Employee turnover in multinational corporations: a supervised machine learning approach

Abstract

This research explores the potential of supervised machine learning techniques in transforming raw data into strategic knowledge in the context of human resource management. By analyzing a database with over 205 variables and 2,932 observations related to a telco multinational corporation, this study tests the predictive and analytical power of classification decision trees in detecting the determinants of voluntary employee turnover. The results show the determinants of groups of employees who may voluntarily leave the company, highlighting the level of analytical depth of the classification tree. This study contributes to the field of human resource management by highlighting the strategic value of the classification decision tree in identifying the characteristics of groups of employees with a high propensity to voluntarily leave the firm. As practical implication, our study provides an approach that any organization can use to self-assess its own turnover risk and develop tailored retention practices.