Comparing the best model of students performance using logistic regression: A case study of final year result performance of student UniKL MiiT at Division of Corporate Policy & Strategic Planning, Majlis Amanah Rakyat (MARA) / Fairuz Atiqah Muhammad Ali

Some students will graduate from school or university, get interesting jobs, and make a lot of money soon after graduation, while others will struggle to move ahead in the working world. Every student deserves the best chance to be successful after graduation. Therefore, one quality indicator of undergraduate students that has been consistently used for entry selection into graduate programs is the undergraduate Cumulative Grade Point Average (CGPA) which in this paper refers to the average of credit points obtained at the end of a program study as opposed to the Cumulative Grade Point Average, which refers to the credit points obtained for a specific semester of study. The aim of this study is to investigate the main factors that contribute to performance of excellent results students UniKL MiiT at Kuala Lumpur. This study used secondary data source at UniKL MiiT which is student’s final year from semester January until May 2013. For this research the researcher used logistic regression analysis to test the significant influence exist between gender ,marital status, program, score A in SPM, loan , and performance result. Besides that, the researcher used Chi Square test to test association exists between gender, marital status, program, score A in SPM, loan, and performance result. It was found that the forward method is the best predictive model using logistic regression analysis because the classification accuracy is highly predicted, higher in sensitivity, lower in specificity compared to another method. Score A in SPM and loan found to contributed towards performance results. Besides that, the researcher get the outcome that if a student get the more A in SPM, chances that student get the excellent result is higher. Although this study provides important implications to performance results, it is suggested that further research should be conducted to find another factor that can be contributing more towards performance results students at UniKL MiiT