Analysis of the spatial impact on Turkey’s life satisfaction

Abstract

Spatial data analysis, whose results depend on the location of the event or the object being analyzed, consists of methods that require the use of both objects features and location. Especially developments in take place at GIS and programmes analyzing data enabled methods used in the analysis of spatial data more feasible. Spatial effect exits to forefront in spatial data analysis and it contains both spatial dependence and spatial heterogeneity. Spatial dependence or spatial autocorrelation reflects the situation in which the values observed in a place or region depend on the values of neighbor observations. Spatial dependence violates independence assumption valid for statistical methods. Studies working spatial dependence should use methods taking this dependence into consideration. One of the most common method used in spatial analysis is spatial regression analysis.

The aim of this study is to examine whether there is a spatial dependency structure in the happiness data at the provincial level in Turkey by using different spatial models including spatial effects in the light of the latest developments in the spatial analysis literature. The results of the regression analysis, in which spatial effects are included in the model or not, were compared by using the data from the Life Satisfaction Survey conducted at the provincial level by the Turkish Statistical Institute (TUIK). As a result of the analyzes, it was determined that there is a spatial effect, and with the help of the estimated spatial regression analysis it has been concluded that the variables of suicide and environmental expenditures were directly; unemployment, income and suicide variables have indirect effects.