Feb 2010: I presented a paper at the 7th annual Educational Symposium for Research and Innovation (ESRI) 2010 hosted by the Graduate School of Education and Human Development (GSEHD) of the George Washington University.
The paper reviewed private-public school comparisons in low-income countries in an attempt to analyze the methodological issues that limit the research and therefore limit the inferences that can be made. Based on 18 studies conducted in 13 countries in Africa, Asia and Latin America, the paper explored the methodologies used in these private-public comparisons based on a five-aspect analytical framework, which covered: study design, choice of treatment variable, dependent variable(s), control variable(s) and statistical methods.
I found that the literature suffers from three main methodological limitations. First is a lack of panel data which limits the possibility of analyzing the effectiveness of private solutions over time. Second is the choice of dependent variables in the studies which are increasingly using test scores as a proxy for educational achievement. This is at the expense of other educational outcomes of concern such as equality, diversity, etc. Finally, each study analyzed here uses a binary treatment variable for enrolment in primary or public schools. This is based on the assumption that there something inherently different between private and public schools. This is an assumption that is not based on any detailed analysis.
The paper and the presentation are here:
Asif Saeed Memon 