e-Behaviour, personality and academic performance analysis for intervention
No Thumbnail Available
Date
2021
Authors
Seota, Serepu Bill-William
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Post-secondary educational institutions have a concerningly low rate of completion.
Problems include a lack of student support, a student’s social, cultural and economic
background, and inability to adapt to the curriculum. This study provides insight into
the relationship between e-Behaviour, personality and performance. Student perfor mance analysis involves data modelling that enables the formulation of hypotheses
and insights about student behaviour and personality. While the automation of such
data pipelines is efficient and economical, it still requires a thorough analysis of
data and the results obtained from its usage. To achieve the exploratory data anal ysis and prediction objectives of this research, the algorithms used are Regression
Analysis, Decision-Tree, Support Vector Machine, k-Means clustering, k-Nearest
Neighbours, and the Long Short-Term Memory. Our procedures provide methodol ogy to timeously identify students who are likely to become at risk of poor academic
performance. For the education sector, this study is valuable because it presents an
approach to examining the extrinsic influences – e-Behaviour and personality – on
performance. We extract online behaviours as proxies to Extraversion and Consci entiousness, which have been proven to correlate with academic performance. The
proxies of Extraversion and Conscientiousness yield significant (p < 0.05) pop ulation correlation coefficients for the personality traits against grade: 0.846 for
Extraversion and 0.319 for Conscientiousness. Using engineered e-Behaviour and
personality features, we obtained a classification accuracy (κ) of students at risk of
0.51. Lastly, we design an intervention process that supplements existing perfor mance analysis and intervention methods
Description
A dissertation submitted in fulfilment of the requirements for the degree Master of Science to the Faculty of Science, School of Computer Science and Applied Mathematics, University of the Witwatersrand, Johannesburg, 2021