Gender-based comparison of students’ academic performance using regression models

Abstract

The process required to gain admission into tertiary institutions is challenging for Nigerian students. This is due to the various examinations and requirements that must be met to be qualified for admission among in numerable applicants. It is therefore imperative that after admission, students must pursue academic excellence to justify the opportunity given to them, and this will also improve their chances of success after graduation. In a university, some first-year students struggle because of cultural disadvantages, as well as their economic and social backgrounds. This has resulted in poor performance by some students and inevitably led to bad grades at graduation and some dropout of the university without graduating. Female students are often said to perform a bit poorer in terms of academic performance. This study is a comparative performance analysis of male and female students in Science, Technology, Engineering and Mathematics (STEM),conducted using regression models. Trend analysis shows that in this case study, female students have a tendency to improve on their academic performance from their first to their final year. The highest R-squared value of 0.7069 was achieved based on a regression analysis of the performance of1,093 female students.

Keywords

• teaching and learning strategies
• data mining algorithms
• educational data mining
• machine learning
• knowledge discovery
• performance evaluation methodologies