Estimating the average duration of employment through predictive modeling to increase retention

Employee retention has become increasingly popular as the costs of turnover and competition for top talent continue to grow. This capstone project aims to analyze employee retention at an organization through a data-driven approach. We are investigating the factors influencing the retention and turnover of both full-time and part-time employees. For the purposes of this project, the word termination refers to an employee quitting their job voluntarily or getting fired. The objective of this project is two-fold: estimating the typical duration of employment for an individual based on other factors, and implementing techniques of survival analysis so the Human Resources department can take a more proactive approach to retaining employees. This study specifically utilizes data from the past three years containing information on demographics, job-related attributes, and eval- uation metrics. The project conducts exploratory data analysis along with identifying different interactions among the existing variables to gain valuable insights and examine patterns. By predicting the average lifespan of an employee and identifying the variables that have the most influence on survival probability, the organization can enhance employee retention, therefore reducing possible costs associated with turnover, and maintaining a skilled and stable workforce.