Millennium Challenge Corporation hired Mathematica Policy Research to conduct an independent evaluation of the BRIGHT II program. The three main research questions of interest are:
• What was the impact of the program on school enrollment, attendance, and retention?
• What was the impact of the program on test scores?
• Are the impacts different for girls than for boys?
Mathematica will compare data collected from the 132 communities served by BRIGHT II (the "treatment group") with that collected from the 161 communities that applied but were not selected for the program (the "comparison group"). Using a statistical technique called regression discontinuity, Mathematica will compare the outcomes of the treatment villages just above the cutoff point to the outcomes of the comparison villages just below the cutoff point. If the intervention had an impact, we will observe a "jump" in outcomes at the point of discontinuity.
Mathematica will perform additional analyses to estimate the overall merit of the BRIGHT investment. By conducting a cost-benefit analysis and a cost-effectiveness analysis and calculating the economic rate of return, Mathematica will be able to answer questions related to the sustainability of the program, and compare the program to interventions and social investments in other sectors. The household survey is designed to capture household-level data rather than community-level data; however, questions have been included to measure head-of-household expectations of educational attainment. These questions ask the head of household what grade level he hopes each child will attain; and what grade level he thinks the child will be capable of achieving in reality.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Anonymized dataset for public distribution
132 rural villages throughout the 10 provinces of Burkina Faso in which girls' enrollment rates were lowest
Unit of Analysis
Households, students, and educators in the 287 villages surveyed
Producers and sponsors
Authoring entity/Primary investigators
Mathematica Policy Research
Millennium Challenge Corporation
The BRIGHT II program was implemented in the same 132 villages that received the BRIGHT I interventions. These 132 villages were originally selected using a scoring process, with eligibility scores based on the villages’ potential to improve girls’ educational outcomes. A total of 293 villages applied to receive a BRIGHT school; the Burkina Faso Ministry of Basic Education (MEBA) selected the 132 villages with scores that were above a certain cutoff point. Whenever possible, the survey will be conducted with the same children in the same households and schools surveyed during the BRIGHT I evaluation. By visiting the same households and schools, the evaluator will be able to better assess the longer-term impacts of the BRIGHT project.
Dates of Data Collection (YYYY/MM/DD)
Type of Research Instrument
Mathematica has developed two surveys, a household survey and a school survey, to collect relevant data from villages in both the treatment and comparison groups. The household survey was administered to a new cross-section of households compared to the BRIGHT I evaluation. Data will be collected on the attendance and educational attainment of school-age children in the household, attitudes towards girls' education, and parental assessment of the extent to which the complementary interventions influenced school enrollment decisions. It will also assess the performance of all household children on basic tests of French and math. The school survey, to be administered to all local schools in the 293 villages, gathers data on school characteristics, personnel, and physical structure, and collects enrollment and attendance records. Data will be gathered by a local data collection firm selected by MCA-Burkina Faso, with Mathematica providing technical assistance and oversight.
Bureau d'Etudes et de Recherche pour le Développement
Following data collection, Mathematica will work with BERD to ensure that the data are correctly entered and are complete and clean. This will include a review of all frequencies for out-of-range responses, missing data, or other problems, as well as a comparison between the data and paper copies for a random selection of variables.
Levy, Dan, Matt Sloan. 2012. Impact Evaluation of Burkina Faso's BRIGHT II Program. Washington, D.C.; Mathematica Policy Research, Inc.
Monitoring & Evaluation Division of the Millennium Challenge Corporation
Millennium Challenge Corporation
Review of Metadata
Date of Production
Version 1.0 (April 2014)
Version 2.0 (April 2015). Edited version based on Version 01 (DDI-MCC-BFA-IE-EDU-2012-v1) that was done by Millennium Challenge Corporation.
Version 02 (March 2019). This version is identical to version 01, except for the section on Related Material which was updated.