Comparison of Cox Weibull and Cox Log-Normal Regression Models for Job Waiting Time Data

The purpose of this study is to evaluate how well the Cox Weibull and Cox Lognormal regression models perform using data on graduate job search duration. Alumni from UIN Sunan Gunung Djati's Faculty of Science and Technology provided the data encompassing variables such as gender, major, and GPA. The impact of independent variables on job search duration was assessed using the Cox Weibull and Cox Lognormal regression models, while the survival distribution was analyzed using the Kaplan-Meier technique. Results revealed that most respondents were female, dominated by Mathematics majors with GPAs ranging from 3,25 to 3,74. Gender significantly influenced job-seeking time, with males being more likely to secure a job faster than females. Based on Akaike Information Criterion (AIC) values, the Cox Lognormal regression model outperformed the Cox Weibull model in describing the data. This study concludes that the Cox Lognormal regression model is the best representation for analyzing job-seeking time data.