<![CDATA[The purpose of this study is to model and segment the health status at the regency/city level in the Southern Sumatra region by considering the complexity of relationships among the variables of Poverty, Economic Welfare, Environment, Utilization of Health Services, and Educational Attainment. The method used to form latent segments is Finite Mixture Partial Least Square Structural Equation Modeling (FIMIX-PLS SEM), which classifies latent heterogeneity based on finite mixture distributions into membership probabilities for each segment. The results obtained from the FIMIX-PLS SEM analysis indicate the formation of 2 optimal segments based on nearly all segment selection criteria. Segment 1 consists of 49 regencies/cities that are more strongly influenced by poverty and economic welfare. Segment 2 consists of 11 regencies/cities that are more strongly influenced by environmental factors, utilization of health services, and educational attainment. The formation of segments also increases the R² value, thereby increasing the variation in Health Degree that can be explained by the exogenous latent variables (Poverty, Economic Welfare, Environment, Utilization of Health Services, and Educational Attainment). Therefore, FIMIX-PLS SEM is capable of improving the accuracy of the analysis results.]]>
Copyrights © 2026
