The purpose of this study was to find out whether there was a relationship between playing chess and learners' achievement at Mathematics. To investigate the relationship an ex post facto, quasi-experimental research design was used. Learners from two Senior Primary Schools in Kwazulu-Natal who had active school chess clubs were selected to participate in the study. The learners' average mathematics marks at entry year to their Senior Primary Schools were considered the pre-test data. These marks were compared to their mathematics marks at their current grade that was considered the post-test data. In addition a further analysis was done with a group of chess players and a carefully selected group of matched non-players using mathematics marks at entry year as the matching criteria. In all cases the treatment was considered to be the current active participation in chess. The data and background information about the groups was obtained from teachers' interviews, existing school records and a questionnaire that was completed by the participating learners. It was found that for the chess players (the test group) the improvement in the average mathematics mark at Grade 7 (post-test) compared to their entry year average mathematics mark (pre-test) was significantly higher than that of non-players. While the chess players' marks improved, the non-players marks (control group) declined. This finding was further supported by analysis of the matched pairs where the same trend was found. Statistical analysis using t-test found that the results were significant. Further detailed analysis of sub groups within the data revealed that current chess players who were below grade average at Grade 4 had improved their mathematical achievement by even more compared to their matched non-players, at Grades 5, 6, and 7. This points to a possible positive causal effect between chess and mathematics achievement for below average achievers. When the amount of exposure to chess was investigated, no correlation was found between the amount of chess played (frequency and length of time) and the level of improvement in mathematical achievement. However, playing chess for a period longer than six months, did positively affect the mathematical achievement of Grade 7 active chess players and the mathematical achievement of the weaker learners at all Grades for which data was collected in the study. This study implies that the incorporation of chess into school activity and further encouragement for all learners to play the game should be seriously considered by the education authorities since it is likely to result in the overall improvement of the mathematical achievements especially in the higher grade of the Senior Primary School. / Thesis (M.Ed.)-University of KwaZulu-Natal, Durban, 2004.