Relumining perceived workplace gender discrimination in South Korea: examining determinants and paths through decision trees

Abstract

ABSTRACT This study explores how women employees perceive gender discrimination in the workplace and how data can be used to predict perceived workplace gender discrimination (PWGD). The research team modelled the decision tree that predicted PWGD in South Korea using the Classification and Regression Trees (CART) algorithm and the data from the 7th Korean Women Manage Panel (KWMP). Three types of PWGD trees – wage, promotion, and evaluation – and one synthesised PWGD tree were built to predict and classify PWGD by discrimination type. The research findings suggest that the chief executive officer’s (CEO) fairness is the cardinal factor in predicting synthesised PWGD, followed by an employee’s exposure to sexual harassment. Whereas the CEO’s fairness is the principal factor in predicting PWGD in promotion, the direct supervisor’s fairness is the most significant factor in predicting PWGD in evaluation. Perceived disparities in pay between women managers and similarly positioned men colleagues are the critical factor in predicting wage PWGD. Lastly, this paper elaborates on important considerations in PWGD and recommendations for continued inquiry.