Case Study: Rutgers Business School achieves employee attrition prediction to improve retention with Alteryx

Predicting Employee Attrition Using Human Resource Data

This Alteryx Community case study, created by Nitesh Sharma (Student, Rutgers Business School), tackles employee attrition prediction using HR data. The dataset includes employee satisfaction, last evaluation, number of projects, average monthly hours, tenure, accidents, promotions, department, salary and whether the employee left. The business challenge was to identify which valuable employees are most likely to leave so the organization can take preventive action.

The solution used a CSV input and R-based predictive tools: independent (employee attributes) and dependent (left) variables were separated, principal component analysis identified features correlated with attrition, and a logistic regression model produced predictions. The workflow outputs Prob_Leave and Prob_Stay (and an accuracy score comparing predicted vs. actual), enabling early identification of staff at risk of leaving.

Rutgers Business School

Nitesh Sharma

Lecturer