Research on School Effectiveness on Pupils’ Achievement in Developing Countries with Special Reference to Malawi: Some Methodological Issues.

Most previous research on the comparative effectiveness of schools in developing and developed countries, particularly the effect of school inputs and resources on academic achievement, have concluded that the effect of school quality on academic achievement is greater than that of pupil Socio-Economic Status (SES). However, the basis on which this conclusion is based is questionable. Of particular concern are the major methodological and conceptual weaknesses of most school effects research in both developed and developing countries. Most of the studies have heavily relied on production function models and as such veiy few process variables have been studied. In addition, the studies also suffer from conceptual limitations, especially in the way family socioeconomic status variables have been specified. Most studies have tended to use conventional social background measures appropriate to developed countries. On the methodological issues, the studies have suffered from over reliance on single-level models, particularly Ordinary • Least Squares (OLS) regression models to analyse hierarchical data. This paper suggests that a deeper understanding of the process of schooling and the determinants of achievement in developing countries can be gained from applying multilevel models using socio-economic status background measures appropriate to developing countries. The paper also presents results of a study undertaken in Malawi which employed multilevel models in order to address in part, some of the methodological limitations leveled against school effectiveness research.

A ZJER comparitive research on school effectiveness in the developing world with emphasis on Malawian schools.