Model angka putus sekolah anak usia wajib belajar di Provinsi Sulawesi Selatan (Pendekatan Generalized Poisson Regression)

Ahmad Zaki
Usman Mulbar
Nasra Nasra

Seminar Nasional Hasil Penelitian LP2M UNM
Prosiding Edisi 6

This study aims to examine the model of school dropout rates for compulsory children in South Sulawesi Province using the Generalized Poisson Regression approach. The type of research used is applied, namely the application of nonlinear regression models. The data that will be used for this research is secondary data obtained from government agencies, namely the Central Statistics Agency of South Sulawesi Province. The response variable used is the number of dropouts in compulsory education and the predictor variable of the ratio of students to the number of schools, the ratio of students to teachers, the percentage of poor people, population density and average length of schooling. The results showed that the best model based on the criteria of AIC was a model that involved all predictor variables in the model, but because the data observed were overdispersed cases so that the dropout rate modeling in South Sulawesi used Poisson regression model said to be unfit for use.  Based on the model obtained, it can be said that the variables that significantly influence the model are the ratio of students to teachers, the number of poor people, population density, and the average length of schooling.