1
BASELINE STUDY: FY 2015 USDA MCGOVERN-DOLE FOOD FOR 
EDUCATION HOME GROWN SCHOOL FEEDING PROGRAM IN 
RWANDA
2016-2020 
July 22, 2016 
FUNDED BY:
McGovern-Dole International Food for Education and Child Nutrition 
Program 
Agreement number: FFE-696-2015/007-00
LED BY:
AUTHORS:
Peninnah Mukiri 
Amos Kaburu
DISCLAIMER: This publication was produced at the request of the 
United States Department of Agriculture. It was prepared by an 
independent third-party evaluation firm. 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.
Accessibility Note: An accessible version of this document can be
made available by contacting 
fas.monitoring.evaluation@usda.gov 
2 | Page
Abbreviations
ADRA Adventist Development and Relief Agency
CSB+ Corn Soya Blend
DEO District Education Officer
ECD Early Child Development
EGRA Early Grade Reading Assessment
FAD Food Assistance Division
FAO Food and Agricultural organizations
FAS Foreign Agricultural Services
HGSF Home Grown School Feeding
HQ Headquarter
KII Key informant interview
MAD Minimum Acceptable Diet
MGD McGovern-Dole
MINAGRI Ministry of Agriculture and Animal Resources
MINEDUC Ministry of Education
MINALOC Ministry of Local Government
MINISANTE Ministry of Health
NISR National Institute of Statistics Rwanda
P4P Purchase for Progress
PTA Parent-teacher association
RAB Rwanda Agriculture Board
REB Rwanda Education Board
SMC School management committee
ToR Terms of Reference
UN United Nations
UNICEF United Nations Children’s Fund
USAID United States Agency for International Development
USD United States Dollars
3 | Page
USDA United States Department of Agriculture
WFP World Food Programme
WV World Vision
4 | Page
Summary of Findings 
WFP Rwanda has been awarded USD 25 million through the United States Department of 
Agriculture (USDA) McGovern-Dole program to support school feeding over the next five 
years (2016-2020) in support of the Government of Rwanda's Home Grown School Feeding 
(HGSF) program. The program will be implemented in four districts; Nyaruguru, 
Nyamagabe, Rutsiro and Karongi districts, where poverty and food insecurity is high. This 
baseline study was commissioned in order to establish indicator baseline data and information 
to allow for regular monitoring of activity outputs and performance indicators; form the 
foundation for the planned midterm and final evaluations; and provide a situation analysis. 
The study was conducted through desk research, school based surveys and 13 key informant 
interviews. The school survey was conducted with a sample of 40 schools drawn from a 
sampling frame of the 104 target schools. School data was collected from a sample of 40 
structured observations; 240 teachers (P1-P6) from selected randomly selected classes; 
Early Grade Reading Assessment (EGRA) test with 1200 randomly selected P1-P3 students;
1200 randomly selected P4-P6 students; 40 head teachers; 39 storekeepers and cooks; 40 
members of Parent Teachers Association (PTA); and 29 members of school management 
committee (SMC). Field work for this study was conducted between June 7, 2016 and June 
28, 2016.
The baseline study focused on indicators that could be measured before project 
implementation. Some indicators could not be measured because their definition is linked 
to the implementation of project activities (see Annex 1). Baseline values for each indicator 
measured against its corresponding target, as per the project document, are summarized 
in Table 1.
Following the baseline study, it is essential that WFP reviews and realigns the targets. In 
preparation for the midterm and final evaluation, there is also a need for a strong program 
monitoring component that collects and compiles data from each of the beneficiary school 
and related activities on a regular basis. All indicators, including policy-related indicators,
require specific project records. 
5 | Page
Table 1: Summary of Baseline Indicator Values Against Target
Result Indicator Baseline 
Indicator Value Target
Increased Access to 
Preventative Health 
Interventions
Number of students receiving 
deworming medication(s)
75,749 127,650
Increased Access to 
Requisite Food 
Preparation and 
Storage Tools and 
Equipment
Number of target schools with access 
to improved food preparation and 
storage equipment (kitchens, cook 
area, storerooms, stoves and kitchen 
utensils
10 104
Increased Student 
Enrolment
Number of students enrolled in 
school receiving USDA assistance 
(female)
41,404 66,378
Number of students enrolled in 
school receiving USDA assistance 
(male)
40,954 61,272
Number of students enrolled in 
school receiving USDA assistance
82,358 127,650
Increased Access to 
Clean Water and 
Sanitation Services
Number of schools using an 
improved water source
62 412
Number of schools with improved 
sanitary facilities
94 412
Increased knowledge 
of Safe Food Prep and 
Storage Practices 
Percent of cooks and storekeepers 
who can identify at least three safe 
food preparation and storage 
practices
92% 95%
Improved Knowledge 
of Health and Hygiene 
Practices
Percent of students who can identify 
at least three key health and hygiene 
practices (female)
47% 80%
Percent of students who can identify 
at least three key health and hygiene 
practices (male)
48% 80%
Improved 
Attentiveness
Percentage of students in classroom 
identified as attentive by their 
teachers
60% 80%
Improved Student 
Attendance
Number of students regularly (80%) 
attending USDA supported 
classrooms/school (female)
37,964 53,102
Number of students regularly (80%) 
attending USDA supported 
classrooms/school (male)
37,512 49,017
6 | Page
Number of students regularly (80%) 
attending USDA supported 
classrooms/school 
75,522 127,650
Reduced Health￾Related Absences
Percent of students who miss more 
than 10 school days per year due to 
illness (female)
8% 20%
Percent of students who miss more 
than 10 school days per year due to 
illness (male)
7% 20%
More Consistent 
Teacher Attendance
Number of teachers in target schools 
who attend and teach school at least 
90 percent of scheduled school days 
per school year
986 653
Improved Literacy of 
School-Age 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 (female)
57% 90%
Percent of students who, by the end 
of two grades of primary schooling, 
demonstrate that they can read and 
understand the meaning of grade 
level text (male)
42% 90%
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 
49% 90%
7 | Page
Table of Contents
ABBREVIATIONS .................................................................................................... 2
SUMMARY OF FINDINGS ........................................................................................ 4
TABLE OF CONTENTS.............................................................................................. 7
1.0 INTRODUCTION................................................................................................11
1.1 Objectives of the Baseline Survey .................................................................................................11
2.0 STUDY APPROACH AND METHODS ..................................................................13
3.0 SITUATION ANALYSIS ......................................................................................18
3.1 School Feeding Program ........................................................................................................................18
3.2 School Gardens .......................................................................................................................................23
3.3 Deworming ..............................................................................................................................................24
3.4 Other organisations working with the primary schools ....................................................................26
3.5 Education promotion Activities at community level........................................................................30
3.6 Teachers reward and recognition program.......................................................................................31
3.7 Health Clubs .........................................................................................................................................31
3.8 Education Policies ..............................................................................................................................31
4.0 RESULTS FOR LITERACY INDICATORS (RESULTS FRAMEWORK #1) .................32
MGD 1.1: Improved quality of literacy instruction ..................................................................................32
MGD 1.2: Improved Attentiveness..............................................................................................................38
MGD 1.3: Improved Student Attendance...................................................................................................39
5. 0 RESULTS FOR HEALTH INDICATORS (RESULTS FRAMEWORK #2) .................41
MGD 2.1 Improved Knowledge of Health and Hygiene Practices ..........................................................41
MGD 2.2 Increased knowledge of safe food preparation and storage practices...................................41
MGD 2.4 Increased Access to Clean Water and Sanitation.....................................................................43
MGD 2.5 Increased Access to Preventative Health Interventions..........................................................46
MGD 2.6 Increased Access to Requisite Food Prep and Storage Tools and Equipment.....................47
6.0 ADDITIONAL FINDINGS ...................................................................................49
8 | Page
6.1 Schools with an ECD Centre and feeding programs ..................................................................49
6.2 Involvement of students in writing competition .............................................................................49
6.3 Schools with Access to a Hand Washing Facility.............................................................................49
7.0 LIMITATIONS OF THE STUDY ...........................................................................51
8.0 CONCLUSIONS AND RECOMMENDATION .........................................................53
ANNEXES 54
Annex 1: Summary of baseline indicators against target........................................................................55
Annex 2: Sampling document .....................................................................................................................62
Annex 3: Additional EGRA findings (that include P1& P2) and tables..................................................67
Annex 4: References .....................................................................................................................................74
Annex 5: Result Framework ........................................................................................................................75
TABLES
TABLE 1: SUMMARY OF BASELINE INDICATOR VALUES AGAINST TARGET............................................................5
TABLE 2: SUMMARY OF PRIMARY DATA COLLECTION ACTIVITY IN SCHOOL, TARGET RESPONDENT AND SAMPLE SIZE
....................................................................................................................................................................14
TABLE 3: SPONTANEOUS MENTIONS BY PTA, SMC AND HEAD TEACHERS OF AREAS OF SUPPORT .....................21
TABLE 4: SUMMARY OF ANTICIPATED CHALLENGES ............................................................................................22
TABLE 5: SUMMARY OF SCHOOL GARDEN ACREAGE IN SAMPLED SCHOOLS.........................................................23
TABLE 6: MAPPING OF STAKEHOLDERS WORKING WITH PRIMARY SCHOOLS BY INTERVENTION ......................27
TABLE 7: LIST OF EDUCATION PROMOTION ACTIVITIES THAT PTA AND SMC ARE ENGAGED IN AT COMMUNITY 
LEVEL ..........................................................................................................................................................30
TABLE 8: ANALYSIS OF LITERACY INDICATOR BY GENDER...................................................................................34
TABLE 9: ANALYSIS OF LITERACY INDICATOR BY DISTRICT..................................................................................34
TABLE 10: PERFORMANCE ON READING SOUNDS AND SYLLABLES ......................................................................35
TABLE 11: AVERAGE OVERALL P3 PERFORMANCES AGAINST THE EXPECTED READING LEVEL ...........................36
TABLE 12: FREQUENCY OF CARRYING OUT DEWORMING ACTIVITIES IN SAMPLED SCHOOLS AS REPORTED BY THE 
HEAD TEACHER. ..........................................................................................................................................47
TABLE 13: DISTRIBUTION OF THE NUMBER OF STUDENTS THAT PARTICIPATED IN THE LAST COMPETITION BY 
SEX AND CLASS ............................................................................................................................................49
TABLE 14: DETERMINING NUMBER OF SCHOOLS REQUIRED ...............................................................................63
TABLE 15: LIST OF SURVEYED SCHOOLS, BY DISTRICT..........................................................................................64
TABLE 16: LIST OF KEY INFORMANT INTERVIEWS CONDUCTED .........................................................................66
9 | Page
TABLE 17: AVERAGE READING SCORES AGAINST THE NUMBER OF TASKS ............................................................71
TABLE 18: AVERAGE READING SCORES BY CLASS.................................................................................................71
TABLE 19: AVERAGE SOUNDS READ BY DISTRICT AND DISAGGREGATED BY CLASS ..............................................71
TABLE 20: AVERAGE WORDS IN A TEXT CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
CLASS...........................................................................................................................................................71
TABLE 21: AVERAGE INVENTED WORDS CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
CLASS...........................................................................................................................................................72
TABLE 22: AVERAGE FAMILIAR WORDS CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
CLASS...........................................................................................................................................................72
TABLE 23: AVERAGE SYLLABLES READ BY DISTRICT AND DISAGGREGATED BY CLASS .........................................72
TABLE 24: OVERALL READING LEVELS BY DISTRICT............................................................................................72
TABLE 25: AVERAGE SOUNDS READ BY DISTRICT AND DISAGGREGATED BY CLASS..............................................72
TABLE 26: AVERAGE SYLLABLES READ BY DISTRICT AND DISAGGREGATED BY CLASS .........................................73
TABLE 27: AVERAGE FAMILIAR WORDS CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
GENDER.......................................................................................................................................................73
TABLE 28: AVERAGE INVENTED WORDS CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
GENDER.......................................................................................................................................................73
TABLE 29: AVERAGE WORDS IN A TEXT CORRECTLY READ PER MINUTE BY DISTRICT AND DISAGGREGATED BY 
GENDER.......................................................................................................................................................73
TABLE 30: AVERAGE READING SCORES BY GENDER............................................................................................73
FIGURES
Figure 1: Summary of foods grown in school gardens .........................................................................................23
Figure 2: P3 detailed literacy performance by district: ........................................................................................37
Figure 3: Proportion of Non-readers in P3 by Task...............................................................................................37
Figure 4: Percent distribution of student across the attentive ladder .................................................................38
Figure 5: Percent of students who can identify at least three key health and hygiene practices by gender and 
grade ...........................................................................................................................................................41
Figure 6: Percent of cooks who cooks and store keepers who could mention at least 3 safe and food and 
storage practices.........................................................................................................................................43
Figure 7: Safe water access in schools..................................................................................................................44
Figure 8: Status of sanitation in sampled schools ................................................................................................45
Figure 9: Summary of schools with specific features of this indicator ................................................................48
Figure 10: Schools’ Access to a hand washing facility ..........................................................................................50
Figure 11: Average number of sounds, syllables and words read per minute .....................................................67
Figure 12: Average Reading Scores by Class.........................................................................................................68
Figure 13: Average Reading Scores by District .....................................................................................................69
10 | Page
Figure 14: Average Reading Scores by Gender.....................................................................................................70
Figure 15: Comparison of reading Sounds and Syllables by grade .......................................................................70
11 | Page
1.0 Introduction
The United States Department of Agriculture (USDA) through the McGovern-Dole program 
is a major funding mechanism for school feeding worldwide. Managed by the Food Assistance 
Division (FAD) under the Foreign Agricultural Services (FAS), the McGovern-Dole program 
strategic objectives are to improve literacy and increase the use of health and dietary 
practices. The program provides US produced agricultural commodities and financial 
assistance, and supports capacity development and enhanced monitoring and reporting, 
with a key emphasis on sustainability and government ownership.
WFP Rwanda has been awarded USD 25 million through the McGovern Dole program to 
support the home grown school feeding (HGSF) program over the next five years (2016-
2020), a program designed to complement in support of the Government of Rwanda’s 
school feeding program. WFP will implement the HGSF program in close collaboration with 
the Ministry of Education (MINEDUC), Ministry of Agriculture and Animal Resources 
(MINAGRI) and Ministry of Local Government (MINALOC). The HGSF vision is to create 
school feeding models and generate evidence on their impact that could eventually be 
adopted by the government at a national scale.
WFP in collaboration with government partners at national, district, and sector levels plans 
to provide approximately 7 million hot lunches and 7.76 million mid-morning and mid￾afternoon porridge meals annually to over 83,000 students. The program will be 
implemented in Nyaruguru and Nyamagabe districts in the Southern Province, and in 
Rutsiro and Karongi districts in the Western Province, districts that are combating poverty 
and food insecurity. World Vision (WV) is a sub-grantee on this project, whose interventions 
and will focus on literacy, health, and WASH.
1.1 Objectives of the Baseline Survey
The grant agreement for this program incorporates a project monitoring plan that 
specifies 70 performance indicators against which progress will be measured. WFP 
Rwanda commissioned this baseline study to establish baseline information against which the 
project will regularly measure its performance.
Research objectives are: 
i. Establish indicator baseline data and information for use to regularly monitor activity 
outputs and performance indicators for lower level results;
12 | Page
ii. Form the foundation for the planned midterm and final evaluations to measure 
performance indicators for McGovern-Dole's two strategic objectives as well as the 
highest level results that feed into the strategic objectives;
iii. Provide a situational analysis before the project begins, and the context necessary 
for the midterm and final evaluations to assess the project's relevance, 
effectiveness, efficiency, sustainability and impact.
13 | Page
2.0 Study Approach and Methods
The baseline survey was conducted in the four program districts namely Nyaruguru, 
Nyamagabe, Rutsiro and Karongi. The population of the study was 104 schools selected for 
the HGSF program focusing on the primary section. 
A combination of secondary and primary data were used to create a situational analysis of 
school feeding in Rwanda and provide reliable measures for the indicators. Primary data 
collection utilized both quantitative and qualitative approaches. Qualitative data was collected 
through 13 key informant interviews (KII) with stakeholders (see Annex 2 for a detailed list of 
KII). The quantitative approach was a school based survey targeting the following categories 
of respondents:
• Head teacher
• Teachers (grade 1-6)
• Students in grade 1-3 (EGRA)
• Students in grade 4-6
• Cooks
• Store keepers
• Parents teachers association (PTA)
• School management committee (SMC)
Sampling of schools 
The baseline survey was conducted in 40 schools, 10 schools in each district. The schools were 
selected using systematic random sampling. Refer to Annex 2 for a detailed write up on 
sample size determination and sampling procedures of each cohort at school.
Summary of school activities, target and sample size 
Table 2 in the next page summarizes the achieved sample for each cohort in the 40 schools. 
14 | Page
Table 2: Summary of primary data collection activity in school, target respondent and sample size
Primary data collection 
activity
Summary description of the activity Target respondent Achieved sample
School facility checklist These were structured observations using 
a checklist
Observation of school 
facilities and infrastructure
40 observations
Early Grade Reading assessment 
(EGRA)
A standard test in Kinyarwanda which 
was conducted with students in P1-P3. 
The students were picked randomly. 
10 interviews per class (5 boys; 5 girls) 
30 interviews per school 
The test was adapted to Rwandan context
before conducting the study
Primary school children in 
P1- P3
1,200 students
400 in each class (P1, 
P2 and P3)
Student survey (health 
indicators)
A short survey on health indicators and 
learning environment targeting P1-P6. 10 
interviews per class (5 boys; 5 girls) 
60 interviews per school 
Primary school children in P 1
- P6
1,200 students in P4-
P6
1,200 students P1-P3; 
these were the same 
students who did EGRA
Teachers’ Survey Interviews with teachers in sampled P1-
P6 classes for student survey. This was 
Class teacher of P 1-P6 240
15 | Page
combined with student attendance 
records document review
6 interviews per school
Interviews with
Store Keeper and Cook
A survey with the school store keeper and 
cook for those schools that had these staff
Store keeper and cook 39 interviews (29 store 
keepers and 10 cooks)
Head teacher interview and 
document review (teacher 
attendance register, classroom 
register, school enrollment, 
training records, records on 
awards and recognition, records 
on health clubs, records on 
harvest, deworming records)
A survey with the head teacher and review 
of documents in school
Head teacher 40 
Interviews with school leadership A short survey with school management 
committee (SMC) member where they 
existed- 1 member per school
School management 
committee 
29
A short survey with a Parents Teachers 
Association (PTA) member. All schools 
had a PTA-1 member per school
PTA representative 40 
16 | Page
Desk research:
A desk review was conducted using documents sourced from WFP, USDA website, publically 
available documents from the Government of Rwanda and other online resources. Desk 
research focused on each indicator in the performance management plan. Analysis of 
secondary data for each indicator used the guidelines provided under Food for Progress and 
McGovern-Dole (MGD) indicators and definitions. One of the major challenges encountered 
is that the secondary data was not disaggregated by student/teacher /school to meet the 
requirements of the indicator as defined by USDA. In addition, the definitions of most 
indicators according to the guidelines required specific information not available in secondary 
data. For example, while statistics exist on types of toilets in schools, the MGD indicators 
require specific information on improved sanitary facilities indicator. This limited the use of 
secondary data to respond to the objectives. To respond to all the baseline indicators, primary 
data has been used.
Design of data collection tools, translation and piloting:
Ipsos designed the quantitative and qualitative tools used in the study in collaboration with 
WFP. All tools were developed in English and later translated in Kinyarwanda except the 
EGRA tool which was already in Kinyarwanda before adaptation. The tools were then 
programmed onto tablets to facilitate electronic data collection. All tools were pre-tested in 
four schools in the target districts before commencing data collection. Table 2 summarizes the 
tools used in this study.
The standard EGRA tool in Kinyarwanda was shared by World Vision. It was recommended 
by education development partners in Rwanda that the tool go through an adaptation process
to ensure that the students have no prior exposure to the content. Adaptation focused on 
replacing the content of the test but maintaining the standard. A one-day adaptation workshop 
was conducted at WFP using four Kinyarwanda teachers from P1-P3. World Vision and WFP 
representatives were present during this activity. Prior to the adaptation workshop, 
informational interviews were held with education representatives from the US Agency for 
International Development (USAID) and the Education Development Center (EDC). The 
adaptation workshop was moderated by the Ipsos team lead for education. A pre-test was 
carried out before using the tool. In addition to the EGRA tool, students were asked additional 
questions to capture their awareness of health benefits. 
Below is a list of all tools used in this study:
• School facility checklist
17 | Page
• EGRA 
• Student tool (P4-P6)
• Teachers interview tool/ attendance records review
• Store keeper/cook interview tool
• Head teacher interview /document review
• Key informant interview tool
Data collection team and training
A centralized 4 days training was conducted in Kigali. The teams conducted a pilot in their 
respective district before engaging in data collection.
The school based survey was conducted by four teams of 40 enumerators and four supervisors. 
Data collection teams were organized around a district. Key informant interviews were 
conducted by three moderators. Data collection teams were locals, fluent in Kinyarwanda. 
Teams that administered the EGRA had teacher training background and experience in 
teaching lower primary students.
Ethical considerations in the study
• A courtesy call was made to the district education office and Mayor’s office before 
commencing the activity.
• The head teacher consented to the study before any activity is undertaken in school.
• The teacher introduced the enumerators to the class before student selection took 
place.
• Participation was voluntary. Individual consent from student and teachers was sought.
• A warm up session with selected students was conducted before participation.
• All the enumerators were friendly and respectful to the students.
• EGRA tests were coded and individual performance of student was not shared with the 
teachers.
Field dates
Data for the baseline study was collected from June 7, 2016 to June 28, 2016. (School based 
survey data collection ended on June 20, 2016).
18 | Page
3.0 Situation Analysis
3.1 School Feeding Program
Data from the key informant interviews show that there are three school feeding programs in 
the country: One Cup of Milk per Child program by MINAGRI, National School Feeding 
program by MINEDUC, and the Home Grown School Feeding Program by WFP.
Overview of One Cup of Milk per Child Program: 
The One Cup of Milk per Child was initiated six years ago by the Government of Rwanda 
through MINAGRI in collaboration with the Ministry of Finance and Economic Planning. The 
program targets school children in ECD and P1 to P3 in selected schools within selected 
districts e.g. Nyamagabe, Muhanga, Kamonyi, Rulindo, Gacyenke. The program reaches
80,000 children with one litre of milk per week, provided in two servings. The program is 
being expanded to reach more children.
Overview of National School Feeding Program
Targeting secondary school students (9 and 12 years’ basic education), since 2004, the 
National School Feeding program is a community driven project, to which parents contribute 
the food that their children eat at school. The government supports families that are not able 
to contribute any food due to their economic situation. The program covers 35 schools in 
Nyaruguru and 38 schools in Nyamagabe. The parents’ contribution is through cash or cash 
equivalent of firewood or foodstuff such as beans, wheat, maize sorghum or porridge flour.
Overview of Home Grown School Feeding Program 
This school feeding program is part of WFP’s portfolio globally. The Home Grown School 
Feeding (HGSF) Program is one of the seven elements of WFP’s Country Programme, a five￾year country program in which UN agencies design a common program to support the 
government in building capacity in various areas, including food assistance i.e. designing and 
managing food assistance programs. The pilot program was implemented in 24 schools in two 
most food insecure districts - Nyamagabe and Nyaruguru in the south in 2014-2015. During 
the pilot, hot lunches were provided to approximately 25,000 primary school children. The 
meals provided consisted of maize meal, beans, salt and vegetable oil. The maize and beans 
used in the program were locally grown and procured in Rwanda. This program was preceded 
by another model of school feeding until 2013.
19 | Page
The USDA-funded HGSF program will support 83,000 children in 104 schools, including 
schools in two additional districts of Karongi and Rutsiro in the Western province. The 
program will continue to provide a hot meal comprising of maize, beans, salt and oil to selected 
schools in Nyamagabe and Nyaruguru. The program in Karongi in Rutsiro will introduce a 
fortified hot porridge made of Corn Soya Blend- or CSB+ and sugar, provided to students at 
mid-day.
The HGSF program will be implemented by WFP in partnership with World Vision, the 
government, and the community. WFP will be providing maize, beans, salt, oil, CSB+ and 
sugar. The maize and beans will be purchased locally through additional funds raised by WFP, 
while USDA will provide CSB+ and fortified vegetable oil. The program will also provide 
training in food handling, stocking, record keeping for storekeepers and teachers, and 
nutrition and hygiene training for the cooks. It will also construct and rehabilitate kitchens, 
latrines and water systems in target schools; this will be done in two phases starting with the 
schools in the West in 2016 before continuing to those in the South in 2017. 
Government involvement began during inception phase and included input in the selection 
criteria for targeting schools as well as on deciding the types of meals to be provided. The 
government is also involved in supporting the project through dedicated human resources 
(consultants) at the national and district level. The USDA capacity development funds are 
being used by WFP to support these project specific consultants within the government 
apparatus to allow them to effectively manage the program and create a skill base, which would 
then be transferred to permanent government officials over the course of the project period.
These personnel have been hired on WFP’s request, with and on the basis of the agreement 
between WFP and the Government and dedicated solely to supporting the school meals 
activities. They are not government agency bureaucrats and are hired specifically for this 
project. The relevant positions include a project coordinator and assistant at MINEDUC, 
advisor at MINAGRI, and a coordinator in each district. Currently, Nyamagabe and Nyaruguru 
have HGSF coordinators. Karongi and Rutsiro are expected to have recruited people by 
August. 
The government will also support the program by providing office space to allow all the 
different stakeholders to work together, technical assistance through the steering (national 
level) and the technical (district level) committees that will be involved in the governance of 
the program. Tasked with ensuring that the key program targets are being met, the steering 
committee is made up of the permanent secretaries of MINEDUC, MINAGRI, MINALOC, 
Country Director of WFP, Rwanda Agriculture Board (RAB), Rwanda Education Board (REB), 
20 | Page
and World Vision. At the district level, the technical committee and districts coordinators will 
be involved in the day to day implementation and documentation of the program. Quarterly 
update meetings of the technical committee will be used to discuss challenges and ways 
forward with district leadership. MINAGRI will support the schools in setting up school 
gardens in its field. The support will involve training on how to design the school garden and 
also in providing selected seeds. Government involvement will also include support to project 
monitoring. While WFP and WVI have the clear mandate of monitoring project 
implementation, district, sector and cell officials have been sensitized on the program, are 
invested in ensuring that it runs efficiently and will support WFP and WVI field officers with 
their monitoring activities. The Government involvement will facilitate capacity enhancement 
in support of nationally owned school feeding program.
World Vision will implement activities in the area of literacy and water, sanitation and hygiene 
(WASH), whereby their role will be primarily to boost the level of reading in primary school 
students from P1 to P3 and improve the knowledge of and access to WASH. These activities 
are aimed at promoting reading, training of teachers to enhance teacher effectiveness within 
the schools, provision of teaching aid and other reading materials for students, and setting up 
libraries. In the area of WASH, World Vision will support schools with infrastructure such as 
sufficient number of latrines that meet national standards, water systems, and hand washing 
stations. Besides infrastructure, World Vision will also carry out communication on messages 
aimed at promoting knowledge and good WASH practices.
Community involvement in the HGSF program will primarily be through Parent Teacher 
Associations (PTAs) and the HGSF committee made up of parents, teachers, store keepers, 
and cooks. The committee will be responsible for implementation of the program at the school 
level. The community will also be involved through the program’s engagement with small
holder farmers. Through the Purchase for Progress (P4P) program, small holder farmers will 
be organized into cooperatives within the district and their capacity to produce the food 
required for the school feeding will be enhanced through trainings on financial literacy, post￾harvest handling and entrepreneurship. Finally, to enable the schools to cook the daily meals, 
the community will also provide firewood and labour.
Sampled Schools with an ongoing school feeding program 
The baseline study sought to establish if the target schools have an ongoing school feeding 
program. Out of the 40 schools covered in this study, two schools had an ongoing school 
feeding program. One school was under the “One Cup of Milk per Child Program”. The other 
school was one of the 24 schools that participated in the WFP school feeding pilot program in 
21 | Page
Nyamagabe. Although, it had a school feeding program for students in secondary school, at 
the end of the WFP program, there was a gap for primary children from poor households that 
come from far. The head teacher set up a local arrangement, whereby the vulnerable children 
are provided with meals at school. The meals are not provided on daily basis but when the 
school can afford extra food.
WFP and the government collaborated very closely to select the areas of implementation. 
Although the selection criteria for schools to be supported by the HGSF program were 
stringently applied to ensure no duplication of efforts, one school out of the 40 sampled 
schools ended-up with both the one cup of milk per child and HGSF programs. However, since 
the one cup of milk per child only provides a liter of milk in two servings over the course of a 
week to students in grades 1-3, it is seen as complementary to the mid-morning porridge. The 
second school had meals only for students in grades 7-12 i.e. no primary school students were 
receiving meals.
School management support for the school feeding program
All the head teachers and school management (PTA and school management committee) who 
were interviewed affirmed that they would support the school feeding program if introduced. 
According to the PTA members the program was perceived to have the following benefits:
• Supports improvements in school enrollment and retention
• Boosts nutrition
• Improves learning
• Assists parents who are unable to feed their children
The sampled schools’ leadership were asked for their opinion on ways the school community 
(management and parents) would contribute to make such a program successful. Table 3 
outlines areas of support mentioned by PTA, SCM and head teachers. The key among the list 
being provision of fuel, payment of cooks and building of the cooking facilities. The intended 
support by the school community indicates willingness to be involved and a strong sense of 
ownership of the program. Although nearly all sampled schools had a school garden, it was 
not mentioned as a resource that could be utilized to support school feeding.
Table 3: Spontaneous mentions by PTA, SMC and head teachers of areas of support
PTA
School 
Management 
Committee
Head 
teachers
Base 40 23 40
22 | Page
Provision of fuel 67% 57% 38%
Payment of cooks 49% 25% 27%
Building of cooking facilities 49% 46% 14%
Provision of labour 38% 11% 19%
Provision of food 23% 4%
Buying utensils 23% 36% 3%
Mobilize parents to support the program 8% 18%
Provision of water 3% 4%
Come up with community work 3% 4%
Management of the program 11%
Source: Interviews with school leadership
The baseline study also sought to assess the challenges based on the previous program in the 
pilot districts. According to the education officials and program participants the key 
challenges are highlighted as follows:
• Perceived discrimination of some grades and some schools. Targeting only a section of a 
school (grade 1 to 6) and excluding pre-school and secondary school grades or targeting 
some schools and excluding others within the same sector or district that face the same 
challenges is perceived to be unfair. Teachers fear that this could potentially lead to the 
relocation of students to the HGSF program schools, with a feeding program which is likely 
to resulting in some schools being over-crowded others being abandoned.
• Not including the parents’ contributions to food provision may have an impact on the 
sustainability of the HGSF school feeding program.
The schools’ leadership were asked for their opinion of the expected challenges when 
implementing a school feeding program to shed light on some of the priority areas when 
implementing the intervention. Table 4 outlines the list of these challenges. These are based 
on perceptions by the school leadership. A review of the facilities that had been done by WFP 
revealed that 46 percent of the target schools did not have storage and cooking facilities. Only 
10 of the 40 sampled school had a cook in place. Although majority of the schools had a water 
source, supply was not always reliable, validating the concern of school leadership on access 
to water.
Table 4: Summary of anticipated challenges
PTA SCM Head Teachers
Base 40 29 40
Lack of money for the food 85% 79% 48%
Lack of utensils 28% 34% 48%
Lack of cooking area 15% 14% 25%
23 | Page
Lack of storage 15% 17% 15%
Lack of water 13% 14% 10%
Lack of cooks 10% 24% 23%
Uncooperative parents because 
they don’t value the program 8% 3% 3%
Lack of security for the food 3% 7% 8%
Other challenges (below 5%
listed below) 3% 11%
Other challenges specified
• Lack of fuel
• Poor roads(Accessibility of schools)
• Inadequate time ( head teachers combining management of the program with their existing 
roles)
• Poor management of the program
3.2 School Gardens
WFP intends to establish and maintain school gardens in the target schools. Interviews with 
the head teachers established that 35 out of 40 schools (88 percent) had a school garden. All 
schools except one had part of their land under cultivation for food crops. The size of the land 
varied from one school to the other. However, it was clear that schools have access to land 
resource.
Table 5: Summary of school garden acreage in sampled schools
Size of land
Number of 
schools %
20 acres and below 17 43%
21-40 acres 6 15%
Above 40 acres 12 30%
No garden 5 13%
Total 40 100%
Note: Size was based on estimation not actual measurements 
As shown in Figure 1, vegetables are the main food crops grown in the school gardens and a 
limited variety of crops are grown in these gardens. Informal discussions with the head 
teachers during the interview revealed that the schools are limited in land utilization due to 
reliance on rain fed agriculture.
Figure 1: Summary of foods grown in school gardens
24 | Page
Note: Base is 34 because one school was not growing any food crops 
The baseline study sought to assess the estimated yield of each food crop grown based on the 
last harvest. Schools do not maintain records on the school garden activities. Therefore, there 
were no reliable records obtained from the yield estimates. From the discussions with the head 
teachers, it was noted that the harvest is not substantial compared to the claimed land size of 
the garden.
According to the head teachers, in more than half of the 34 schools that were growing crops, 
the food produced in the school gardens do not benefit the students. The food crops are either 
sold (56 percent) or consumed by teachers (15 percent). It was established that in three out of 
34 schools growing food crops students carried home the food crops produced in school 
gardens. 
3.3 Deworming
The current school based deworming activities in the country, including the four target 
districts, are carried out regularly (twice a year) by Ministry of Health (MINISANTE). The 
activity is carried out through health centers in the district under the supervision of the District 
Education and Health Officers. First, a record of all students (for the age group being targeted) 
is taken from the register, then nurses from the nearby health centers visit the school to 
administer the deworming medication, starting with those in the morning shift followed by 
those in the afternoon shift. In each school, the exercise takes about three days to a week to 
ensure that those who are absent are covered, before the health team moves to the next school.
Although it is supposed to be a blanket intervention, there appear to be some discrepancies in 
3%
6%
6%
24%
24%
26%
29%
35%
56%
Sweet potatoes
Wheat
Mushroom
Potatoes
Bananas
Carrots
Maize
Beans
Vegetables
Base: 34
25 | Page
data provided by MINISANTE and the districts. In 2015, 89 primary schools in Nyaruguru 
and 77 in Nyamagabe were targeted. 
The community health mobilizers are used to sensitize the community about the exercises as 
well as inform them about the benefits of the medication. Awareness is raised through parents 
through their children before the exercise begins and using community platforms.
Deworming is a component of the WFP program and the plan is to support MINISANTE 
through advocacy and awareness creation activities. While the ministry covers the entire 
country, the advocacy and awareness activities by WFP will be limited to its four program
districts.
Challenges faced by deworming in schools
Based on the key informant interviews, challenges faced by schools include religious beliefs, 
(where religion is opposed to conventional medicine) and misconception associated with 
medication given for deworming. Key informants revealed that due to these issues some 
parents keep the children away from school during the period of the activity to avoid being 
dewormed.
Sometimes children refuse them, therefore, he/she abstains himself/herself from coming to 
school during that whole week and he/she says it is the parents who have done that. That is 
the reason we involve the community health mobilizers. That way they can’t miss school 
within that activity, since community health mobilizers are very close to people*. DEO
*Paraphrased to capture meaning
Suggestions for improvement 
Related to the challenge above, there is the need for more sensitization and raising awareness 
around deworming prior to any deworming activity so that parents are well informed. 
Although findings reveal that WFP will be involved in advocacy and awareness creation, the 
scope will be limited to the four HGSF program target districts.
“I think that if one thing can be improved, it is communications, ---on time… “Yes! and to 
explain it to the community and everyone know his part and do it, if it is the head of village 
he has to know it in order to inform people from this village. If it is that one of cell or if it is a 
teacher, director of a school the health centre…they have to know it before and make sure 
that all material needed are available because it is him the 1st implementer who has to take 
the 1st step (that of health centre) and others follow him, what I’m saying is that they have to 
improve the communication.” DEO
26 | Page
3.4 Other organisations working with the primary schools
A review of the materials and discussions with key informants and head teachers established 
that there are a large number of players working with schools in different areas. Table 6 maps 
these stakeholders by intervention.
WFP is the Government of Rwanda’s primary partner for all school feeding activities. The 
project was carefully designed and schools selected to ensure that no duplication of efforts 
takes place. If at all, there would be only complementary activities. In addition, the HGSF 
programme has established a Steering Committee and Technical Committee and the HGSF 
partners are members of the Education Sector Working Group, the Education Development 
Partners Group and the Agriculture Sector Working Group.
27 | Page
Table 6: Mapping of stakeholders working with primary schools by 
intervention
Program Organization Activities District
Some of the schools 
where activities are 
conducted
School feeding
(Although the pilot 
program ended in 
December, these two 
schools still feel that 
WFP is current player)
World Food 
Programme
(WFP)
Providing food 
and utensils
Nyamagabe Rugogwe GS
Nyaruguru Ruhinga PS
Other nutrition activities ADENYA
Providing 
mushrooms 
seeds
Nyaruguru Ruhinga PS
Sanitation (water and 
toilets)
Grandshule 
planning
Construction of 
toilet Karongi EP Gitanga
World Vision
Provision of 
hand washing 
materials called 
Kandangira 
ukarabe
Nyamagabe Rugogwe GS
Kiyumba PS
Kibirizi PS
Construction of 
toilets Nyamagabe Rugogwe GS
REMA
Provision of 
hand washing 
system called 
Kandagira 
ukarabe
(Tread-on and 
wash), water 
tanks, whoes.
Nyaruguru Ruheru GS
PEPAPS
Provision of 
water Nyaruguru Ruhinga PS
Kiyonza GS
EP Ruhororo
Construction of 
toilets Nyaruguru Fugi GS
Kiyonza GS
Concern
Provision of 
hand-wash 
system called 
Kandagira 
ukarabe (tread 
on and wash)
Nyaruguru Kimina PS
Health Poverty 
Action (HPA)
Construction of 
toilets and 
room for girls
Nyaruguru Kivuru PS
EP Ruhororo
28 | Page
School governance
Concern
Training on 
school 
management
Nyaruguru
Nyamagambe
Zirambi PS
Ruheru GS 
Kimina PS 
Rugerero GS 
Kiyonza GS 
Kivuru PS
EP Gitanga
Ep Kinyami
GS Ngoma
Training PTA 
members
Nyamagabe Rugogwe GS
Kiraro Protestant GS
Training 
teachers
Nyamagabe
Nkore PS
Training school 
administrators
Nyamagabe
Masagara PS
Action Aid
Training 
teachers and 
PTA
Nyaruguru Ruheru GS
VSO
Training 
administrative 
staff on 
administration 
methodology
Nyaruguru Ruheru GS
ADRA
Training of 
school 
administrator
Karongi EP Kinyovu
GS Shoba
Provision of school 
materials, textbooks, 
book
Ministry of 
Education
Provision of 
books and text 
books
Rutsiro EP Mwufe
GS Kivumu
EP Rutsiro
REB
Provision of 
books and text 
books
Rutsiro EP Nyakarera
Concern
Provision of 
books Nyamagabe Kiyumba PS
Nyampinga
Provision of 
books Nyaruguru Kiyonza GS
EDC
Provision of 
books and text 
books
Rutsiro EP Umubano
29 | Page
HDP (Health 
Development 
and 
Performance)
Training on 
sexual 
reproductive 
health
Nyaruguru Kiyonza GS
Renovation/construction 
of infrastructure in 
school e.g. classes, 
kitchens, stores
World Food 
Programme
(WFP)
Construction of 
kitchen and 
store
Nyaruguru Ruhinga PS
Renovation of 
store and 
kitchen
Nyaruguru
Kivuru PS
Construction of 
the Kitchen
Nyaruguru
EP Ruhororo
World vision
Construction of 
classrooms Nyamagabe
Baro PS
Rugogwe GS
Kiyumba PS
EP Kamegeri
Action Aid
Construction of 
new class 
rooms and new 
closed water 
holes
Nyaruguru Gakaranka PS
Construction of 
three new 
classrooms
Nyaruguru Ruheru GS
Langenlonshein
Construction of 
classrooms Rutsiro EP Mpingamabuye
Catholic church
Construction of 
classrooms Nyamagabe Kibirizi PS
District
Construction of 
new class 
rooms
Nyaruguru Kiyonza GS
Training of teachers
Concern
Training on 
how to make a 
new curriculum
Nyamagabe Kiyumba PS
Training on 
literacy Nyaruguru Kivuru PS
VSO
Training on 
administration, 
teaching and 
learning 
process
Nyaruguru Ruheru GS
Training on 
teaching tools Nyaruguru Ruhinga PS
30 | Page
UNFPA
Comprehensive 
sexuality 
education
Nyaruguru Kiyonza GS
USAID
Training on 
how to make a 
new curriculum
Nyamagabe Baro PS
Health education
Ministry of 
Health
Reproductive 
health for girls Nyaruguru Kimina PS
HPA
Sexual and 
reproductive 
health training
Nyaruguru Kivuru PS
Concern
kandagira 
ukarabe Karongi EP Gitanga
HDP (Health 
Development 
and 
Performance)
Training on 
sexual and 
reproductive 
health
Nyaruguru
Zirambi PS
Ruheru GS
Rugerero GS
Kiyonza GS
Fugi GS
Village health
Training 
teachers and 
students in 
health
Nyamagabe Rugogwe GS
3.5 Education promotion Activities at community level 
The program was keen to learn if the school leaders are involved in education promotion 
activities at the community level. The PTA and SMC were asked about education promotion 
activities they engage in outside the school. Table 7 summarizes the spontaneous mention of 
these activities. The main role that the PTA and SMC engage is in mobilizing parents to enroll 
children and increasing sensitization on girl child education. Members of SMC seem to be 
engaged in community libraries.
Table 7: List of education promotion activities that PTA and SMC are engaged in at 
community level
PTA SMC
Base (those that are engaged in education promotion activities) 26 16
Following/encouraging parents to take children to school 62% 56%
Sensitizing families and communities to promote girl child education 50% 81%
Reading clubs 27% 31%
Recognition of high performers 27% 38%
Establishing community libraries 12% 50%
Education to parents on technology 8%
**Base is very small. List is interpreted qualitatively
31 | Page
3.6 Teachers reward and recognition program 
Out of all the 40 schools visited, all except two reported to have an existing teacher reward and 
recognition scheme. The scheme focuses on incentivizing teachers based on academic 
performance of the students. However, the scheme does not appear to be active in all schools. 
When asked how many teachers were awarded in 2015, 12 schools did not have any awardee
and only 46 teachers received some award in 2015.
3.7 Health Clubs 
Thirty-five percent (35 percent) of the head teachers of visited schools reported that there was 
an existing Health Club in the school. Children in the Health Clubs are mainly involved in the 
training of health matters after which they sensitize other children on health issues. Most 
schools did not have any recent documentation verifying the membership. The data was used 
as it is with clarification from the head teacher or teacher involved. It was estimated that health 
clubs in the 14 schools had 658 student members; 360 girls and 298 boys. The average 
membership in a school is 47 students; 26 girls and 21 boys.
3.8 Education Policies 
Currently the only school feeding related policy is the national school feeding policy which is 
in the initial stages of formulation and is being drafted by MINEDUC in partnership with 
MINAGRI, with technical support from WFP. The objective of the policy is to ensure the 
effective and efficient implementation of a national school feeding program. This includes, but 
is not limited to, providing regulations, setting standard for service delivery, establishing 
appropriate institutions, and strengthening the governance, partnerships and multi-sectoral 
coordination of school feeding. 
32 | Page
4.0 Results for Literacy Indicators (Results Framework #1)
This section presents findings of the baseline literacy indicators. Not all indicators could be 
measured at baseline because they are based on program activities (see Annex 1). These will 
be tracked during the program implementation.
MGD 1.1: Improved quality of literacy instruction
Indicator Baseline Target
Number of teachers in target schools who attend and teach at 
least 90% of scheduled school days per school year
986 653
Data for this indicator was collected through a review of teachers’ attendance records in the 
40 sampled schools and then estimated for target schools. The records indicated if the 
teacher was present or not. These records did not specify if the teacher was present for the 
whole day or part of the day. Each teacher’s record was reviewed for the whole year in 2015. 
Where the teacher had taught for less than one year in the school, the base was the total 
days the teacher was in the school. Although the indicator requires both attendance and 
teaching, there were no records on class attendance. It is expected that teachers in school 
attend classes. Therefore, this baseline survey assumed that the numbers of days a teacher 
was in attendance is equivalent to the number of days taught.
A total of 543 teachers’ records in the sampled schools were reviewed for the year 2015. At 
baseline, 68 percent of the teachers attended and taught at least 90 percent of scheduled 
school days in 2015 school year. Male teacher attendance was at 80 percent compared to 
female at 57 percent. According to DEOs in the four districts, the target schools have 1450 
teachers. It can therefore be estimated that 986 teachers, attended and taught at least 90 
percent of scheduled school days in 2015 school year.
Indicator Baseline Target
Percentage of students who by the end of two grades of primary 
schooling, demonstrate that they can read and understand the 
meaning of grade level text.
49% 90%
Percentage of students who by the end of two grades of primary 
schooling, demonstrate that they can read and understand the 
meaning of grade level text (male). 
42% 90%
Percentage of students who by the end of two grades of primary 
schooling, demonstrate that they can read and understand the 
meaning of grade level text (female). 
57% 90%
33 | Page
This indicator looks at the proportion of those children in P3 who have acquired the basic 
literacy skills. This indicator was measured using the scores of the literacy on the EGRA test. 
All the students were assessed using the same standardized procedures that allows for 
comparisons of the scores and therefore periodic monitoring to enable comparison with the 
base value.
On each of the task, the learners were presented with 100 phonemes and syllables, 50 invented 
words and short familiar words, and given a short text of 66 words to read. They were assessed 
on their ability to read and also comprehend the text read. 
The students were given one minute to read as many words as possible at each of the five 
levels of the assessment test, which included:
1. Phonemic awareness (understanding the sounds letters making words in a particular 
language, naming the beginning sound of a word); the learners were presented with 100 
letter sounds in Kinyarwanda.
2. Reading syllables (a consonant and a vowel together—e.g. ba); the learners were presented 
with 100 commonly found Kinyarwanda syllables (combination of the phonemes of 
consonants and vowels). 
3. Reading short, familiar words; these are words commonly used in Kinyarwanda. The 
learners were presented with 50 words.
4. Reading invented words (a way to assess decoding skills). Invented words are words that 
lack meaning in a given language but follow the same morphological structure of the same 
language that include syllable formation and how syllables combine to make words. 
Invented words assess whether the children have acquired automaticity (ability to read 
any word in a given language). That means that, for a child who has mastered the principles 
of word formation in a given language, they would read any word in that language 
regardless of whether that word is new. This task presented such 50 words to the learners. 
5. Reading words in a text. This is the measure of reading fluency. Fluency in reading is 
important in helping learners to comprehend facts. For fluency to occur, a child should 
read without skipping or omitting words in a text. Furthermore, the child should read a 
certain threshold of words. In addition, the child should not read words like a string of 
sentences and should not replace words in a text. Often, children who cannot read fluently 
will struggle to comprehend the facts in the text although the vice versa may not necessarily 
happen. The fluency reading was administered by giving the children a text in prose of 65 
words and they were given a minute to read the text. 
34 | Page
This indicator was specific to the percentages of the children who could read a text and 
comprehend at the end of the grade of two. Since the baseline study was conducted in the 
middle of an academic year, the results present the scores for the students currently enrolled 
in grade 3, although learners in grades one and two also undertook the assessment. 
This indicator was measured using the last task on the EGRA test that involves reading words 
in a text. This is a composite indicator that involved those students who were able to not only 
read but also comprehend the inferred meaning from the text. This was analyzed by focusing 
on the proportion of the children who were able to answer the third comprehension question.. 
Overall, 49 percent of the children by the end of grade 2 can fluently read and understand the 
meaning of the same grade level text. These are the learners who were able to read and answer 
a question that required inferring meaning from the text. 
When comparisons are made across the two genders, girls outperform boys. On average, 57
percent of the girls read and comprehended compared to 42 percent of the boys. This was the 
situation in all the districts where the learners were tested. The analysis by gender are 
presented in Table 8 below. 
Table 8: Analysis of literacy indicator by gender
Gender N %
Boys 83 42%
Girls 114 57%
Total 197 49%
Base: 200 children per gender 
As shown in Table 9 below, learning levels and outcomes are highest in Nyamagabe (65 
percent), followed by Karongi (62 percent), Rutsiro at 37 percent and least in Nyaruguru (33 
percent). The proportion who are competent in reading and comprehending is almost twice in 
Nyamagabe (65 percent) than in Nyaruguru (33 percent). In Nyamagabe and Karongi, more 
than a half of the learners can read a grade 2 text and comprehend an inferred question. On 
the other hand, 67 percent of the P3 students in Nyaruguru and 63 percent of the children in 
Rutsiro cannot read and comprehend a P2 text. 
Table 9: Analysis of literacy indicator by district
District N %
Nyamagabe 65 65%
Karongi 62 62%
Nyaruguru 33 33%
Rutsiro 37 37%
Total 197 49%
Base: 100 children per district 
35 | Page
How learners in Grade 3 performed in the other EGRA Tasks 
The EGRA test also has other tasks. The tasks included the letter sounds, syllables, and 
familiar word reading and invented word reading as discussed earlier on. 
REB and the Language, Literacy and Learning (L3) initiative have been working on developing 
literacy-reading benchmarks. However, these benchmarks are only applicable to the words 
read and not specific on whether they are individual words or words in a text. 
The following facts emerge from these findings:
1. Even in the absence of benchmarks for the other literacy skills in Kinyarwanda, P3 students 
are able to read 16 letter sounds on average (per minute), decode 46 syllables per minute, 
decode 15 correctly invented words (per minute) and correctly read 21 familiar words per 
minute. 
This demonstrates the following
1. The P3 learners are struggling to read letter sounds compared to the 
syllables. 
Phonemic awareness is low compared to decoding syllables. However, it can be speculated that 
the children are struggling to read the sounds based on the nature of the Kinyarwanda 
language. Although the foundation of literacy is the grasping of sounds, this seems not to be 
the case with the target learners. It is evident that learners are struggling to read the letter 
sounds. In fact, the scores in syllable reading are higher than the scores in letter sound reading. 
This could be perhaps due to the nature of reading the sounds that is often accompanied by 
decoding by adding a vowel and therefore consonants rarely being read independently. For 
instance, whereas the learners decoded 46 syllables correctly per minute, the children could 
only decode and identify 16 letter sounds. 
Table 10: Performance on reading sounds and syllables
No Literacy Skills Average Performance 
(per minute)
No of tasks Benchmark 
1 Reading Sounds 16 100 None 
2 Syllables 46 100 None 
2. Children have not mastered the word formation principles as measured 
by the number of invented words correctly read 
36 | Page
Ability to read invented words is a clear characteristic of developed automaticity and ability to 
decode any word in Kinyarwanda. However, when words are read (whether independently or 
in a text), children are decoding fewer invented words. The ability to read invented words 
reveals the extent to which children have mastered the principles in word formation in a 
language. However, the analysis shows that on average, P3 children read 21 words per minute 
(wpm) (familiar words), 15 wpm (invented words) and 20 wpm (words in a text). It is therefore 
possible that the children are able to identify and read the familiar words without necessarily
mastering the principles behind those word formations. This comparison is presented in Table 
11 below. 
Table 11: Average overall P3 performances against the expected reading level
No Literacy Skills Average Performance 
(per minute)
No of tasks Benchmark 
3 Familiar Words 21 50 None 
4 Invented Words 15 50 None 
5 Words in a text 20 65 20-32
3. Girls in P3 are performing better than the boys in literacy skills
Further, analysis by gender of the student reveals that the girls are performing better than the 
boys across the literacy skills. On average, girls read three more letter sounds than the boys, 
decoded 13 more syllables, read six more familiar words, five more invented words and seven 
more words in a text than the boys. Basically, on any given literacy task, the girls performed 
better than the boys. 
4. The performance in any literacy skill vary across the districts; highest in 
Nyamagabe District and lowest in Nyaruguru District in any literacy skill
The detailed performance for the P3 by district is presented in Figure 2 below. It is evident 
that the performance in Nyamagabe is the highest when making comparison across the four 
districts followed by Karongi, Rutsiro while the learning levels are the lowest in Nyaruguru. 
37 | Page
Figure 2: P3 detailed literacy performance by district:
Base: Total=400, each district=100 P3 students
5. There are non-readers in P3 in all the literacy tasks
This was measured by establishing the proportion of the children with 0 scores in each task. 
For instance, 5 percent of the P3 students cannot decode any letter sound, 14 percent of the 
same P3 students cannot identify any Kinyarwanda syllable. Furthermore, 27 percent of the 
students in this class cannot read any familiar word in Kinyarwanda while almost a third (29 
percent) cannot read any invented word. When presented with words in a text, almost a 
quarter (24 percent) of all the learners in P3 cannot read a single word. This brings to the fore 
the reality that many children are struggling to learn the basic skills and are graduating to 
higher levels without acquiring the basic reading skills. This is illustrated in Figure 3 below.
Figure 3: Proportion of Non-readers in P3 by Task
16
50
25
17
24 22
60
27
19
26
13
35
14 10 13 15 15 18
12 16 16
46
21
15
20
Sounds Syllables Familiar words Invented words Words in Text
P3 Reading Scores by Task 
Karongi Nyamagabe Nyaruguru Rutsiro Total Average
5%
14%
27%
29%
24%
Letter Sounds
Syllables
Familiar words
Invented words
Words in a text
Base: P3 sampled students (400)
38 | Page
MGD 1.2: Improved Attentiveness
Indicator Baseline Target
Percentage of students in classrooms identified as attentive by 
their teachers
60% 80%
Data for this indicator was collected using the perception of teachers of sampled classes. A 
show card with a ladder of attentiveness was used as a projective technique to collect this 
information. The ladder was divided into four equal segments ranking the lowest segment to 
the highest segment. These segments represented four levels of attention as follows.
• lowest segment of ladder (no attention),
• second segment of ladder (low attention),
• third segment of ladder (medium attention) and 
• fourth segment of ladder (high attention)
The teacher was asked to place the students in the class on this ladder of attentiveness. For 
this indicator, the third and fourth segment was combined to represent attentiveness. An 
average score of all classes was used to calculate the percentage of attentive students.
Considering the third and fourth segment as the attentive students, the percentage of students 
considered to be attentive by teachers is 60 percent. 
Figure 4: Percent distribution of student across the attentive ladder
Note: The assessment of this indicator is based on a subjective approach that is founded on the 
perception of class teachers and the assumption that the teachers are aware of student attentiveness 
in class
1 15% 25% 36% 24%
Base: All students in sampled classes (10, 397)
Lowest ladder(No attention) Second ladder(low attention)
Third ladder(medium attention) Fourth ladder(high attention)
39 | Page
Indicator Baseline Target
Number of students enrolled in schools receiving USDA 
assistance
82,358 127,650
Number of students enrolled in schools receiving USDA 
assistance (female)
41,404 66,378
Number of students enrolled in schools receiving USDA 
assistance (male)
40,954 61,272
The indicator counts students enrolled at schools that are directly benefiting from USDA 
assistance. For this indicator, USDA assistance to schools includes the provision of 
commodities for school feeding and/or the rehabilitation of school infrastructure. The focus 
of this indicator is 104 schools that have been pre-selected to participate in the home grown 
school feeding program. Data for this indicator is required to be disaggregated by gender.
Based on the secondary data by WFP the total number of students enrolled in schools selected 
to receive USDA assistance is 82,358 students. This considers grades 1-6 only. Data was not 
disaggregated by gender in 13 schools. Based on schools that had gender disaggregation, the 
mean gender split is approximately 50 percent (50.2 percent female vs. 49.7 percent male). 
Applying the universal gender split, the estimated number of female students enrolled in these 
schools are 41,404 and 40,954 for male students. 
MGD 1.3: Improved Student Attendance
Indicator Baseline Target
Number of students regularly (80%) attending USDA supported 
classrooms/schools.
75,522 127,650
Number of students regularly (80%) attending USDA supported 
classrooms/schools (female).
37,964 53,102
Number of students regularly (80%) attending USDA supported 
classrooms/schools (male).
37,512 49,017
This indicator measures the number of boys and girls attending school regularly. According to 
the definition of this indicator, it requires the analysis of data for each student during normal 
school hours for a given school year and establish that each student was in school at least 80 
percent of the school days. Desk research established that data on attendance is not 
summarized by student per year. Therefore, this indicator could not use secondary data at 
district or sector level. To respond to this indicator, it would have required analysis of more 
than 100,000 records of students using the attendance records at school level. Given time and 
resource constraints, alternate methodology was used.
40 | Page
To obtain a proxy for this indicator at baseline, the sampled classes were used and their 
attendance reviewed for the first seven weeks in the second term (April 18 to June 3, 2016). 
Using the same standard of 80 percent attendance, the baseline study counted the number of 
students who have not been in school for seven or more days during this period. A total of 
10,397 students’ records were analyzed for this period, which constituted 13 percent of the 
population records. Based on the reviewed records, 91.7 percent of the students regularly 
attended classes; 90.6 percent for boys and 92.7 percent for girls. When projected to the total 
population of target students of 82,358, it can be estimated that 75,522 students regularly 
attended USDA supported schools. When disaggregated by gender, 37,964 girls and 37,512 
met the indicator.
Indicator Baseline Target
Percent of students who miss more than 10 school days per year 
due to illness (female)
8% 20%
Percent of students who miss more than 10 school days per year 
due to illness (male)
7% 20%
Similar to MGD indicator 1 described above, this data used students’ records for a period of 
seven weeks. An ordinary school year is 180 days. The indicator is measuring absence of 5.6
percent or more of the school days. This is equivalent to more than two days for 35 school days. 
Therefore, the enumerators counted the students in each sampled class who had missed more 
than two days during the last 35 school days due to illness. The percentage was calculated in 
each class and overall statistic was generated by gender.
Based on 10,397 students’ records studied for seven weeks, overall, 7 percent of students 
missed more than 2 school days during this period because of illness. Girls recorded 8 percent
and boys 7 percent. On the day the schools were visited, overall, 15 percent of the enrolled 
students in sampled classes were absent. The proportion was higher for boys at 17 percent
compared to girls at 14 percent.
41 | Page
5. 0 Results for Health Indicators (Results Framework #2)
This section presents the results of the indicators on health. 
MGD 2.1 Improved Knowledge of Health and Hygiene Practices
Indicator Baseline Target
Percent of students who can identify at least three key health and 
hygiene practices (female)
47% 80%
Percent of students who can identify at least three key health and 
hygiene practices (male)
48% 80%
Information for these two indicators was collected through interviews with students from P1 
to P6. The students were asked to mention at least three key health and hygiene practices
unprompted. Overall 48 percent of students could identify at least three key health and 
hygiene practices. There was no major variation across gender. However, there was a 
statistically significant difference between lower classes and upper classes (p value 0.00).
Figure 5: Percent of students who can identify at least three key health and hygiene 
practices by gender and grade
MGD 2.2 Increased knowledge of safe food preparation and storage 
practices
Indicator Baseline Target
Percent of cooks and store keepers who can identify at least three 
safe food preparation and storage practices.
92% 95%
The data for this indicator was gathered through interviews with store keepers and cooks in 
sampled schools. A total of 29 store keepers and 10 cooks were interviewed. Although only two
of the sampled schools had a school feeding program, eight school had cooks to serve the 
teachers.
48% 48% 47%
76% 76% 77%
19% 21% 17%
Total Boy Girl Total Boy Girl Total Boy Girl
Total (upper primary P4-P6) Lower primary(P1-P3)
42 | Page
The respondents were asked to mention safe food preparation and storage practices
unprompted. The baseline focused on the following:
1. Using clean containers to collect food from the store
2. Removing foreign matters from food before cooking, and then washing it with clean 
water thoroughly 
3. Check expiry date of food before cooking
4. Hand washing before handling food and often during food preparation
5. Cleaning the kitchen and working area 
6. Storing cooked food in covered cooking pots in a clean, safe place before serving the 
students
7. Safe storage of food
• Taking measures to prevent food from contamination from pests and rodents
• Storing food off the ground 
8. Washing cooking utensils and materials with water and soap
If a respondent mentioned three, he/ she was given a score of one. As shown in Figure 6, 92
percent could mention at least three safe food preparation and storage practices against a 
target of 95 percent.
43 | Page
Figure 6: Percent of cooks who cooks and store keepers who could mention at least 3 
safe and food and storage practices
MGD 2.4 Increased Access to Clean Water and Sanitation 
Indicator Baseline Target
Number of schools using an improved water source 62 412*
*target that was agreed upon in signed project document – needs to be revised
Indicator: 
Information for this indicator was gathered from the head teacher interviews and 
observations. For a school to fulfil the indicator criteria, it should comply with the following:
1. Main water source for the school was from the following sources:
• Piped water into dwelling, plot, or yard
• Public tap/standpipe
• Tube well/borehole
• Protected dug well
• Protected spring
• Rainwater collection
2. Water is normally available from the identified source(s) 
3. Water was not unattainable from the identified source(s) in the past two weeks for a day or 
longer 
Based on sampled schools 60 percent (24/40) are using an improved water source. This 
translates to 62 schools at baseline, which had an improved water source.
Percent who 
could name at 
least 3 safe food 
preparation and 
storage practices, 
92%
Percent who could not 
name at least 3 safe food 
preparation and storage 
practices, 8%
Base: 39
sampled cooks and store keepers
44 | Page
The study explored further the nature of water access in schools. Reliability of water source is 
where there is a slight challenge. Even though 35 schools had an improved water source, only 
28 had water every day in the two weeks preceding the baseline survey.
Figure 7: Safe water access in schools
*Base: 40 sampled schools
*The first bar represents the indicator as defined by USDA: a composite variable that has to meet the three criteria listed 
above. 
Indicator Baseline Target
Number of Schools with Improved Sanitary Facilities 94 412*
*target that was agreed upon in signed project document – needs to be revised
This indicator measures whether there are adequate sanitary facilities at the school and 
whether that sanitary facility meets the improved sanitation standards defined in the 
Millennium Development Goals (MDGs). To be considered adequate, the school must have 
separate improved sanitation facilities available for the use of both males and females. In 
defining this indicator only toilets for students were used. An improved sanitation facility was 
defined as: 
• Flush or pour/flush facilities connected to a: 
• Piped sewer system 
o Septic system 
o Pit latrine 
• Pit latrines with a slab 
• Composting toilets 
• Ventilated improved pit latrines 
24
35 33
28
Number of schools with
improved water source
(composite indicator -
meeting the following three
criteria)
Number of schools whose
main source of water is
improved
Number of schools where
water is normally available
from this source
Number of schools where
water was available all days
in the previous two weeks
45 | Page
To meet the criteria, the school had to fulfill the following two conditions:
• Has separate boys’ and girls’ toilets
• The type of toilet was classified as improved sanitation facility
Based on this definition, out of the 40 schools visited at baseline 36 (90 percent) had improved 
sanitary facilities. Baseline value for this indicator for all target schools is 94 schools out of 
104 program schools.
Figure 8 summarizes the status of sanitation facilities in schools. Out of the 40 schools visited, 
three schools did not have separate toilets for teachers (one in Nyaruguru, one in Rutsiro and 
one in Karongi)1
. The teachers and students shared the same toilets. Two schools in 
Nyamagambe did not have separate toilets for boys and girls.
Figure 8: Status of sanitation in sampled schools
Looking at the number of users per toilet in sampled schools the toilet ratio i.e. number of 
students per stance was 77:1 for girls, 84:1 for boys and 80 overall. Number of users per toilet 
was much higher compared to the national toilet ratio in Rwanda which is 54 for girls, 56 for 
boys and 55 overall.2 Long ques were observed during break time. Toilets were not adequate 
to cater for the school population. 
1 The specific names of schools will be shared with program staff in a separately. Consent form administered to 
head teachers promised that identity of the school names would be revealed in the report. 
2 Ministry of Education; 2015 Educational Statistical Yearbook, June 2016.
36
14
3 2 1 1
Schools meeting
the indicator￾Improved
sanitation
Schools where
toilets were not
clean (visible feacal
matter)
Schools with no
separate toilet for
teachers
Schools with no
separate toilet for
girls
Schools where girls
toilets have no
walls
Schools where girls
toilets have no
doors
Base: 40 Sampled schools
46 | Page
MGD 2.5 Increased Access to Preventative Health Interventions
Indicator Baseline Target
Number of students receiving deworming medications 75,749 127,650
This indicator measures the number of students in a fiscal year that have received deworming 
medication(s), usually through the distribution of deworming tablets at school. This data was 
collected through interviews with the head teacher and review of documents of deworming 
exercise. The information was also collated from key informant interviews with the District 
Education Officers (DEOs). 
Based on reports from head teachers of sampled schools, all schools had the students 
dewormed. However, focusing on the past one year, a total of 29,839 students representing 92 
percent of the enrolled students in the sampled schools had been dewormed. Total number of 
students enrolled in target schools are 82,358. Projecting to the total population enrolled in 
the target schools, baseline estimates that the number of students receiving deworming are 
75,749.
McGovern-Dole definitions recommends use of recommended medications, doses and 
treatment guidelines stipulated by the World Health Organization (WHO) for use in large￾scale school deworming programs. Baseline study was limited by access to the specific 
guidelines for the target district. However, there are indications that students in Rwanda are 
at high risk of worm infestations. The District Education Officers confirmed that deworming 
is supposed to be done twice a year. The baseline study assumed a standard of twice a year as 
recommended by WHO for soil-transmitted helminths using Albendazole (400mg); 
Mebendazole (500mg), or Levamisole (80mg).34 In addition, baseline was limited on the 
dosage used. Given that Ministry of Health is solely in charge of this initiatives, baseline 
assumed that the correct dosage was used.
Based on findings from head teachers of sampled schools 65 percent (26/40) of the schools 
were conducting deworming exercise at least twice a year. None of the schools visited in 
Karongi districts met this criterion. Most of the schools visited in Rutsiro districts were 
carrying out deworming exercise four times in a year.
3 http://www.who.int/features/2015/rwanda-deworming-campaign/en/
4 Food for Progress and McGovern-Dole Indicators and Definitions, Food Assistance Division, Offices of 
Capacity Building and Development
47 | Page
Table 12: Frequency of carrying out deworming activities in sampled schools as 
reported by the head teacher.
Frequency 
of 
dewormin
g exercise
Total Karongi
Nyamagab
e
Nyarugur
u Rutsiro
Number 
of 
mentions %
Number 
of 
mention
s
Number
of 
mentions
Number 
of 
mentions
Number 
of 
mention
s
Every 3 
months 7 18% 7
Twice a year 19 48% 10 8 1
Once a year 13 33% 10 2 1
Was a one￾off event 1 3% 1
Total 40
100
% 10 10 10 10
Caution: There was a contradiction with information obtained from KII with DEOs on frequency of deworming. DEOs 
reported a higher frequency of at least twice a year.
MGD 2.6 Increased Access to Requisite Food Prep and Storage Tools and 
Equipment
Indicator Baseline Target
Number of target schools with access to improved food 
preparation and storage equipment (kitchens, cook area, 
storerooms, stoves and kitchen utensils)
10 104
Based on secondary data from information collected earlier by WFP in the 104 target schools, 
48 schools (46 percent of total target schools) did not have cooking and storage facilities5
. 
However, this data does not clearly capture the whole indicator. Therefore, data for this 
indicator was collected from structured observations in sampled schools. To meet the 
indicator, the school had to have the following: 
• School has a kitchen 
• Cooking area is clean 
• Cooking area has an improved food preparation area (raised area for food preparation) 
• School has a fuel-saving stove
• School has an improved storage are (store exclusively for food)
5
List of schools in the 4 district selected to participate in the program by WFP.
48 | Page
Based on this criteria, only four of the sampled schools (10 percent) have access to improved 
food preparation and storage equipment (kitchens, cook area, storerooms, stoves and kitchen 
utensils
Projected to all target schools, the baseline value for this indicator is 10 schools.
Figure 9 illustrates that 58 percent of schools have an existing kitchen. Similar to available 
secondary data, all schools had space for constructing a kitchen.
Figure 9: Summary of schools with specific features of this indicator
Base: Sampled schools (40)
10%
58% 55%
37%
23% 20%
Schools that meet
the indicator
Schools with a
kitchen
Schools with a
fuel-saving stove
Cooking area is
clean
Schools with an
improved storage
area (store
exclusively for
food)
Cooking area has
an improved food
preparation area (
raised area for
food preparation)
49 | Page
6.0 Additional Findings
6.1 Schools with an ECD Centre and feeding programs
Out of the 40 schools visited 22 (55 percent) had an ECD center, however, only one school had 
a school feeding program for these children. The ECD center provided children with hot 
porridge and the feeding program was financed by parents.
6.2 Involvement of students in writing competition
Out of the 40 schools visited, students in 13 of them participated in writing competition 
according to the head teachers. Although the ratio of the student is 1:1 between male and 
female, there were more girls participating in the writing competition compared to boys.
Table 13: Distribution of the number of students that participated in the last 
competition by sex and class
Grade
Number of 
boys
Number of 
girls Total
P1 7 8 15
P2 6 6 12
P3 7 5 12
P4 7 10 17
P5 10 15 25
P6 7 17 24
44 61 105
6.3 Schools with Access to a Hand Washing Facility
Although not a McGovern-Dole requirement, the baseline study sought to establish the 
schools’ access to hand washing facility. As shown in Figure 10, only 50 percent of the 
schools hand a functional hand washing facility.
50 | Page
Figure 10: Schools’ Access to a hand washing facility
% of schools with 
no hand washing 
facility, 43%
% of schools with a 
hand washing 
facility but no 
running water, 7%
% of schools with a 
hand washing 
facility with running 
water, 50%
Base: 40 sampled schools 
51 | Page
7.0 Limitations of the Study
This baseline study was conducted within the limitations cited here below:
1. Timing of the EGRA exercise: The indicator requires data to be collected at the end of 
grade two. The ideal timing for this indicator should have been the end of the academic 
year or first week of the new term. By the time this study was commissioned and data 
collected, P3 students had covered seven weeks in second term. On the other hand, P2 
students had not yet completed P2 syllabus.
2. Secondary data: The ToR anticipated that the baseline study would heavily rely on 
secondary data and minimal use of primary data. A review of available secondary data and 
corresponding indicators revealed that definitions of indicators required specific 
information, a gap that could not be filled through secondary data. Aggregated secondary 
data from schools at sector and district level were not in a format that could be utilized in 
this study, thus creating the need for this data to be collected from all the target schools. 
The study was limited to 40 sampled schools and estimates have been projected to the total 
population.
3. Incomplete secondary data: Data on teachers and students’ records were not 
comprehensive in indicating reason for absenteeism for all days. Given this limitation, 
WFP and World Vision have redesigned the class register to facilitate data collection on 
attendance for the midline and end line studies.
4. Missing secondary data: Schools maintain records on teacher attendance but don’t 
track if a teacher in attendance has taught or not. The indicator below requires data on 
both attendance and teaching.
“Number of teachers in target schools who attend and teach school at least 90 percent of 
scheduled school days per school year”
The baseline study made an assumption that any teacher in attendance taught. Monitoring 
records should track both attendance and teaching to facilitate the midline and final 
evaluations.
5. Factors contributing to variations in learning outcomes across the district 
and gender: Learning outcomes differ across the districts and gender. The baseline study 
did not manage to unveil the contributing factors and barriers to learning among boys and 
Nyaruguru district. A further study can inform the program on specific strategies to 
promote literacy and tailor make the literacy activities.
52 | Page
6. Research approval in Rwanda for a household survey was not feasible in the 
timeframe of the baseline study: The process of acquiring research approval for a 
household survey from the National Institute of Statistics in Rwanda takes at least three 
months, which was not possible given the timeline of the study. This limitation confined 
the baseline study to a school-based survey, with no access to the community or parents, 
except through the PTC. Due to this, the baseline value for the following indicator could 
not be established: Percent of parents in target communities who can name at least three 
benefits of primary education. WFP will conduct a follow-up survey with parents in 
September, 2016 and the baseline value for this indicator will be published in the next 
USDA semi-annual report. 
7. The baseline value for the Minimum Acceptable Diet (MAD) was not 
established due to methodological limitations of applying MAD to school-going children 
in grades 1-6. MAD is internationally validated only for children aged 6-23 months. 
Consultations are currently ongoing between WFP and USDA on the feasibility of asking 
school-age children about their diet intakes and the reliability and accuracy of such 
information.
53 | Page
8.0 Conclusions and Recommendation
Following the baseline study, the following conclusions can be drawn:
• Findings from the sampled schools illustrate the absence of, and the need for, 
school feeding for students in grades 1-6. In addition, the baseline study 
revealed a very limited risk of overlap with other school feeding programs.
• The school management committees support the school feeding program and 
are willing to make contributions towards the program. The plan to incorporate 
inputs from the school community is well aligned with the school leaderships’ 
expectations.
• While majority of the target schools use an improved water source, reliability of 
water availability is the challenge. This program needs to incorporate a strategy 
that will ensure there is clean water available to the beneficiary schools.
• Although most of the schools have improved sanitation facilities according to 
the definition of the indicator, the number of toilets are inadequate to cater for 
the student population. There is also a need to ensure that all target schools 
have separate girls and boys toilets and reserved toilets for teachers.
• Nearly all target schools have school gardens. There is potential to utilise school 
gardens for increasing access to food in primary school.
• Based on findings from EGRA, levels of literacy in the four program districts 
across genders are low, exemplified by the number of non-readers across all 
three grades tested in the four program districts. This intervention with a 
literacy improvement component is much-needed to promote learning in 
beneficiary schools.
• The targets set by the program for some indicators don’t correspond to the 
current situation in the program schools and need to be revised. For example, 
the indicator for health related absence has been set at 20 percent. However, 
the baseline value established that only eight percent of girls and seven percent 
of boys are missing more than ten days of school per year due to illness. 
54 | Page
Annexes
Below is the list of documents annexed in this report:
1. Summary of baseline indicators against target
2. Sampling document 
3. Additional EGRA findings that include P1&2 and additional tables
4. References
5. Results framework 
55
Annex 1: Summary of baseline indicators against target
Result Indicator Baseline Indicator 
Value Target
Increased Access to 
Preventative Health 
Interventions
Number of students 
receiving deworming 
medication(s) 75, 749 127,650
Increased Access to 
Requisite Food 
Preparation and 
Storage Tools and 
Equipment
Number of target schools 
with access to improved 
food preparation and 
storage equipment 
(kitchens, cook area, 
storerooms, stoves and 
kitchen utensils
10 104
Improved School 
Infrastructure
Number of educational 
facilities (i.e. school 
buildings, classroom, and 
latrines) 
rehabilitated/constructed as 
result of USDA assistance 
(kitchens, cook areas)
Zero at baseline
Program activities have not 
started
104
Number of educational 
facilities (i.e. school 
buildings, classroom, and 
latrines) 
rehabilitated/constructed as 
result of USDA assistance 
Zero at baseline
Program activities have not 
started
26
Increased Student 
Enrolment
Number of students 
enrolled in school receiving 
USDA assistance (female) 41,404 66,378
Number of students 
enrolled in school receiving 
USDA assistance (male) 40,954 61,272
Number of students 
enrolled in school receiving 
USDA assistance 82,358 127,650
Increased Community 
Understanding of 
Benefits of Education
Percent of parents in target 
communities who can name 
at least three benefits of 
primary education
Research approval in 
Rwanda for a household 
survey was not feasible in 
the timeframe of the 
baseline. WFP decided that 
baseline data for this 
indicator will be collected 
through a separate survey 
in September. The baseline 
value for this indicator will 
be shared with USDA in 
90%
56
the next semi-annual 
report.
Increased Access to 
Clean Water and 
Sanitation Services
Number of schools using an 
improved water source
62 412
Number of schools with 
improved sanitary facilities
94 412
Increased Knowledge 
of Nutrition
Number of individuals 
trained in child health and 
nutrition as a result of 
USDA assistance (female)
Zero at baseline
Program activities have not 
started
145
Number of individuals 
trained in child health and 
nutrition as a result of 
USDA assistance (male)
Zero at baseline
Program activities have not 
started
63
Number of individuals 
trained in child health and 
nutrition as a result of 
USDA assistance 
Zero at baseline
Program activities have not 
started
208
Increased knowledge 
of Safe Food Prep and 
Storage Practices 
Percent of cooks and 
storekeepers who can 
identify at least three safe 
food preparation and 
storage practices
92% 95%
Improved Knowledge 
of Health and Hygiene 
Practices
Percent of students who can 
identify at least three key 
health and hygiene practices 
(female)
47% 80%
Percent of students who can 
identify at least three key 
health and hygiene practices 
(male)
48% 80%
Improved Policy and 
Regulatory Framework
Number of Child health and 
nutrition policies, 
regulations or 
administrative procedures 
in the following stages of 
development as result of 
USDA assistance (stage 5)
Zero at baseline
Program activities have not 
started
1
57
Number of educational 
policies, regulations, or 
administrative procedures 
in each of the following 
stages of development as a 
result of USDA assistance 
(stage 5)
Zero at baseline
Program activities have not 
started
1
Increased Capacity of 
Government 
Institutions
Number of government staff 
trained or certified as a 
result of USDA assistance 
(female)
Zero at baseline
Program activities have not 
started
58
Number of government staff 
trained or certified as a 
result of USDA assistance 
(male)
Zero at baseline
Program activities have not 
started
86
Increased Engagement 
of Local Organizations 
and Community 
Groups
Number of Parent- Teacher 
Associations (PTA) or 
similar "school" governance 
structures supported as a 
result of USDA assistance
Zero at baseline
Program activities have not 
started
104
Number of public-private 
partnerships formed as a 
result of USDA assistance 
Zero at baseline
Program activities have not 
started
31
Value of public-private 
sector investments 
leveraged as a result of 
USDA assistance (private)
Zero at baseline
Program activities have not 
started
0
Increased Government 
Support
Value of public-private 
sector investments 
leveraged as a result of 
USDA assistance (host 
government)
Zero at baseline
Program activities have not 
started
0
Percent increase in budget 
allocated by the 
Government of Rwanda to 
Home-Grown School 
Feeding Program
Zero at baseline
Currently Home Grown
School Feeding program 
has no budget allocated to 
it
5%
Increased Use of 
Health and Dietary 
Practices
Percent of school-age 
children receiving a 
minimum acceptable diet 
(female)
WFP dropped this 
indicator because of 
methodological limitations
65%
Percent of school-age 
children receiving a 
minimum acceptable diet 
(male)
WFP dropped this 
indicator because of 
methodological limitations 65%
58
Reduced Health￾Related Absences
Percent of students who 
miss more than 10 school 
days per year due to illness 
(female)
8% 20%
Percent of students who 
miss more than 10 school 
days per year due to illness 
(male)
7% 20%
Increased Economic 
and Cultural 
Incentives (Or 
Decreased 
Disincentives)
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance (female)
Zero at baseline
Program activities have not 
started
66,378
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance (male)
Zero at baseline
Program activities have not 
started
61,272
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance (new)
Zero at baseline
Program activities have not 
started
127,650
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance 
(continuing)
Zero at baseline
Program activities have not 
started
71,970
Improved Student 
Attendance
Number of students 
regularly (80%) attending 
USDA supported 
classrooms/school (female)
37,964 53,102
Number of students 
regularly (80%) attending 
USDA supported 
classrooms/school (male)
37,512 49,017
Number of students 
regularly (80%) attending 
USDA supported 
classrooms/school 
75, 522 127,650
Increased Access to 
Food (School Feeding)
Number of social assistance 
beneficiaries participating 
in productive safety nets as 
a result of USDA assistance 
(female)
Zero at baseline
Program activities have not 
started
66,378
59
Number of social assistance 
beneficiaries participating 
in productive safety nets as 
a result of USDA assistance 
(male)
Zero at baseline
Program activities have not 
started
61,272
Number of social assistance 
beneficiaries participating 
in productive safety nets as 
a result of USDA assistance 
(new)
Zero at baseline
Program activities have not 
started
127,650
Number of social assistance 
beneficiaries participating 
in productive safety nets as 
a result of USDA assistance 
(continuing)
Zero at baseline
Program activities have not 
started
71,970
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance (female)
Zero at baseline
Program activities have not 
started
66,378
Number of school-aged 
children receiving daily 
school meals (breakfast, 
snack, lunch) as a result of 
USDA assistance (male)
Zero at baseline
Program activities have not 
started
61,272
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance (new)
Zero at baseline
Program activities have not 
started
127,650
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance 
(continuing)
Zero at baseline
Program activities have not 
started
71,970
Number of school-aged 
children receiving daily 
school meals (Breakfast, 
snack, lunch) as a result of 
USDA assistance
Zero at baseline
Program activities have not 
started
127,650
Reduced Short-Term 
Hunger
Number of daily school 
meals (breakfast, snack, 
lunch) provided to school￾age children as a result of 
USDA assistance
Zero at baseline
Program activities have not 
started
74,808,333
Improved 
Attentiveness
Percentage of students in 
classroom identified as 
attentive by their teachers 60% 80%
60
Increased Skills and 
Knowledge of School 
Administrators
Number of school 
administrators and officials 
in target schools who 
demonstrate use of new 
techniques or tools as a 
result of USDA assistance
Zero at baseline
Program activities have not 
started
93
Number of school 
administrators and officials 
trained or certified as result 
of USDA assistance 
Zero at baseline
Program activities have not 
started
139
Increased Skills and 
Knowledge of Teacher
Number of 
teacher/educators/teaching 
assistants in target schools 
who demonstrate use of new 
techniques or tools as a 
result of USDA assistance
Zero at baseline
Program activities have not 
started
653
Number of 
teacher/educators/teaching 
assistants trained or 
certified as result of USDA 
assistance 
Zero at baseline
Program activities have not 
started
816
Improved Literacy of 
Instructional Materials
Number of target schools 
with supplemental reading 
materials available to 
students as result of USDA 
assistance
Zero at baseline
Program activities have not 
started
104
Better Access to School 
Supplies and Materials
Number of textbooks and 
other teaching and learning 
materials provided as a 
result of USDA assistance
Zero at baseline
Program activities have not 
started
132,990
More Consistent 
Teacher Attendance
Number of teachers in 
target schools who attend 
and teach school at least 90 
percent of scheduled school 
days per school year
986 653
Improved Quality of 
Literacy Instruction
Number of teachers in 
target schools who 
demonstrate use of new and 
quality teaching techniques 
or tools as a result of USDA 
assistance
Zero at baseline
Program activities have not 
started 
693
61
Improved Literacy of 
School-Age Children
Number of individuals 
benefiting directly from 
USDA-funded interventions 
(new)
Zero at baseline
Program activities have not 
started
133,328
Number of individuals 
benefiting directly from 
USDA-funded interventions 
(continuing)
Zero at baseline
Program activities have not 
started
77,503
Number of individuals 
benefiting directly from 
USDA-funded interventions 
(male)
Zero at baseline
Program activities have not 
started
61,240
Number of individuals 
benefiting directly from 
USDA-funded interventions 
(female)
Zero at baseline
Program activities have not 
started
69,088
Number of individuals 
benefiting directly from 
USDA-funded interventions
Zero at baseline
Program activities have not 
started
133,328
Number of individuals 
benefiting indirectly from 
USDA-funded interventions
Zero at baseline
Program activities have not 
started
175,379
Percent of students who, by 
the end of two grades of 
primary schooling, 
demonstrate that they can 
read and understand the 
meaning of grade level text 
(female)
57% 90%
Percent of students who, by 
the end of two grades of 
primary schooling, 
demonstrate that they can 
read and understand the 
meaning of grade level text 
(male)
42% 90%
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 
49% 90%
62
Annex 2: Sampling document 
Detailed Sampling Procedure for School based Survey
Sample size determination
Early grade reading assessment (EGRA) exercise was used in determining the sample size of 
schools. This activity targeted students in grade 1 to 3 although the focus of this indicator is 
grade 3. (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)
This baseline indicator is reported at total level and disaggregated by gender. Sampled schools 
for this indicator was used to collect data for other cohorts. This facilitated coordination of 
data collection activities.
Sample size determination used the following formula. This formula is used in estimating the 
sample size for small populations. It takes into consideration population size, level of 
confidence and margin of error.
n = Nz^2pq/( E^2(N-1)+ z^2pq)
Where 
 N=population size target students (83,000)
 z=confidence level (1.96)
 E=margin of error (2.8%)
 p=the proportion of a population with a particular characteristic (0.5)
 q=the proportion of a population not having this characteristic (0.5)
Based on the above formula and parameters, the sample size calculation yields a sample of 
1208. To distribute the sample equally in all schools a Figure of 1200 was used.
Using the same formula and 95% confidence level with an interval of +/-5% a sample of 382 
can be used. However, this cannot allow analysis by gender and by district. Therefore, Ipsos 
recommends a sample of 1,200 for this indicator.
Sample distribution across districts
Sample considerations
63
EGRA involved getting students from classes. Therefore, consideration in sampling was taken 
to ensure interruption of school learning activities was minimised by carrying out the study 
within one hour in a class. This time factored in sampling of students, warm up and 
assessment. School activities could also be interrupted by having many assessors in the school. 
Effectively there are 3-4 hours when this can be done. Lower grades attend school up to lunch 
time for those schools that do not have a double shift system. The teams were to complete the 
activities in one day. Based on these considerations, Ipsos proposed to have a sample of 10 
students per grade; (5 girls and 5 boys). Therefore, in each school, a sample of 30 students.
Distribution of the sample across the district was done using probability proportionate to 
population size as shown in Table 14. First the target population of each district was 
determined by summing up the population of schools in each district. The proportion of 
student population in each district to the total was calculated. Expected sample for each 
district was calculated by multiplying 1200 with the proportion of each district. To estimate 
the number of schools in each district the rationale of 30 interviews in each school was used. 
Number of schools was rounded to the nearest number. Based on this rationale 40 schools 
were sampled used for this study; 10 schools in each district.
Table 14: Determining number of schools required
(a)
District
Total number of 
students Proportion Sample
Estimate 
of 
schools
No. of 
schools
Karongi 20028 24.3% 292 9.73 10
Nyamagabe 20887 25.4% 304 10.14 10
Rutsiro 19698 23.9% 287 9.57 10
Nyaruguru 21745 26.4% 317 10.56 10
82358 100.0% 1200 40
Student population data source: Sampling frame provided by WFP
Sample size of students for health 
Data for this indicator was collected through a student survey.
• Percent of students who can identify at least three key health and hygiene practices
Data is disaggregated by gender. In addition, the first indicator required data disaggregation 
by classes. Although class 1-3 was included in collecting data for these indicators, it was 
expected that comprehension and recall may be lower. Therefore, focus of these indicators was
grade 4-6.
64
Data was collected from 2,400 students; 1200 from grade 4-6 and 1200 from grade1-3. Given 
that a sample was drawn for EGRA which is a simple tool, Ipsos proposed to ask the sampled 
students additional few questions to capture this indicator. In grade 4, 5 and 6, data was 
collected from a sample of 10 students in each grade. 
Sample size for this indicator has been determined using the same rationale as EGRA sample. 
Selection of sampled schools
Systematic random sampling was used to select the schools. A sampling frame was provided 
by WFP which comprised of all the 104 schools target beneficiary schools. These schools were 
grouped per district and given serial numbers. For each district a sampling interval was 
calculated. A random start was picked using True Random Number Generator. This was 
the first school to be picked on the list. The other 9 schools were selected by adding the 
sampling interval. 
Table 15: list of surveyed schools, by district
Karongi District Nyamagabe District
EP Kinyami GS Rugogwe 
EP Gitesi EP Kiyumba
EP Gitanga GS Rususa
GS Ngoma EP Baro
EP Kinyovu GS Kiraro Protestant
EP Biguhu A EP Kibirizi
GS Nyabikeri EP Nyagisozi 
EP Nyamugwagwa EP Nkore
GS Shoba Muramba EP Masagara
EP Nyamabuye EP Kamegeri
Rutsiro District Nyaruguru District
EP Mujebeshi EP Ruhinga
EP Mpingamabuye GS Ruheru 
GS Bitega B EP Zirambi 
EP Umubano EP Gakaranka
EP Rundoyi EP Kimina 
EP Gasovu GS Rugerero
EP Nyakarera GS Kiyonza
EP Mwufe GS Kivuru 
GS KIVUMU GS Fugi
EP Rutsiro EP Ruhororo
*GS – Groupe Scolaire: primary and secondary school
**EP – École Primaire: primary school
65
Selection of participating students
Selection of participating students was done upon arrival in the school using systematic 
random sampling. The attendance register for that day was used. In each class two sampling 
frames were developed; one for girls and the other one for boys. A sampling interval for each 
frame was determined by dividing the total number of girls/boys by 5. Then a random start 
was obtained by which any number is picked randomly, that is between 1 and the sampling 
interval. The other four students were picked by adding the sampling interval. Each selected 
student was given a unique code which was used to identify the in the data
In the event that there were multiple streams in a class one class was selected using simple 
random sampling. 
School Based surveys
Entering the districts and school
WFP assisted in contacting the district education offices and organizing for the first courtesy 
call at district education offices or Mayors office. Prior appointments were made with schools 
before the research activity day. The team leader organized the logistics to the school prior to 
the activity day. 
The team would ensure that they reached the school first thing in the morning. The team 
leader will introduce the team and the activity. Each team member would introduce their role 
that day without giving details that would bias the answers.
The team leader would go through the consenting process with the head teacher. After getting 
consent, he/she would seek to understand the school program for that day so that they can 
effectively plan the activities within the school program.
The team leader would request the head teacher to organize for a brief meeting with the 
teachers, cooks, storekeeper so that he can brief them and give them his/her blessings to 
cooperate in the survey. This was an opportunity to inform the teachers about the activity and 
the processes. It also helped to manage the misconceptions about the activities which can bias 
the responses. After introductions the team would commence sampling and data collection. 
After completing the school activities, the team leader will congregate his/her team and give a 
vote of thanks to the head teacher and the teachers.
66
Table 16: List of Key Informant Interviews Conducted
Organization Designation
Nyamagabe District DEO
Karongi District DEO
Rutsiro District DEO
Nyaruguru District DEO
Nyamagabe District HGSF District Coordinator
Nyaruguru District HGSF District Coordinator
WFP National Program officer
WFP Head of Program Unit in WFP
WFP Project Manager of HGSF
WV Technical Program Manager
WV Technical Program Officer
MINAGRI
Works at Rwanda Agricultural Board, in charge of One 
Cup of Milk per Child
MINEDUC
67
Annex 3: Additional EGRA findings (that include P1& P2) and tables
The following facts emerge from the analysis of the literacy scores for all the students including 
P1 and P2.
1. The learning levels in literacy in the four districts, across the classes and 
cutting across the two genders are low;
The average learning levels across the four districts in all the classes and cutting across the two 
genders are low. On average, the learners could only identify 11 sounds correctly per minute 
with a standard deviation of nine, meaning that there is a high proportion of children who 
could not identify any letter sound at all. On the other hand, learners decoded an average of 
25 syllables per minute, 11 familiar words per minute, 7 invented words per minute and 10 
words in a text. This is against a list of 100 sounds, 100 syllables, 50 familiar and invented 
words and 65 words in a text given. It is evident that children are struggling to read. There are 
also wide variations among the learners as demonstrated by high standard deviations in the 
scores as shown in the Figure 11 below.
Figure 11: Average number of sounds, syllables and words read per minute
Base: 1200 students 
2. Students in P3 have higher learning outcomes in literacy than their peers in 
P2 and P1;
It is evident that the learning levels are comparatively higher in P3 and lowest in P1. This is 
perhaps expected considering that the P3 have more content exposure than their peers in P2 
and P1 respectively. However, even the P2 for whom the test has been developed based on 
their level, it is evident that their learning levels are low. For instance, on decoding the sounds, 
11
25
11
7
10
Reading Sounds Reading Syllables Familiar Words Invented words Words in a text
68
P3 on average 16 sounds in a minute compared to their peers in P1 who on average, decoded 
6 sounds in a minute. When it comes to reading words in a text, the average number of words 
read in a text in P3 was 20wpm (words per minute) compared to 1 wpm in P1. The higher 
performance is also replicated in other literacy tasks that include reading the syllables, familiar 
words and invented words. Figure 12 below breaks down the average performance of the 
classes by task. 
Figure 12: Average Reading Scores by Class
Base: 400 students in each class
3. Learning outcomes are highest in Nyamagabe and lowest in Nyaruguru 
district;
When the learning levels in the four districts are compared, learning levels are highest in 
Nyamagabe and lowest in Nyaruguru while Rutsiru and Karongi fall in between. This occurs 
across all the literacy task and thus, this cannot be happening by chance. Nyamagabe has the 
highest literacy levels, followed by Karongi, Rutsiro and Nyaruguru has the lowest. However, 
the differences between the highest and the lowest vary from one task to the other. For 
instance, the average syllables read in Nyamagabe was 34 syllables per minute compared to 
20 syllables correctly read in Nyaruguru. This is illustrated in Figure 13.
6 6
1 1 1
11
25
9
6
8
16
46
21
15
20
Reading Sounds Reading Syllables Familiar Words Invented words Words in a text
P1 P2 P3
69
Figure 13: Average Reading Scores by District
*Total refers to average
4. Girls demonstrate higher learning outcomes in literacy than the boys in the 
four districts; 
When boys and girls are compared, girls fair better than boys on all the literacy tasks. On 
average, girls identified 12 sounds compared to boys who identified 10 sounds, read 29 
syllables compared to boys who read 22 syllables, read 12 familiar words than 9 read by boys, 
read 8 invented words compared to 6 read by the boys and read fluently (23 wpm) in a text 
compared to boys (8wpm). This is illustrated in Figure 14 below.
10
7
11
25
11
8
6
9
21
10
8
6
9
20
9
13
9
14
34
14
11
9
12
26
11
0
5
10
15
20
25
30
35
40
Words in a text Invented Words Familiar Words Syllables Sounds
Average Scores per task by District
Total Rutsiro Nyaruguru Nyamagabe Karongi
70
Figure 14: Average Reading Scores by Gender
Base: Girls (600); Boys (600)
5. Learners in the target classes have a problem decoding the letter sounds in 
Kinyarwanda language
Although the foundation of literacy is the grasping of sounds, this seems not to be the case 
with the target learners. It is evident that learners are struggling to read the letter sounds. In 
fact, the scores in syllable reading are higher than the scores in letter sound reading. This could 
be perhaps due to the nature of reading the sounds that is often accompanied by decoding by 
adding a vowel and therefore consonants rarely being read independently. For instance, the 
results in P1 are level at 6 sounds/syllables per minute but this takes a stead difference in P2 
and P3 where the difference between the two tasks is 14 sounds/syllables and 30 
sounds/syllables respectively. This is illustrated in Figure 15 below.
Figure 15: Comparison of reading Sounds and Syllables by grade
Base: 400 per class
6. There are high proportions of non-readers across the three classes in all the 
four target districts. 
It is evident that a high proportion of the children attending the target classes completely lack 
reading skills. When presented with a broad range of reading tasks that are timed, the P1 
10
22
9
6
8
12
29
12
8
12
Sounds Syllables Familiar words Invented words Words in Text
Boys Girls
6
11
16
6
25
46
P1 P2 P3
Sounds Syllables
71
learners have the highest proportion of non-readers. However, it is also observable that the P3 
who have covered the entire test content record high proportions of the non-readers. For
instance, almost a quarter (24 percent) of all the learners in P3 cannot read any word in a text 
while a third (29 percent) of those in the same class cannot decode an invented word. While 
focusing on the same group of learners (P3), 27 percent of them cannot read a familiar word. 
This brings to fore the reality that many children are struggling to learn the basic skills and 
are graduating to higher levels without the acquiring the basic reading skills. 
Table 17: Average reading scores against the number of tasks
Analysis Type Sounds Syllables Familiar 
words
Invented 
words
Words in 
Text
Mean 11 25 11 7 10
Number of tasks 100 100 50 50 65
Table 18: Average Reading scores by class
Grade Sounds Syllables Familiar 
words
Invented 
words
Words in 
Text
P1 6 6 1 1 1
P2 11 25 9 6 8
P3 16 46 21 15 20
Total 11 25 11 7 10
Table 19: Average sounds read by district and disaggregated by class
District P1 P2 P3 Total
Karongi 5 10 16 11
Nyamagabe 7 14 22 14
Nyaruguru 6 9 13 9
Rutsiro 4 10 15 10
Total 6 11 16 11
Table 20: Average words in a text correctly read per minute by district and disaggregated by class
District P1 P2 P3 Total
Karongi 1 8 24 11
Nyamagabe 1 12 26 13
Nyaruguru 1 6 13 7
Rutsiro 1 8 16 8
Total 1 8 20 10
72
Table 21: Average invented words correctly read per minute by district and disaggregated by class
District P1 P2 P3 Total
Karongi 0 2 17 10
Nyamagabe 1 8 19 12
Nyaruguru 1 6 10 11
Rutsiro 0 6 12 10
Total 1 6 15 7
Table 22: Average familiar words correctly read per minute by district and disaggregated by class
District P1 P2 P3 Total
Karongi 1 9 25 12
Nyamagabe 2 12 27 14
Nyaruguru 2 7 14 8
Rutsiro 1 9 18 9
Total 1 9 21 11
Table 23: Average syllables read by district and disaggregated by class
District P1 P2 P3 Total
Karongi 7 23 50 26
Nyamagabe 8 32 60 34
Nyaruguru 6 20 35 20
Rutsiro 4 10 15 21
Total 6 25 46 25
Table 24: Overall reading levels by District 
So
un
ds
Syllable
s
Familiar 
words
Invented 
words
Words in 
Text
Karongi 11 26 12 8 11
Nyamagabe 14 34 14 9 13
Nyaruguru 9 20 8 6 7
Rutsiro 10 21 9 6 8
Total 11 25 11 7 10
Table 25: Average sounds read by district and disaggregated by class
District Boy Girl Total
Karongi 10 11 11
Nyamagabe 13 15 14
Nyaruguru 8 10 9
Rutsiro 9 11 10
Total 10 12 11
73
Table 26: Average syllables read by district and disaggregated by class
District Boy Girl Total
Karongi 24 29 26
Nyamagabe 29 38 34
Nyaruguru 17 24 20
Rutsiro 18 25 21
Total 22 29 25
Table 27: Average familiar words correctly read per minute by district and disaggregated by gender
District Boy Girl Total
Karongi 10 13 12
Nyamagabe 12 15 14
Nyaruguru 6 9 8
Rutsiro 7 11 9
Total 9 12 11
Table 28: Average invented words correctly read per minute by district and disaggregated by gender
District Boy Girl Total
Karongi 7 9 10
Nyamagabe 8 11 12
Nyaruguru 4 7 11
Rutsiro 5 8 10
Total 6 8 7
Table 29: Average words in a text correctly read per minute by district and disaggregated by gender
District Boy Girl Total
Karongi 10 12 11
Nyamagabe 11 15 13
Nyaruguru 5 9 7
Rutsiro 6 10 8
Total 0 1 1
Table 30: Average Reading Scores by gender 
Gender Sounds Syllables Familiar 
words
Invented 
words
Words in 
Text
Boys 10 22 9 6 8
Girls 12 29 12 8 12
Total 11 25 11 7 10
74
Annex 4: References
Food Assistance Division, Office of Capacity Building and Development United States 
Department of Agriculture, Foreign Agricultural Service (2013). Monitoring and Evaluation 
Policy 
FAS, WFP(2015). Framework of Agreement between The Foreign Agriculture Services and 
the World Food Programme on the Donation of Agricutural c odities and the Reated 
Assistance Under the MacGovern -Dole International Food for Education and Child 
Nutrition. 
Republic of Rwanda, Ministry of Education (2014). National School Health Policy. 
Kigali
Republic of Rwanda, Ministry of Education (2014). National School Health Strategic 
Plan 2013/2014-2017/2018. Kigali.
Republic of Rwanda, Ministry of Education (June 2016). 2015 Educational Statistical 
Yearbook. Kigali
Republic of Rwanda, (2013). Homegrown School Feeding Programme White Paper; 
“Developing local agriculture, nourishing young minds. Kigali
Republic of Rwanda, Ministry of Local Government (January 2013). District Development 
Plan 2013/14-2017/18. Nyaruguru
USAID, (March 2015). Overview of the Early Grade Reading Assessment (EGRA); Designing 
and Implementing Early Grade Reading Assessment 
World Food Program (2016). List of schools and summary information in the 4 district 
selected to participate in the Home School program by WFP.
World Food Program. Rwanda FY2015 McGovern‐Dole Propo
75
Annex 5: Result Framework
76
77