 Primary
Pete 13058 Primary Podor 12036 Primary
Pete 13076 Primary Podor 12028 Primary
Pete 13051 Primary Podor 12039 Primary
Dagana 13013 Primary Podor 22017 Preschool
Dagana 23003 Preschool Podor 12011 Primary
Dagana 23005 Preschool Pete 22006 Preschool
Dagana 13026 Primary Podor 12003 Primary
Dagana 13077 Primary Podor 12005 Primary
Dagana 13014 Primary Pete 22018 Preschool
Dagana 13011 Primary Podor 12004 Primary
Dagana 13027 Primary Podor 12006 Primary
Dagana 23017 Preschool Pete 22013 Preschool
Dagana 13033 Primary Pete 11022 Primary
Dagana 13044 Primary Pete 22008 Preschool
Dagana 13041 Primary Podor 12002 Primary
Dagana 13017 Primary Pete 11012 Primary
Dagana 13032 Primary Pete 11035 Primary
Dagana 23009 Preschool Pete 11026 Primary
Podor 12018 Primary Pete 11016 Primary
Dagana 13019 Primary Podor 12033 Primary
Dagana 13039 Primary Pete 11033 Primary
Podor 22001 Preschool Pete 22015 Preschool
Podor 12020 Primary Podor 11024 Primary
Podor 12024 Primary Pete 11006 Primary
Saint Louis 24029 Preschool Podor 22010 Primary
Saint Louis 24008 Preschool Podor 12056 Primary
92
Dagana 13036 Primary Podor 12078 Primary
Podor 12075 Primary Pete 11011 Primary
Podor 12073 Primary Pete 11009 Primary
Podor 12064 Primary Dagana Primary
Podor 12051 Primary Pete 11041 Primary
Podor 12050 Primary Pete 11008 Primary
Dagana 13024 Primary Pete 11013 Primary
Podor 12025 Primary Podor 12035 Primary
Podor 12055 Primary Podor 12029 Primary
Podor 12012 Primary Podor 12042 Primary
Podor 12068 Primary Podor 12052 Primary
Podor 22016 Preschool Podor 12040 Primary
Dagana 13040 Primary Pete 11025 Primary
Saint Louis 24018 Preschool Pete 22007 Preschool
Podor 12065 Primary Pete 11036 Primary
Dagana 23018 Preschool Pete 11027 Primary
Dagana 23013 Preschool Pete 11002 Primary
Saint Louis 24020 Preschool Pete 11044 Primary
Saint Louis 24023 Preschool Podor 12046 Primary
Saint Louis 24022 Preschool Podor 22010 Preschool
Saint Louis 24002 Preschool Podor 12080 Primary
Podor 12079 Primary
Podor 12041 Primary
Annex 6: Evaluation Team Bios
Mr. Jason Wares, Team Leader, 
93
Mr. Jason Wares, a Director of Programs at ISG, is an international development expert with 15 years’ 
experience specializing in catalyzing development and aid programming focused on livelihoods, 
competitiveness, and market development. Of most relevance to this project, Mr. Wares is currently leading 
baseline evaluation for ActionAid that is focused on Climate Smart Agriculture, women’s empowerment, and 
agriculture development. Other relevant experience includes an assessment of FINCA International’s Food 
for Progress (USDA-funded) projects in Haiti and Tanzania. Both assessments sought to understand how 
FINCA’s microfinance clients use loans to create sustainable microenterprises that translate into improved 
quality of life for their communities. Previously, Mr. Wares served as a senior economic development advisor 
at RTI International. At RTI, Mr. Wares led the institute’s business environment improvement technical area. 
Mr. Wares holds an MBA from the Thunderbird School of Global Management.
Mr. Chris Root, Co-Evaluator
Mr. Root is an agricultural economist with ten years of experience including in M&E, and impact evaluation. 
He recently worked with ISG leading the Final Evaluation of FINCA’s Tanzania program (USDA-funded) as 
well as for the Mid-term Evaluation of the McGovern-Dole USDA-funded program in Senegal. While 
working in Sudan, Mr. Root assessed the use of crop byproducts in livestock feed production. He made 
recommendations on how these byproducts, which are currently underutilized, could be better used to feed 
the country's huge livestock population. For a Somalia project he produced a brief on the camel milk value 
chain making recommendations on how a new project could work in the sector. While working at Land 
O'Lakes, Chris conducted research on livestock lending. Mr. Root has experience and expertise in propensity 
score matching (PSM) and difference in difference evaluation methodology. In Nepal he led a large 
agricultural value chain impact evaluation, conducting propensity score matching to identify impact. This built 
off his MA research titled "Assessing the Impact of Agricultural Value Chain Interventions" which developed 
the approaches he used in Nepal and built off his coursework. For RTI, he created guidelines on assessing the 
impact of economic development projects which detailed the PSM methodology. Recently at Michigan State 
he updated these skills through coursework on these quasi-experimental evaluation methodologies for his MS 
in Agricultural Economics.
Ms. Noreen Mucha, Technical Specialist
Ms. Mucha brings over 15 years’ experience implementing and evaluating child nutrition, sanitation and 
education projects. Ms. Mucha has been engaged with ISG since 2014 conducting evaluations, most recently 
for the Irish AID Program Fund recipient GOAL in Kenya and Zimbabwe centered on nutrition practices, 
child protection and water, sanitation and hygiene.
94
Annex 7: Indicator Framework – Baseline vs Mid-Term Data
Strategic Objectives
Indicator Indicator description Final 
Evaluation 
Result
Baseline 
Result57
SO 1: Improve 
Literacy of School 
Aged Children
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
All 53.68
Boys 66 31
Girls 57 33
1.2.1: reduced 
Short-Term 
Hunger
Percent of students in target 
schools who indicate that they are 
not hungry during the school day
All 64.54 66
Boys 61.9
Girls 65.9
1.2.1: reduced 
Short-Term 
Hunger
Percent of students in target 
schools who indicate that they are
not hungry (or hungry from "time 
to time") during the school day
All 86.72 66
Boys 85.35
Girls 87.43
1.4.3: Increased 
Government 
Support
Percent of teachers who received 
government supported training in 
pedagogy in the past three months
All 31.73 28
SO 2: Increased 
Use of Health and 
Dietary Practices 
Percent of school-aged children 
receiving a minimum acceptable 
diet at the school level
All 57.89 2
2.2: Increased 
Knowledge of Safe 
Food Prep and 
Storage Practices
Percent of beneficiaries (students, 
cooks) who use appropriate 
handwashing practices29 (i.e. with 
soap, before meals, before food 
prep, after latrine use, and diaper 
changing)
All 73.68
Boys 71.43 50
Girls 74.86 55
All 51.88
 
57 The baseline study did not disaggregate all indicators. The table includes baseline indicators at the level each was 
reported, resulting in some blank cells in the table. 
95
2.1: Improved 
Knowledge of 
Health and 
Hygiene Practices
Percent of students in target 
schools who can correctly identify 
at least 2 ways to prevent intestinal 
worms
Boys 52.75 8
Girls 51.43 10
Difference between baseline, adjusted baseline, mid-term and final evaluation reading levels
Baseline Adj. baseline58 Midterm^ Final 
Grade 3 (CE1) 3.31 4.1 5.04 4.88
Grade 4 (CE2) 4.89 5.54 6.25 6.39
Grade 5 (CM1) 6.19 6.7 7.34 7.69
COHORT ANALYSIS
Baseline Adjusted Baseline Midterm Final
BASELINE CE1 COHORT 3.31 4.1 6.25 7.69
BASELINE CE2 COHORT 4.89 5.54 7.34
BASELINE CM1 COHORT 6.19 6.7
 
58Adjusted baseline accounts for the fact that children in the midterm have had more schooling than those at baseline. 
The baseline data was collected in Jan-Feb, the midterm data in May-June and the endline in April-May. Therefore 
kids in each grade have had nearly half a year longer to learn and should be expected to do better on reading tests, 
independent of the project. 
Adjustment made by dividing the difference betweren grades at the baseline by two (half a school year) and adding to 
the preveious year. I.e adjusted grade 3 = grade 3 + ((grade 4 - grade 3)/2)
For grade 5, the project increase in reading score to grade 6 is based on the difference between the difference from grade 
3 to 4 and grade 4 to 5. That is 1.58 - 1.30 = .28. 1.3 - .28 = 1.02.
96
Annex 8: Evaluation Terms of Reference
Please find the terms of reference here