How Do we Know What to do? The Importance of Experimentation in Designing Programmes and Policies to Improve South African Schooling

The quality of schooling in South Africa is a continuing cause for concern. There is broad consensus around the goals of reform, but not about how these goals might or should be achieved. This is because there is little evidence on the effects of alternative reforms allowing assessment of their efficacy and utility. This paper argues for much more experimentation in educational reform, i.e. running experimental reforms and monitoring closely their effects. Some evidence is available from projects and programmes funded by foreign donors, business donors or occasionally provincial education departments. This paper examines some of these, focusing especially on the design and evaluation methodology of the Imbewu programme run in the Eastern Cape funded through British aid. Unfortunately, little attempt has been made to standardize or collate research on the effects of these initiatives.