Who’s Next: Evaluating Employee Churn in Retail using Machine Learning algorithm CHAID

Abstract

Researchers have undertaken numerous studies on employee churn, but retail employers are facing challenges to retain employees. Academicians have emphasized the factors promoting employee churn and have also incorporated ML algorithms to predict employee churn. But the research related to employee churn prediction in the retail sector is missing. The data was collected from 446 retail employees using a questionnaire. This research attempts to demonstrate the performance of ML algorithm i.e., CHAID for predicting retail employee churn using real-time data. The performance assessment of CHAID is based on accuracy, confidence matrix, CHAID decision tree, and AUC. The deployment of efficient employee churn strategies is essential for businesses to survive in today's competitive marketplace, given that lowering employee churn increases productivity and profitability.