Journal of Social Sciences

Effects of Students' Beliefs on Mathematics and Achievement of University Students: Regression Analysis Approach

Velo Suthar and Rohani Ahmad Tarmizi

DOI : 10.3844/jssp.2010.146.152

Journal of Social Sciences

Volume 6, Issue 2

Pages 146-152


Problem statement: At present, after almost more than 20-decades, Malaysia can boast of a solid national philosophy of education, despite tremendous struggles and hopes. The professional learning opportunities are necessary to enhance, support and sustain student's mathematics achievement. Approach: Empirical evidence had shown that student's belief in mathematics is crucial in meeting career aspiration. In addition mathematical beliefs are closely correlated to their mathematics achievement among university students. Results: The literature exposed that a few studies had been done on university undergraduates. The present study involves a sample of eighty-six university undergraduate students, who had completed a self-reported questionnaire related to student mathematical beliefs on three dimensions, viz-a-viz beliefs about mathematics, beliefs about importance of mathematics and beliefs on one's ability in mathematics. The reliability index, using the Cronbach's alpha was 0.86, indicating a high level of internal consistency. Records of achievement (GPA) were obtained from the academic division, University Putra Malaysia. Based on these records, students were classified into the minor and major mathematics group. The authors examined student's mathematical beliefs based on a three dimensional logistic regression model estimation technique, appropriate for a survey design study. Conclusion/Recommendations: The results illustrated and identified significant relationships between student beliefs about importance of mathematics and beliefs on one's ability in mathematics with mathematics achievement. In addition, the Hosmer and Lemeshow test was non-significant with a chi-square of 8.46, p = 0.3, which indicated that there is a good model fit as the data did not significantly deviate from the model. The overall model, 77.9% of the sample was classified correctly.


© 2010 Velo Suthar and Rohani Ahmad Tarmizi. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.