Addressing Challenges in Primary Education in Developing Countries

Opinions abound when it comes to making up suggestions on how to solve problems in basic education in developing countries. In contrast, solutions based on evidence are unfortunately scarce. Patrick McEwan of Wellesley College has recently appraised published studies on education research and found less than eighty experiments conducted in primary schools in developing countries during the past four decades that are worthwhile to examine. The dearth in research-based evidence in this area is quite dreadful especially when the need for effective interventions is so high. Worse, when various studies are combined, results are all over the map, that findings based on an average over various countries are effectively nil. Nonetheless, there are few interventions that seem to benefit primary education universally. One is teacher training as shown in the following figure:
Above copied without permission from
Improving Learning in Primary Schools of Developing Countries:
A Meta-Analysis of Randomized Experiments
The red vertical line above marks zero (no effect). Obviously, all but a handful of studies suggest that teacher training leads to improved learning outcomes. One of the very few studies that does not bear this finding is actually from the Philippines (The following is a magnified view of the bottom portion of the above figure):

Study 69 is Tan, J.-P., Lane, J., & Lassibille, G. (1999). Student outcomes in Philippine elementary schools: an evaluation of four experiments. World Bank Economic Review, 13(3), 493-508. The 95% confidence interval in this case is so wide that the results are probably meaningless.

Seeing that teacher training is nearly universal in improving learning outcomes should not be surprising. In fact, most would suggest this as an important step in improving basic education. There are, however, some results that may seem counterintuitive or at least, contrary to popular opinion. In developing countries, it is more likely that pupils likewise face problems like malnutrition and poor health. Thus, a common suggestion is to provide interventions that address the nutritional and health needs of the students. One example is feeding the children in school. The graph below shows that the effects on learning outcomes are not always positive:
Above copied without permission from 
Improving Learning in Primary Schools of Developing Countries:
A Meta-Analysis of Randomized Experiments
In the above case, about a third of the studies lies on the negative side of zero. Another example is de-worming:
Above copied without permission from 
Improving Learning in Primary Schools of Developing Countries:
A Meta-Analysis of Randomized Experiments
Here, the studies are almost evenly divided between the two sides. Addressing the nutritional and health needs of students are, of course, good interventions on their own regardless of learning outcomes. The reason why the above interventions seem ineffective in improving education is that learning is truly multifactorial. The reason why the effect sizes of various interventions when averaged over various studies from different countries gravitate toward zero is the simple fact that primary schools in developing countries need a lot. Addressing only one factor at a time therefore often leads to no noticeable change in learning outcomes....









Comments