ANALISIS STRUCTURAL EQUATION MODELLING PENDEKATAN PARTIAL LEAST SQUARE DAN PENGELOMPOKAN DENGAN FINITE MIXTURE PLS (FIMIX-PLS) (Studi Kasus: Kemiskinan Rumah Tangga di Indonesia 2017)

*Esta Dewi Anggita  -  Departemen Statistika, FSM, Universitas Diponegoro, Indonesia
Abdul Hoyyi  -  Departemen Statistika, FSM, Universitas Diponegoro, Indonesia
Agus Rusgiyono  -  Departemen Statistika, FSM, Universitas Diponegoro, Indonesia
Published: 28 Feb 2019.
Open Access Copyright 2020 Jurnal Gaussian
License URL: http://creativecommons.org/licenses/by-nc-sa/4.0

Citation Format:
Abstract

Poverty is a complex and multidimensional problem that links several dimensions. Statistical method that can explain the relationship between one latent variable with others is Structural Equation Modelling (SEM). The purpose of this study is to create a structural model of the relationship between education, health, economy and poverty in Indonesia in 2017 by using Structural Equation Modeling with Partial Least Square approach (SEM-PLS) based on predetermined indicators with the results of 11 valid indicators. Based on the model obtained, health has a significant positive effect on education, health and education have a significant positive effect on the economy and the economy has a significant negative effect on poverty. Segmentation based on the relationship of latent variables in structural models can be overcome by Finite Mixture Partial Least Square (FIMIX-PLS) so that it can identify poverty areas in each province in Indonesia with more homogeneous characteristics. The best segmentation result is number of segments (K) = 2 obtained based on the criteria of AIC, BIC, CAIC and Normed Entropy (EN) with an EN value of 0.964 which means the quality of segment separation is very good. Papua and West Papua provinces form one segment in segment 2, while 32 other provinces form one segment in segment 1.

Keywords: Poverty, Structural Equation Modelling, Partial Least Square, Finite Mixture, Segmentation.

Keywords: Poverty, Structural Equation Modelling; Partial Least Square; Finite Mixture; Segmentation.

Article Metrics: