What Works in Developing Nations to Get Children Into School and Keep Them There? A Systematic Review of Experimental and Quasi-Experimental Research

What are the effects of interventions implemented in developing countries on measures of enrollment, attendance, retention, and progression?

What are the effects of interventions in developing countries on learning outcomes as measured by test scores, grades, and other achievement measures?

Funded by the International Initiative for Impact Evaluation (3ie), WestEd sought to answer these key research questions by systematically reviewing rigorous studies of programs and interventions designed to increase school enrollment in developing nations (as classified by the World Bank).

To be included in WestEd’s systematic review, studies had to use randomized controlled designs or quasi-experimental designs that controlled for baseline differences on the major outcome measure.

The research methods WestEd used to conduct the synthesis were peer reviewed and published by the Campbell Collaboration to ensure transparency in the review process.

WestEd researchers searched approximately 30 electronic databases along with the use of other methods to identify studies. They used a 50-item coding instrument to extract data from each eligible study. Following two rounds of screening, researchers included 73 studies in the final sample.

When possible, researchers created effect sizes for four primary outcomes: enrollment, attendance, retention, and grade progression. Researchers used standardized mean effects to transform data contained in the retrieved evaluation reports.

Study Findings

Results from the 73 included studies suggest that the average effect of intervention was small but positive in keeping children in school in developing nations. Studies of the effects of building new schools and other infrastructure interventions reported the largest average positive impacts.

Effects on enrollment, attendance, and progression were larger than those reported across studies that measured dropout. For secondary learning outcomes, average effects on math and language scores were larger than those reported across studies that measured global test scores or “other achievement.”

A journal article based on this research was one of the top five downloaded social work articles via Sage Publication’s Social Work twitter page in 2014. The methods of the review were also highlighted in the online Sage Research Methods Cases (published in 2015).