[N(poster)] ポスターセッション

英語

[1O14] Heterogeneous Effects of Pre-primary Education on Child Development Outcomes in Bangladesh: Insights from Project Data and Application of Causal Forest Analysis

*Kohei UNO¹ (1. Kobe University/ PADECO)

キーワード：

Pre-primary Education、Bangladesh

Pre-primary education (PPE) is widely and globally recognized as a critical foundation for lifelong learning and development. In line with global trends, the Government of Bangladesh has made significant efforts to implement PPE policy reforms in collaboration with multilateral agencies over a decade. Given the rapid expansion of PPE access without sufficient supporting evidence or evaluation, it is timely to examine the potential effects of PPE on child outcomes. A substantial body of evidence from intervention studies in high-income countries, such as the Perry Preschool Project and the Azerbaijan PPE program, demonstrates that PPE is a cost-effective public investment, yielding particularly large benefits for children from disadvantaged backgrounds. In contrast, evidence from low- and middle-income countries (LMICs) remains limited. For example, Pimenta (2023) documented heterogeneity in the association between PPE attendance and child outcomes across family wealth, parental education, and rural–urban divides. Yet, to date, no causal inference studies have been conducted in LMICs, leaving a significant research gap. This study addresses this gap by leveraging data from a randomized controlled trial (RCT) of the Early Years Preschool Program (EYPP), conducted by the World Bank in Meherpur district, Bangladesh. Using data from approximately 1,800 children, this study applies recent machine-learning methods, called the causal forest to estimate Conditional Average Treatment Effects (CATEs) across child and family characteristics. The findings reveal that across all developmental domains (literacy, numeracy, social-emotional, and motor skills), children from poorer households and those with fewer home learning opportunities benefited more from the EYPP compared to their more advantaged peers. CATE estimation further indicates that children with mid- to low-level baseline cognitive skills (−1 < SD < 0) experienced the largest gains, whereas those with extremely low baseline skills (SD < −1) showed smaller improvements. These results suggest that a quality-assured preschool program can mitigate early developmental disparities in LMIC contexts, echoing findings from high-income countries. This study contributes new causal evidence on the heterogeneous impacts of PPE in Bangladesh, with broader implications for designing equitable early childhood policies in developing countries.