Quantitative understanding of relationships between students' behavioral patterns and academic performances is a significant step towards personalized education. In contrast to previous studies that mainly based on questionnaire surveys, in this paper, we collect behavioral records from 18,960 undergraduate students' smart cards and propose a novel metric, called orderness, which measures the regularity of campus daily life (e.g., meals and showers) of each student. Empirical analysis demonstrates that academic performance (GPA) is strongly correlated with orderness. Furthermore, we show that orderness is an important feature to predict academic performance, which remarkably improves the prediction accuracy even at the presence of students' diligence. Based on these analyses, education administrators could better guide students' campus lives and implement effective interventions in an early stage when necessary.
@article{arxiv.1704.04103,
title = {Orderness Predicts Academic Performance: Behavioral Analysis on Campus Lifestyle},
author = {Yi Cao and Jian Gao and Defu Lian and Zhihai Rong and Jiatu Shi and Qing Wang and Yifan Wu and Huaxiu Yao and Tao Zhou},
journal= {arXiv preprint arXiv:1704.04103},
year = {2018}
}
Comments
11 pages, 6 figures and 2 table, Appendix included