Penerapan Structural Equation Modelling-Partial Least Squares pada Faktor Kemiskinan di Jawa Tengah

Abstract

The number of poverty-stricken people in Central Java in March 2020 was 3.98 million people (11.41%), the second-largest in Java. The approximately high number of poverty-stricken people is a priority for the government to reduce poverty. One of the solutions to reduce poverty is knowing the factors that may affect it. The purpose of this study is to identify the factors that affected poverty in Central Java using the Structural Equation Modeling-Partial Least Squares (SEM-PLS) method. This study used data from districts/ cities in Central Java in 2020. In this case, there is one exogenous latent variable for health and three endogenous latent variables for poverty, economy, and human resources. The problem encountered that the observed data is relatively small, specifically for 35 observations and the data distribution is suspected not fulfilled the normal assumptions. In conclusion, the appropriate analysis used in this study is Structural Equation Modeling-Partial Least Squares (SEM-PLS). The results showed that the economic latent variable had a positive but not significant effect on the latent variable of poverty, Human Resources also had a positive but not significant effect, while the latent health variable had a negative and significant effect on the latent variable of poverty. The Q2 value for the latent variable of poverty is 0.333, this shows that 33.3% of the diversity of the latent variable of poverty can be explained by the latent variables of economy, health, and human resources.