Individual Submission Summary
Share...

Direct link:

What we learn about girls’ education from interventions that don’t focus on girls

Mon, April 15, 1:30 to 3:00pm, Hyatt Regency, Floor: Atrium (Level 2), Waterfront E

Proposal

Despite dramatic global gains in access to education, 130 million school-age girls remain out of school. Among those who do enter, too many fail to gain the essential skills to succeed after they complete their schooling. Many previous efforts to synthesize evidence on how to improve both access and learning for girls have focused on interventions that are principally targeted to girls, such as girls’ latrines or scholarships for girls. However, if non-targeted interventions – those that benefit both girls and boys – significantly improve girls’ education, then such a strategy may miss some of the best investments for improving educational opportunities for girls. In this updated review, we ask three research questions:

1. Are girl-targeted interventions the most effective for improving access and learning?
2. For non-targeted interventions, do impacts on girls tend to be larger?
3. What specific interventions tend to be the most effective for improving girls’ access and learning?

Using a recent education database, we identified 105 general education studies that reported program effects for girls and 24 girl-targeted studies. With the support from authors of studies that originally did not report gender-differentiated impacts, we were able to obtain additional results of 51 studies. In total, we synthesized the effects for girls of 180 studies.
We find that to increase girls’ access to school, most girl-targeted interventions have similar effects to non-targeted interventions, but that the top-performing girl-targeted interventions are more effective than the top-performing non-targeted interventions in improving access for girls. On average (at the 50th percentile), the effect sizes of these two categories are very similar. The middle interventions increased girls’ enrollment or attendance by 0.05-0.07 standard deviations. The effect sizes of less effective interventions at the 10th and 25th percentiles are also close to each other. However, the top girl-targeted interventions at the 90th percentiles have larger effect sizes than those of general interventions. To increase girls’ learning, targeted and non-targeted interventions perform similarly. Non-targeted interventions tend to have similar impacts in terms of access and learning for boys and for girls, so while girls do not benefit differentially, they do benefit as much as their male classmates.

Finally, we explore the most effective interventions for improving access and learning among girls. Six of the top ten programs for improving access include some form of conditional cash transfer. For learning, the top interventions are more varied, including programs to provide structured pedagogy for teachers in the early grades.
These results show that if the objective is to improve girls’ learning in school, the best strategy may be to improve the quality of schooling for all children.

Authors