Modelling student dropout using statistical and data mining methods

DOI

10.2991/amse-19.2019.8How to use a DOI?

Keywords

student dropout, logistic regression, decision trees, association rules

Abstract

Not completing the study by a large portion of students is a serious problem at universities worldwide. Regardless of the country, numbers are very similar: about one-half of students who enrolled for the bachelor study leave university before obtaining the degree. To deal with this problem, we create models to distinguish between students who successfully completed their study and students who dropped out of university. Models created using traditional statistical analysis techniques (logistic regression) are compared with models created using data mining methods (decision trees, association rules). We use data about students who enrolled for their bachelor study at the University of Economics in Prague in the academic year 2013/2014 in our analysis.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Petr Berka
AU  - Luboš Marek
AU  - Michal Vrabec
PY  - 2019/10
DA  - 2019/10
TI  - Modelling student dropout using statistical and data mining methods
BT  - Proceedings of the 22nd International Scientific Conference on Applications of Mathematics and Statistics in Economics (AMSE 2019)
PB  - Atlantis Press
SP  - 70
EP  - 80
SN  - 2589-6644
UR  - https://doi.org/10.2991/amse-19.2019.8
DO  - 10.2991/amse-19.2019.8
ID  - Berka2019/10
ER  -