Determinants of employment in Turkey with spatial econometrics analysis

Regional studies have become common in recent years. Spatial econometric analysis has been developed due to the inadequacy of the use of classical econometric methods for the analysis of the relationship between geographically close places. Spatial econometric analysis with spatial autocorrelation and spatial heterogeneity features include environmental effects in the model. Spatial econometric analysis, which is widely used in regional analysis, has started to be used more with the advancement of computer programs. In this context, the spatial econometrics technique was explained in the study, and at the end of the study, the factors that have an impact on employment, which is an important issue for Turkey, were analyzed by including the proximity factor in the model. In the study, spatial cross-section analysis was made with the data of 2020 of Turkey's 81 provinces. Employment variable is dependent variable; Number of illiterate people, number of undergraduates, population, net migration rate, number of natural gas consumption, number of unemployment and gross domestic product per capita were evaluated as independent variables. While examining the direction of the relationship between the variables, it was also analyzed whether there was a spatial effect. Regions with spatial effect were analyzed with Moran I test statistic and LISA test statistic. It has been determined that there is spatial autocorrelation in the positive direction. The spatial error model was determined as the appropriate spatial model for the analysis of the factors affecting employment, the direction of the correlation between the variables was analyzed and evaluated economically. Keywords: Employment, Employment in Turkey, Spatial Econometrics, Spatial Regression Models, Specification Tests