2009年-世界发展银行全球_Educational_and_Health_Impacts_of_Two_School_Feeding_Schemes___Evidence_from_a_Randomized_Trial_in_Rural_Burkina_Faso_44页_885kb

Educational and Health Impacts of Two School Feeding Schemes: Evidence from a Randomized Trial in Rural Burkina Faso

Core Content

This paper evaluates the impact of two school feeding schemes—school meals and take-home rations (THR)—on educational and health outcomes for children in low-income households in rural Burkina Faso. The study uses a randomized controlled trial (RCT) design to isolate the effects of the interventions.

Main Objectives

The study aims to:

• Assess the impact of school feeding programs on enrollment, academic performance, cognitive development, and nutrition.
• Compare the effectiveness of school meals and THR in different contexts.
• Explore spillover effects on younger siblings of program beneficiaries.

Key Findings

1. Enrollment

• Both school meals and THR increased new enrollment for girls by 5 to 6 percentage points.
• School meals increased enrollment for girls by 4.9 percentage points.
• THR increased enrollment for girls by 5.6 percentage points.
• No significant impact on boys' enrollment was observed.

2. Academic Performance

• School meals led to a slight improvement in time-adjusted math scores for girls.
• No significant impact on raw math scores was observed.
• No significant effect on other cognitive development measures (e.g., memory, reasoning).

3. Absenteeism

• School meals increased absenteeism in households with low child labor supply.
• THR decreased absenteeism in households with high child labor supply.
• This suggests that labor constraints influence the relationship between school feeding and school attendance.

4. Nutritional Impact

• THR significantly improved nutrition for younger children (aged 12–60 months) not in school:
  • Weight-for-height increased by 0.33 standard deviations.
  • Weight-for-age increased by 0.38 standard deviations.
• School meals had no significant impact on the nutrition of younger children.

Program Design and Implementation

• The study was conducted in northern Burkina Faso, where primary school participation is very low (around 20% of school-age children).
• School meals provided lunch daily, accessible to both boys and girls.
• THR provided 10 kg of cereal flour monthly to girls, conditional on 90% attendance.
• The World Food Program (WFP) managed the implementation of both programs.
• Randomized assignment of 46 schools to three groups: school meals, THR, and control.
• Baseline survey was conducted in June 2006, with follow-up in June 2007.

Methodology

• The study uses a difference-in-differences (DID) model to estimate the intent-to-treat (ITT) effect.
• Random assignment of villages ensures causal interpretation of the results.
• Anthropometric data and mathematical tests were collected from children aged 6–15.
• Cognitive tests (e.g., Raven's Colored Progressive Matrices) were administered in the follow-up round.

Policy Implications

• THR may be more effective in reaching malnourished children not enrolled in school, due to household-level food sharing.
• School meals are more targeted but may lead to intra-household reallocation of food, potentially diluting the nutritional benefits.
• The effectiveness of school feeding programs depends on local conditions and household labor dynamics.
• The study highlights the importance of considering spillover effects when evaluating school feeding interventions.

Limitations and Considerations

• The random assignment of villages to treatment and control groups may be affected by crossover effects.
• The one-year duration of the program might limit the ability to observe long-term impacts.
• The results on older boys are mixed and require further investigation.

Conclusion

The study provides new insights into the effectiveness of school feeding programs in improving enrollment and nutrition, particularly for girls and younger siblings. It underscores the need for policy design that accounts for household behavior, labor constraints, and spillover effects. The results suggest that THR may be more beneficial for nutrition, while school meals may be more effective for academic performance if the learning environment is improved.