BINARY LOGISTIC REGRESSION MODEL WITH APPLICATION IN STUDENTS G.P.A PERFORMANCE

1Sule Omeiza Bashiru, 2Aminu Ibrahim, 3Alhaji Modu Isa and 1Ibrahim Ismaila Itopa

1 Department of Mathematical Sciences, Prince Abubakar Audu University, Anyigba, Kogi State
2Department of Statistics, Confluence University of Science and Technology, Osara, Kogi State.
3Department of Mathematics and Computer Science, Borno State University, Maiduguri, Nigeria

*Corresponding author: bash0140@gmail.com, doi.org/10.55639/607mlkji


ABSTRACT

The quality of results obtained by Statistics students of Prince Abubakar Audu University, Anyigba in their first semester 100 level examination has been called into question owing to their poor performances in some courses they offered. Considering the pivotal role of students’ early academic achievement in determining the overall success of their university education, university educators must gain insight into how various courses influence students’ main indicator of academic success which is the grade point average (GPA). This empirical study examined how the various courses offered by 100-level students of statistics in the aforementioned university influence students’ academic performance at the end of the first year using the Binary logistic regression model. The study comprises nine eight (98) students. Negrerke R2 test shows that 84.2% variation in the log odd ratio was explained by the fitted model. Also, 63.7% explained variation from Cox and Snell R test, the Hosmer and Lemeshow test and the classification results further confirmed the model is robust enough for such a study. Five courses namely MAT 101, PHY 101, STA 111, CHM 101 and BIO 101 were identified to have significantly influenced students’ early academic achievement. One of the recommendations made is that an extra learning period for each of these courses should be added to the school lecture timetable to enable students to grasp the understanding of basic concepts of these courses

Keywords:

GPA
Academic Performance,
Binary Logistic Regression,
First Semester Courses,