When the goal is to improve literacy in developing countries, common sense would say you should distribute more textbooks to schools. But common sense is often wrong, according to the Poverty Action Lab, a research group that conducts rigorous evaluations of which aid programs work and which ones do not.
Established in 2003 by three professors at the Massachusetts Institute of Technology, the Poverty Action Lab uses randomized trials and has reached some counterintuitive findings.
Providing more textbooks to children in rural Kenya did not significantly improve literacy, for example, but giving them drugs to treat intestinal worms worked wonders.
The reason was twofold. First, health problems caused by intestinal worms were so pervasive in Kenya that they were keeping even many of the most motivated students out of the schools that supplied the textbooks. Second, while the textbooks were written in the appropriate language, their curriculums were designed with Kenya’s urban children in mind, and simply did not speak to the rural children.
The conventional wisdom also says that if you want kids to improve their performance at school, you need to provide more educational materials and better schools. However, researchers found that incentives for student learning, largely ignored in previous educational studies, boosted student effort among girls in Kenya. Merit-based scholarships to girls led to significant gains in standardized exam scores and school attendance. Interestingly enough, boys who were ineligible for the scholarship but who attended the same schools also achieved higher test scores, presumably the result of a positive classroom environment. The study’s researchers believe that the performance incentives both created an environment more suitable to learning and spurred parents to take more of an interest in their children’s education.
A study measuring a school voucher program to the urban poor in Colombia found a link between private education and likelihood of early marriage. Recipients of vouchers were found to be less likely to marry as teenagers than their counterparts. This outcome allowed the students more educational opportunities that could increase their future earnings while stopping poverty’s vicious cycle.
“The takeaway is that you can’t just rely on your instincts or conventional wisdom when considering what is going to work,” explains Rachel Glennerster, executive director of the Poverty Action Lab.
The lab believes that its methodology of randomized trials is the key to generating reliable and accurate answers to a wide variety of questions on social programs. Speaking to the Lancet last August, Esther Duflo, a professor of economics at MIT and one of the lab’s founders, said, “Creating a culture in which rigorous randomized evaluations are promoted, encouraged, and financed has the potential to revolutionize social policy during the 21st century, just as randomized trials revolutionized medicine during the 20th.”
Read more stories by Andrea Orr.