Data mining approach to predicting the performance of first year student in a university using the admission requirements

• Published in: Education and information technologies 2019-03, Vol.24 (2), p.1527-1543
• Main Authors: Adekitan, Aderibigbe Israel, Noma-Osaghae, Etinosa
• Format: Article
• Language: English
• Subjects: Academic Achievement; Accuracy; Admission Criteria; Algorithms; Analysis; College Admission; College admissions; College Freshmen; Computer Appl. in Social and Behavioral Sciences; Computer Science; Computers and Education; Data Collection; Data mining; Education; Educational Research; Educational Technology; Foreign Countries; Grade Point Average; Information Systems Applications (incl.Internet); Mathematics; Peer Influence; Prediction; Secondary Schools; User Interfaces and Human Computer Interaction
• ISSN: 1360-2357; 1573-7608
• DOI: 10.1007/s10639-018-9839-7

The academic performance of a student in a university is determined by a number of factors, both academic and non-academic. Student that previously excelled at the secondary school level may lose focus due to peer pressure and social lifestyle while those who previously struggled due to family distractions may be able to focus away from home, and as a result excel at the university. University admission in Nigeria is typically based on cognitive entry characteristics of a student which is mostly academic, and may not necessarily translate to excellence once in the university. In this study, the relationship between the cognitive admission entry requirements and the academic performance of students in their first year, using their CGPA and class of degree was examined using six data mining algorithms in KNIME and Orange platforms. Maximum accuracies of 50.23% and 51.9% respectively were observed, and the results were verified using regression models, with R 2 values of 0.207 and 0.232 recorded which indicate that students’ performance in their first year is not fully explained by cognitive entry requirements.