Classroomlevel factors affecting mathematics achievement : a comparative study between South Africa and Australia using TIMSS 2003

Abstract

The purpose of the study was to explore and compare key classroom level factors affecting mathematics learner achievement for South Africa and Australia. The study focused in the classroom where teaching and learning takes place. This is a secondary analysis of classroom level factors influencing Grade 8 mathematics learner achievement using the Trends in International Mathematics and Science Study (TIMSS) 2003. TIMSS 2003 was chosen because it was the latest international study available to measure trends in mathematics learner achievement, where South Africa had participated. Quantitative research approach was employed and a survey research method was used which seeks, among others, to explore relationships and patterns. Survey research method was suitable to provide data that responded to the research questions. The data collection in South Africa and Australia was conducted in October-December 2002 as both countries are located in the Southern Hemisphere. The sample for South Africa consisted of 255 schools with 100% coverage and stratification done by a total of nine provinces, and language. This resulted in 8952 learners tested across the provinces (Joncas, 2004, p. 212). For Australia, the sample consisted of 207 schools with 100% coverage and stratification done by a total of 8 States and Territories and school type. This resulted in 4791 learners participating in the study. The sample included teachers of learners who were selected to participate in the TIMSS 2003 study for South Africa and Australia. The intended target was teachers of all learners at the end of their eight year of schooling. For each participating school, a single mathematics class was sampled and the mathematics teacher of the selected class was asked to complete a mathematics questionnaire. Mathematics teachers of sampled learners responded to questions about teaching emphasis on the topics in the curriculum frameworks, instructional practices, professional training and education and their views on mathematics. The mathematics teacher questionnaire was designed to take about 45 minutes to complete The main question for this study was “What are the key classroom factors that influence learner performance in mathematics?” The three sub questions for the study were: What key variables on classroom level are related to learner achievement in mathematics for South Africa? What key variables on classroom level are related to learner achievement in mathematics for Australia? How do the classroom level factors in mathematics performance of South Africa compare with classroom level factors in Australia? The conceptual framework for the study stressed classroom level factors including instructional quality, which includes teacher background factors, classroom climate, teaching requirements and mathematics curriculum. The framework describes the factors related to classroom interactions within the comprehensive education system, with regard to inputs – process – outputs – outcomes. The selection of variables for the inclusion in the models was guided by the conceptual framework and extensive preliminary analyses. Preliminary statistical analyses included exploring descriptive statistics, Varimax factor analysis, reliability, correlation analysis and stepwise multiple regression analysis. The results of the study indicate that several specific classroom level factors were associated with the higher levels of mathematics achievement of South Africa and Australia. The results for the final South African model were: age of teacher; years been teaching; outside school day grading tests; outside school day other; and computer shortage were identified to predict learner achievement. For Australia ten classroom factors, namely, teacher perception of school climate; teacher perception of school safety; teacher emphasis on mathematics homework; teacher repeat mathematics limiting factors; homework contribute towards learning; work conditions; unhappy learners; shortage of instructional equipment; geometric shapes; and algebraic functions were identified to predict learner achievement. South Africa has factors like teacher background and outside school activities by the teacher. Australia has factors like classroom climate, work conditions and curriculum quality. In the light of schools effectiveness research and school improvement research, a comparative study like this one would require more than one level (classroom level), two or three levels would have been ideal to draw other variables and enrich the analysis, especially the learner level and school level. School effectiveness places an emphasis on the ability and social background of the learners as factors that shape academic performance