IMPLEMENTATION OF GEOGRAPHICALLY WEIGHTED LOGISTIC REGRESSION ON POPULATION GROWTH RATE IN MAJALENGKA

Abstract

This study investigates the factors that influence the population growth rate in Majalengka Regency using Geographical Weighted Logistic Regression (GWLR) approach. The data used includes the number of births (X1), the number of fertile age couples participating in family planning (X2), and the number of couples of childbearing age (X3) in each sub-district. Descriptive analysis showed significant variation in the variables used across Kecamatans. The simultaneous logistic regression test showed a significant effect of the three predictor variables on the overall population growth rate. However, the partial logistic regression test results showed that not all predictor variables had a significant influence individually in each Kecamatan. The logistic regression model proved to be feasible and suitable for making predictions based on observed data. The spatial heterogeneity test shows that there is heterogeneous variation in the population growth rate in each subdistrict. The best kernel weighting was selected using the Fixed Gaussian kernel function based on the lower AIC value. The results of GWLR modeling showed different effects of predictor variables on population growth rate in each subdistrict. Model evaluation shows that GWLR provides important information regarding the influence of predictor variables on population growth rate in Majalengka Regency. This study provides in-depth and contextual insights into the factors that influence population growth at the  local level, which can be the basis for decision-making in regional planning and development.