This post summarizes findings from a study in education presented on March 2, 2012 at a half-day event hosted by Innovations for Poverty Action (IPA) and the University of California’s Center for Evaluation for Global Action (CEGA). Please forgive any errors or omissions. Cross-posted from Philanthropy Action.

Read about the other presentations from this conference in Part One and Part Two.

Michael Kremer from Harvard University presented in the second half of the morning on the results of a study he conducted to assess whether empowerment and political engagement are higher for girls who stay in school longer.

The results were part of a ten-year follow-up on merit scholar recipients from a study Kremer conducted in 2001. In the original study, Kremer and his colleagues Edward Miguel and Rebecca Thornton found that offering merit scholarships to the top performing girls in poor rural schools increased school attendance and performance across the board—even among those girls and boys who were well outside the range of achievement (to say nothing of gender) to receive a scholarship.  In sum, scholarships help girls do better in school and they go to school longer. This is particularly important for this population, which is extremely poor, socially marginalized, and lives in a part of Kenya that is not politically empowered. But does the extra schooling also make the young women more empowered and more actively engaged in their communities and society, as many have suggested?

The evidence in favor is muted to nonexistent. Ten years later, the girls in treatment schools did have more progressive views about their rights. They were less likely to agree that it was okay for a man to beat his wife, and less likely to have had their parents choose their spouse. They were much more likely to read a newspaper regularly and their political knowledge was much higher than that of their peers. Yet that knowledge did not necessarily translate into action or engagement. These more educated women did not have different feelings about democracy or their ability to affect change within a democratic system.

These results fail to provide strong support for the theory that education empowers marginalized groups and replaces assigned identities (caste, tribe, gender) with a self-empowered identity based on personal achievement. But in many ways this study is more interesting for its long-term outlook. Kremer is one of the pioneering researchers applying randomized techniques to the social sciences. Just last year he and Edward Miguel published another paper offering a ten-year look at young adults who had been the first cohort to receive deworming pills in Kenya’s schools. To be able to look back ten years later to assess the long-term impact of programs is itself a huge achievement.