Constraints to school effectiveness: what prevents poor schools from delivering results?

The poor state of quality education in South Africa is confirmed by the weak performance of South African students on international tests, even when compared to countries with comparatively poorer education systems. This paper aims to shed light on this issue through the use of the PIRLS 2006 dataset and education production function techniques. A unique feature of this dataset is that schools were able to choose the language in which the test was conducted. This provided a proxy for former school department, a feature that has not been captured in international survey datasets. A clear distinction between the historically black and the historically white, coloured and Indian school systems is needed in order to identify the different data generating processes at work. The regression model results reveal that family and student characteristics are undoubtedly important for performance within both school samples. At the level of the school, quite divergent school factors and classroom processes were found to have significant impacts on student performance across the two school systems. It is concluded that a lack of enabling conditions such as effective leadership, flexibility and autonomy, and a capable teaching force may contribute to certain school and classroom processes not playing a significant role in determining performance in the less affluent black school system.

C20, C21, I20, I21, I30

South Africa, Education, Education production function, Educational Achievement, Educational Inequality