Longitudinal Studies of Random Effect Model on Academic Performance of Undergraduate Students

Abstract

This paper discussed the longitudinal studies of random effect model on academic performance of student using Federal University of Technology, Owerri Imo State Nigeria as a case study. Secondary data were adopted for the research work, and a SAS software package was used for the analysis. There appears to be some curvature in the average trend and individual profile plots, and hence a quadratic time effect was fitted to the data. From the individual profiles are the total observations collected for the analysis. From the profiles of the type of SSA, Entry Age, Entry Aggregate and Gender, it could be assumed that each profiles evolution follows a quadratic trend. Also, it could be concluded that most students who started with low GPA at semester one, improved in their performance to semester three and there was a downward trend before semester seven. Further, the mean profile for SSA was explored. From the chosen model among all models fitted to the data set, we conclude based on the results obtained that student’s GPA depends on the SSA, Entry Age, Entry Aggregate and Gender). Student with high and medium admission aggregates scores high GPA and student with low admission aggregates scores low GPA at semester one, but on the average students with Low and Medium Entry Aggregate score higher GPA than students with High Entry Aggregate. The performance of GSS students is better as compare to that PSS at semester one and on the average. Meanwhile, in all the models it appeared, student GPA’s increase from semester one to semester three and decreases after semester three. Generally students tend to perform better at the third semester. The analysis also revealed that the academic performance is dependent on the SSA, Entry Age, Entry Aggregate and Gender.