After deliberations since 2006 between The Shuttleworth Foundation, the Western Cape Education Department and the Human Sciences Research Council, an agreement was reached late in 2006 to undertake a study during 2007 to pilot or demonstrate an intervention procedure that would be aimed at increasing the contact time of Grade 8 learners in Mathematics and English with a view to improving their performance in the two learning areas. This would be an attempt to explore whether the perceived serious underperformance of learners, that may also be the result of poor foundational knowledge, could be remedied through a 20-hour tuition programme delivered after school hours. It was hoped that within a year's time, a 10 %-point increase in learner marks would not be unrealistic. The implementation of the intervention programme got delayed by the public service labour action and incumbent teacher strike during June 2007. The baseline learner performance assessment and contextual survey information collection was fitted in before that, but the tuition roll-out could only start at the beginning of the third term in 2007. Special permission was obtained to extend the programme into the fourth term, instead of it running from Term 2 to Term 3. The study comprised a paired/matched control group design. It took place in only one "district" in the Western Cape (Metropole-South EMDC). Eight potential control schools were initially recruited, reduced to four later, and matched with the four project schools. Baseline information comprised learner performance assessments on a multiple-choice question Mathematics and English learner performance test, as well as the collection of various pieces of contextual and background information from learners, their parents, teachers, tutors and school principals. Some observation of classrooms and school sites, as well as document review at the level of schools, teachers and learners also took place to collect information on aspects such as facilities, systems, management practices, and more. After completion of the tuition, learner performance was again assessed. This was followed by the required data capture and verification procedures, after which difference-indifference analyses were conducted to assess if project-school learners' performance indeed improved more than that of control-school learners. The techniques used mainly comprised analysis of variance, chi-square analysis and correlation analysis. The findings were inconsistent, and the effects smaller than expected on the whole. The main consistency, albeit of a smaller extent than hoped for, was that better attendance at Mathematics tuition sessions correlated positively and significantly with higher performance improvements over time among the project-school learners. In terms of English tuition, the slightly more consistent finding that project-school learners' performance improvements slightly exceeded those of control-school learners was heartening. However, many contextual and background factors, as well as aspects related to the conceptualisation and delivery of the tuition sessions, appear to have probably influenced the findings further. In short, the expertise of tutors, some school and classroom factors such as levels of order and discipline, and time on task, and many parental and learner factors related to exposure to opportunity to read and write, and other general learner support, seem to have put a ceiling on how much learner performance could improve. A key suggestion flowing from this, also based on the very low levels of baseline performance of all the learners, is that backlogs and low foundation knowledge levels, as created as early as the Foundation Phase, seriously hamper current performance levels and the ability of learners to benefit from extra tuition in secondary school. As a result, it is recommended that similar and other intervention programmes, also aiming at increasing contact time, especially in the learning areas of Literacy and Numeracy, be established at the Foundation and Intermediate Phase levels. Their conceptualisation and implementation should ideally be dealt with within a peer-support model, where secondary schools twin with their primary feeder schools, to identify, address and prevent the kinds of knowledge gaps that seem to appear to hamper learner performance later down the line. Apart from many more detailed observations and recommendations (selected and summarised in more detail in the Executive Summary on p.5), the importance of language ability for performance in Mathematics (and other learning areas), has been observed clearly in a number of cases.
Human Sciences Research Council (Wed,) studied this question.