Sitti Sham Amir and Siti Rahayu Mohd Hashim and Dg Siti Nurisya Sahirah Ag Isha and Mohd Khairuddin Abdullah @Jerry Abdullah (2024) Multiple regression modelling for mathematics performance: best model selections. International Journal of Academic Research in Progressive Education and Development, 13. pp. 1137-1150.
![[img]](https://eprints.ums.edu.my/style/images/fileicons/text.png) |
Text
ABSTRACT.pdf Download (39kB) |
|
Text
FULL TEXT.pdf Restricted to Registered users only Download (521kB) |
Abstract
Initiating mastery of mathematics in primary school is pivotal for successful learning at higher levels. Multiple regression analysis stands as a cornerstone in statistical methods for modelling mathematical achievement. However, despite its prevalence, earlier studies often neglect to disclose the essential assumptions requisite for effective multiple regression modelling. Moreover, the impact of variable selection methods on model generation and subsequent identification of the optimal model remains insufficiently explored. Considering these gaps, this study was undertaken to identify significant factors influencing students' mathematics achievement while ensuring adherence to multiple regression analysis assumptions. Utilizing demographic data, the number of books, home educational resources, student attitudes, and mathematics anxiety as independent variables, two models were derived: Model 1 incorporated all variables without domains, while Model 2 included domainspecific variables and adhered to multiple regression assumptions. The findings revealed that the Model 2 is the best model since it has highest , adjusted , lowest standard error of estimation, lower values in 8 selection criteria which also fulfilled assumptions of multiple regression analysis. In conclusion, key determinants of mathematics achievement were identified as the number of books (101-200), student confidence, and mathematics learning anxiety. The constructed model elucidated 27.6% of the variance in mathematics achievement. This study underscores the importance of meeting regression test assumptions for modelling accuracy and provides actionable insights for schools to design interventions aimed at enhancing mathematics achievement among fifth-year students and the broader elementary school population.
| Item Type: | Article |
|---|---|
| Keyword: | Mathematics Achievement, Multiple Regression, Mathematics Anxiety, Attitudes towards Mathematics |
| Subjects: | L Education > LB Theory and practice of education > LB5-3640 Theory and practice of education > LB1501-1547 Primary education Q Science > QA Mathematics > QA1-939 Mathematics |
| Department: | FACULTY > Faculty of Science and Natural Resources |
| Depositing User: | SITI AZIZAH BINTI IDRIS - |
| Date Deposited: | 18 Oct 2024 15:49 |
| Last Modified: | 18 Oct 2024 15:49 |
| URI: | https://eprints.ums.edu.my/id/eprint/41485 |
Actions (login required)
 |
View Item |