A study of achievement in mathematics with special reference to the relationship between attitudes and attainment.
This study was designed to explore the patterns and trends in attitudes towards mathematics and attainment in mathematics among Indian pre-matriculants. It also aimed at investigating the relationships between attitudes and attainment and the effects of differences in sex, grade, levels and certain background variables on mathematics attainment and attitudes. A set of 24 multiple-choice mathematics test items, compiled to test cognitive outcomes in mathematics at three taxonomic levels, served as a measure of attainment in mathematics. An attitude scale of 48 Likert-type items comprising six dimensions was developed to measure pupil's attitudes towards mathematics (affective outcomes). In addition, questionnaires were used to collect data on selected background variables and teachers' ratings of the pupils. The test, attitude scale and questionnaires were administered to 680 pupils selected randomly from 17 secondary schools in Durban. The 53 teachers responsible for the 151 mathematics class units (Std 9 and Std 10) at these schools also participated in the research project. The data was subjected to statistical analyses (item analyses, correlational analyses, z-scores and ANOVAS) by computerization. The reliability and validity of both the mathematics test and the attitude scale were demonstrated. The potential value of these instruments as measures of cognitive and affective outcomes in mathematics has been presented and argued. The significances of the relationships in respect of the background variables (including sex, grade and levels) and mathematics attitudes and attainment have been carefully documented (see summary - section 6.4). The relationship between attitudes towards mathematics and attainment in mathematics was found to be positive and significant, with no difference between males and females. It was also demonstrated that attainment in mathematics might be predicted from attitude and ability (IQ) scores, and a regression equation was derived for this purpose. Finally, consideration was given to implications of the major findings and problems for future research.