<![CDATA[Structural Equation Modeling (SEM) is a multivariate statistical analysis method that combines regression analysis with factor analysis. SEM can be used to describe simultaneous linear relationships between observed variables (indicators) and variables that cannot be directly measured (latent variables). In the development of covariance-based SEM, there are still weaknesses based on parametric assumptions that must be met in regression analysis, and one of the classic assumptions that must be met is the assumption that the data is normally distributed. Partial Least Square (PLS) is one solution or alternative method of model estimation to manage SEM modeling with reflective or formative indicators. PLS was created to overcome the limitations of the SEM method. Structural Equation Modeling-Partial Least Square (SEM-PLS) is a powerful analysis method because it allows structural equation modeling with the assumption that the data used does not have to be normally distributed, SEM-PLS can use a relatively small sample size, and the indicators used are reflective, formative, or a combination of both. This study aims to determine the effect of latent variable indicators, namely public service expenditure, economy, health, and education on the Gross Regional Domestic Product (GRDP) in each district/city in South Sulawesi Province in 2022. The indicators used for each latent variable are Public services (employee expenditure, goods and services expenditure, capital expenditure, other expenditure), Economy (goods and services expenditure, capital expenditure), Health (employee expenditure, goods and services expenditure, capital expenditure), Education (employee expenditure, goods and services expenditure, capital expenditure, other expenditure). The results of the study show that the other expenditure indicator on the latent variable of public services and the other expenditure indicator on the latent variable of Education are excluded in the study because they do not pass the loading factor test. The model equation obtained is GRDP = 0.052 Economy - 0.087 Health + 0.321 Education + 0.706 Public Services. The R2 value obtained from the model equation is 0.982, which means that the latent variables of public service expenditure, economy, health, and education can explain the latent variable of GRDP by 98.2%. The latent variables that significantly influence the model equation are public service expenditure and education.]]>
