Abstract:
Spatial regression analysis is a statistical method that is useful for analyzing spatial data.
Spatial analysis assumes the presence of spatial dependencies. One way to find out if
there is a spatial dependency is to do a spatial autocorrelation test. Spatial autocorrelation
is used to analyze the similarity of values at observation locations with neighboring
locations on the same variable. There are several spatial models, including the Spatial
Autoregressive Model (SAR) and Spatial Error Model (SEM). The purpose of this study
was to determine the best spatial regression model to model the percentage of tuberculosis
cures in Riau Province and to determine the factors that influence it, by examining spatial
autocorrelation using the Moran Index. Based on the results of the analysis by testing the
Moran Index hypothesis, it was found that there was positive spatial autocorrelation and
on examining spatial dependence, it was found that there was a lag dependence on the
dependent variable, which means that the modeling was done with SAR. Based on the
results of the SAR analysis, it was found that there were three independent variables that
significantly influenced the percentage of TB cures, including the percentage of
households with proper drinking water (𝑋1), the number of medical personnel at the
puskesmas health facility (𝑋3) and the total number of public places (TTU) that met the
requirements health (𝑋5).
