MODELING EMPLOYEE RESIGNATION USING A SEMIPARAMETRIC APPROACH COX PROPORTIONAL HAZARD

Abstrak

Survival analysis is a research method that studies the duration individuals or experimental units endure against events like death, disease, recovery, or other experiences. This study employs a semi-parametric survival analysis model using the Cox proportional hazards regression method to identify factors such as age, gender, marital status, and education influencing how long employees stay with a company before resigning. The aim is to describe and interpret significant factors affecting employee resignation using the Cox Regression method. The results indicate that age significantly influences employee tenure. The average tenure is eight years. The probability of an employee still working at age 32 for up to eight years is 0.0057, while the likelihood for an employee who has worked more than eight years at age 32 is 0.9943. The study uses secondary data on the tenure of 521 employees, analyzed with the Cox proportional hazards regression method. The data, however, has limitations due to type III censoring, where some subjects leave observation, resulting in incomplete data. The study concludes that age significantly impacts employee tenure. Younger employees tend to explore career opportunities, while older employees seek stability, pension benefits, and a comfortable work environment.