Font Size: 

The loss of participation in the informal sector of the Work Accident Insurance program with the type of work that has a high risk of work accidents, is still a problem by the province of East Kalimantan with the number of informal sector reaching the third highest nationally, amount to 48.12% in 2020 to date. The purpose of this study was to compare the best predictive models from the results of logistic regression tests, using the logit and probit models, to predict the factors that affect the ownership of Work Accident Insurance in 3,869 informal sector workers in East Kalimantan province This research is a non-reactive research, with the secondary data source used being the Susenas of East Kalimantan province in 2020. The results showed that the variables of gender, ownership of a Population Identification Number, type of area, marital status, the latest education, and PKH recipients had a significant influence on the ownership of Work Accident Insurance. The best prediction model is generated by the logit model with an AIC of 1976.8 which is smaller than the AIC value of the probit model of 1981.2 with the result that the R-Square value of the logit model is 6.12% and the percentage is greater than the probit model which is only 6.05%. The value of R-Square percentage result is strengthened by the hit ratio in both models of 92.2%, which can represent the accuracy of the two predictive models on the ownership of the informal sector Work Accident Insurance. The suggestion of this research for further research is that further studies are needed to explore more deeply the causal factors that influence the role of gender differences in workers' participation in employment security between men and women workers in the informal sector

logit model;probit model; pendidikan