May 29, 2012

What does the evidence show about improving school participation and performance in Sub-Saharan Africa? This question was the focus of the first morning of the two-day evidenced-based education conference held in Accra, Ghana last week. Although there have been large investments in promoting primary and secondary school enrollment in the last 30 years, many enrolled children still do not attend school regularly, and learn little when they do attend classes. The day’s discussion focused on where investments have the most impact in improving school participation and performance, and what implications the evidence has for future investments in education. We take a look at some of the key themes from day 1 below.

A significant body of evidence already exists

The conference showed that there is already quite a bit of high-quality evidence in education providing policy makers a robust pool of information to pull from. This information allows practitioners and policy makers to tackle ineffective systems and programs with high-quality evidence on of the best ways to improve enrollment rates and teacher quality, and increase student achievement.

The body of high-quality research has grown, and the means by which program are designed are becoming increasingly creative. Take, for example, a randomized evaluation in rural Udaipur, India that measured the effect of pay incentives on teacher attendance, and of increased teacher attendance on students’ attendance and ability in math and language. Seva Mandir, a local NGO, provided students with cameras and instructions to take a picture of the teacher and the class at the start and close of each school day. The camera’s timestamp feature allowed the implementing NGO to determine when and for how long the teacher was at school. At the end of the month, the photos were tallied and the teacher’s salary was determined partly by his or her attendance rate. In participating schools, teacher absenteeism was cut in half, translating into 30 percent more instruction time for students. Moreover, a year into the program, test scores in those schools were higher than in the schools in the comparison group.

Value for money matters

A common theme throughout many of the presentations was that the best way to spend a dollar matters. A great example of value for money is the Primary School Deworming Project in Kenya. This program provided deworming drugs to children, reducing the prevalence of soil-transmitted intestinal worms and schistosomiasis, and provided preventive health education to 75 primary schools. At a cost of less than 50 cents per child per year, school-based deworming can reduce the incidence of infection by 25 percentage points and reduced school absenteeism by 25 percent.

There is scope for rethinking resource allocation

Effective innovations are not necessarily tied to large amounts of money. There is a lot of latitude for rethinking resource allocation, and  concentrating resources on effective approaches can have a big impact. A great example of this is the Balsakhi Program, a remedial instruction program from India. In a program run by Indian NGO Pratham, a tutor (balsakhi), usually a young woman recruited from the local community and paid a fraction of the cost of civil-service teachers ($10-15 per month), worked with children in grades 2, 3, and 4 who were identified as falling behind their peers. Researchers evaluated the outcomes from the Balsakhi Program in 122 public primary schools in Vadodara and 77 schools in Mumbai. The program had substantial positive impacts on children’s academic achievement: the number of students in the bottom third of program classes who passed basic competency tests increased by nearly 8 percent, while those in the top third who passed increased by 4 percent.

Click here to view the presentations from the Evidence-Based Education conference, which was co-hosted by J-PAL and presented in collaboration with the Ghana Educational Service (GES). This conference was made possible through the generous support of USAID and the American people.