Identifying Turnover Contagion in Enterprise Social Networks

Turnover contagion has created salient challenges for companies and society as turnover itself causes more turnover. Technically, due to the nonlinear growth of overall turnover rate within a short time, turnover is difficult to accurately predict. As a result, predicting and managing turnover contagion have become key issues in people analytics systems. However, existing studies on turnover contagion has primarily focused on its consequence such as time-lagged turnover, while how to identify turnover contagion effect and leverage it to improve the interpretability of turnover prediction algorithm remains under explored. Building upon the behavioral contagion theory, this research-in-progress paper proposes using machine learning method to calculate the turnover contagion effect based on the susceptible-infective epidemic model. This study lays a theoretical foundation to leverage social network data to examine turnover contagion, and practically contributes to managerial practice in designing turnover prediction systems for both efficiency and interpretability