Towards understanding student academic performance in South Africa: a pilot study of grade 6 mathematics lessons in Gauteng province

Prepared for the Spencer Foundation, AprilDespite widespread acceptance of the notion that improving student performance may have a high economic and social payoff, policy analysts in all countries have surprisingly limited hard data on which to base educational strategies for raising achievement. In South Africa this question is all the more pressing. South African students score at low levels in mathematics and language tests even when compared with students in other African countries (van der Berg and Louw, 2006). Further, the South African government's own evaluations of ten years of democracy show little improvement in educational outcomes despite significant policy changes (DoE, 2006). While some reasons for this poor performance may be evident, and there is widespread agreement that the main challenge in South Africa is the quality of education, there is little empirical analysis that helps policy makers understand the low l level of student performance in South African schools or how to improve it. As a first step toward an empirical approach to unpacking the factors contributing to low levels of learning in South African schools, the Human Sciences Research Council in partnership with a consortium of South African universities and researchers at the School of Education at Stanford University engaged in a small scale empirical pilot study that focuses on the role that teacher skills and practice play in South African students' learning within the socioeconomic and administrative conditions in those schools (and South African society more broadly). The pilot was conducted in a sample of forty primary schools (Grade 6) in Gauteng, a geographically small and highly urbanized province in the northern half of the country. The pilot focused on mathematics lessons. Students in the selected sixth Grade classrooms filled out a short questionnaire on their socio-economic situation and some observations about their school. They took a predominantly fifth Grade mathematics test administered in July, 2007, and then a subset of the sample took the same test a second time in October. Students could choose to do the questionnaire and the test either in English, Afrikaans, or one of several African languages. Almost all chose to use the language of instruction (English or Afrikaans). Their teachers and their principals also filled out short questionnaires. The teacher instrument included questions about mathematics teaching, specifically content and pedagogical content knowledge questions. Each teacher was also asked to do a mathematics lesson with his/her class, and this was videotaped. Researchers visiting the schools/classrooms provided additional notes about the general situation at the school. The data provided by these instruments form the basis of this report. Although the information is copious, the pilot nature of the research means that its main purpose is to test the instruments and to assess the viability of the models. The first chapter contextualizes these schools and teachers. We profile public school teachers in South Africa and in Gauteng and examine the curriculum, teacher education, supervision and evaluation and socio-economic context of the education system in which they teach. In the second chapter, we present our conceptual framework and methodology for the empirical analysis. We also discuss how we gathered the data Student Academic Performance in South Africa and some of the lessons we learned from that process. The third chapter shows the results we were able to glean from the pilot. In the fourth chapter, we discuss the results and what they teach us about designing future research on a larger and comparative scale