REGIONALISM, ELECTION BEHAVIOR, AND ELECTION RESULT: A SPATIAL REGRESSION ANALYSIS

Abstract

The objective of this study is to describe about the relationship between voting behavior of PheuThai party and Democrat party in 2011 election results and geographical, behavioral, socio-economic, and demographic and election result in 2005 and 2007 components. To develop model for predicting election result. Model selection compares three models are OLS, Spatial error and spatial lag, criteria for consideration model selection are R², AIC and BIC. The model that was selected must be maximum R², minimum AIC and BIC. This study finds  out that the spatial error model is the best model, the geographic variable and election results in 2005 and 2007 were the good variable for predict election results in 2011. Moran’s I statistics and Local Indicators of Spatial Association (LISA) were used to investigate the spatial autocorrelation and spatial pattern of election results among election districts. The study shows the spatial autocorrelation among each electorate district, Moran’s I statistic of PheuThai Party and Democrat party were 0.73 and 0.79.