Case Study: Brightening the Future of Girls’ Education in Burkina Faso
The BRIGHT program was implemented in the 10 provinces in which girls’ enrollment rates were the lowest in Burkina Faso from 2005-2008.
Mathematica and its partners conducted a series of rigorous impact evaluations to examine the impacts of the BRIGHT program 3, 5, and 10 years after implementation started in 2005.
The program had large positive impacts on the school enrollment rates and test scores of both girls and boys after 3 years and continued to have positive impacts after 7 and 10 years. However, the magnitude of the impacts declined with time.
Notes on Data Collection. Mathematica led the data collection efforts for the BRIGHT evaluation. Households in the 293 villages that applied for a BRIGHT school made up the sample frame. In total, 132 “participant” villages received a BRIGHT school, and 161 “comparison” villages did not receive a BRIGHT school. We collected household-level data from 30 randomly selected households in each village and school-level data from three schools within 10 kilometers of the village center that village children were enrolled in. The interviewers used paper surveys and were fluent in French as well as local languages; they were organized into teams by linguistic group.
Fewer than half of the girls living in Burkina Faso in 2004 attended primary school. In the same year, only a quarter were enrolled in the highest grade of primary school. To address these issues, the government of Burkina Faso launched “Burkinabé Response to Improve Girls’ Chances to Succeed,” or BRIGHT, in 132 villages throughout 10 provinces in the country, where primary school enrollment rates were the lowest nationwide. Starting in 2005, BRIGHT funded the construction of primary schools, one in each village, and implemented a set of complementary interventions. The schools had three classrooms, one each for grades 1 to 3. To sustain the program’s success beyond its first three years, the Burkina Faso government extended BRIGHT in 2009 by constructing three more classrooms, one each for grades 4 to 6, and by continuing the original complementary interventions.
The BRIGHT schools were based on a model that consists of three classrooms, housing for three teachers, and separate latrines for boys and girls. The schools' locations in each selected village were deliberately chosen because they were near a water source, and a borehole was installed close by. Complementary interventions included daily meals for students, take-home rations to girls with a 90 percent attendance rate, school textbooks, a mobilization campaign, literacy training and mentoring, and local capacity building.
Mathematica researchers, along with two academic consultants (Harounan Kazianga and Leigh Linden) completed a series of rigorous evaluations of the BRIGHT program after 3, 5, and 10 years of implementation. We used a regression discontinuity design to assess how children in BRIGHT villages fared relative to how they would have fared had the BRIGHT program not been implemented. Data for the three evaluations were collected by University of Ouagadougou, Bureau d’Etude et de Researche pour le Développement (BERD), and the Laboratoire d’Analyse Quantitative Appliquée au Développement-Sahel (LAQAD-S), respectively. For each evaluation, the data collection teams visited the 132 participant and 161 comparison villages across the country, completing thousands of household questionnaires and hundreds of school questionnaires.
First three years of operation. BRIGHT increased enrollment by 20 percentage points, based on self-reports in the household survey data collected in 2008 for children ages 6 to 12. The impact on enrollment was accompanied by a large positive impact of 0.4 standard deviation on student test scores, which covered math and French. The short-term impacts of BRIGHT were positive for both boys and girls. In terms of enrollment, the impact for girls was about 5 percentage points higher than the impact for boys. However, the impacts on test scores for girls and boys were statistically indistinguishable.
After seven years of operation. BRIGHT continued to have large positive impacts on self-reported enrollment, of 15.5 percentage points, and on test scores, of 0.29 standard deviations. However, BRIGHT did not have any impact on child health in terms of arm circumference and four other anthropometric measures (height-for-age z-index, weight-for-age z-index, weight-for-height z-index, and body mass index). The program modestly reduced the number of children engaged in six common household activities by 2.1 to 5.2 percentage points. The evaluation also found that BRIGHT had larger positive impacts on girls compared to boys in terms of enrollment and test scores. Girls’ enrollment increased by 11.4 percentage points more than boys’ did, and their test scores increased by 0.21 standard deviations more. There was no differential impact for girls in terms of health outcomes, but the program succeeded in a greater reduction in the child labor index for girls, by 0.07 standard deviations, than boys. The seven-year impacts were based on surveys conducted in 2012 that collected data for children ages 6 to 17.
After 10 years of operation. Based on data collected in 2015 from the same set of 293 villages as in the two previous evaluations, BRIGHT raised enrollment rates and test scores for both girls and boys, with larger impacts on girls. Girls’ enrollment increased 5.4 percentage points more than boys’ did, and girls’ test scores increased by .08 standard deviations more. Also, BRIGHT boosted the rates of primary school completion and enrollment in school while decreasing employment and marriage rates for young women. For girls and women ages 13 to 22, rates of primary school completion increased by 13.5 percentage points, enrollment increased by 10.3 percentage points, and youth employment and marriage rates declined by 5.5 and 6.3 percentage points, respectively. This round collected data for children and young adults 6 to 22 years old.
This case study is for informational purposes only. Mathematica Policy Research, a nonpartisan research firm, provides a full range of research and data collection services, including program evaluation and policy research, survey design and data collection, research assessment and interpretation, and program performance/data management, to improve public well-being. Its clients include federal and state governments, foundations, and private-sector and international organizations. The employee-owned company, with offices in Princeton, N.J.; Ann Arbor, Mich.; Cambridge, Mass.; Chicago, Ill.; Oakland, Calif.; and Washington, D.C., has conducted some of the most important studies of education, disability, health care, family support, employment, nutrition, and early childhood policies and programs.
The Millennium Challenge Corporation (MCC) sponsored the “Burkinabé Response to Improve Girls’ Chances to Succeed,” or BRIGHT program to improve the educational outcomes of children in Burkina Faso. Its primary focus was girls, and it was implemented in 132 villages. The first phase of the program (BRIGHT I) operated from 2005 to 2008 under the Burkina Faso Threshold Program and consisted of constructing primary schools with three classrooms and implementing a set of complementary interventions. To continue the success of BRIGHT I, the government of Burkina Faso extended it, using $28.8 million in compact funding. This second phase of BRIGHT was implemented from 2009 to September 2012 and consisted of constructing three additional classrooms for grades 4 through 6 in the original 132 villages and continuing the complementary interventions begun during the first three years of the program. BRIGHT was implemented by a consortium of NGOs–Plan International, Catholic Relief Services, Tin Tua, and the Forum for African Women Educationalists–under the supervision of USAID.
Mathematica Policy Research and its partners conducted a series of rigorous impact evaluations of the BRIGHT program to assess the impact of the program after 3, 7, and 10 years of implementation. All three evaluations examine impacts on school enrollment, test scores, and child labor. Additionally, the 7-year evaluation examines impacts on child health and the 10-year evaluation examines impacts on young adult outcomes.