The ability to predict voluntary turnover is crucial for every company to prevent a broad range of financial losses. This prediction is usually based on time-consuming surveys with low response rates. The research question of this paper is whether the available data in the HR system can result in reliable predictions.
We will discuss the usage of decision trees and logistic regression. Each technique will require different data preparation and will bring different determinants of voluntary turnover to the surface.
An illustration will be given on a real life dataset of our own university. The model performance will be evaluated by the AUC measure, making it clear that the available data can be used by HR managers to prevent and reduce voluntary turnover.
Original languageEnglish
Title of host publicationSMTDA2016 Book of abstracts 4th Stochastic Modeling Techniques & Data Analysis International Conference
PublisherISAST-International Society for the Advancement of Science and Technology
ISBN (Electronic)978-618-5180-15-7
ISBN (Print)978-618-5180-14-0
Publication statusPublished - 2016
EventStochastic Modeling Techniques and Data Analysis International Conference - University of Malta, Valetta, Malta
Duration: 1 Jun 20164 Jun 2016


ConferenceStochastic Modeling Techniques and Data Analysis International Conference
Abbreviated titleSMTDA2016
Internet address

    Research areas

  • Human Resource Management, turnover, wastage, decision support system

ID: 25187212