Detailed Spatial Analysis of Parliamentary Elections in 2010 Using Geographically Weighted Regression

The thesis deals with the geographical analysis of the parliamentary elections in the Czech Republic in 2010. Its main goal is a detailed analysis of the elections using the method of geographically weighted regression (GWR), statistical method, which allows to uncover spatial context of the examined phenomenon more precisely. Furthermore, it is investigated whether and how the entry of new political parties on the scene (TOP 09, Věci veřejné) in 2010 changed the territorial relationship between voting behavior and the socioeconomic characteristics of the population compared to 2006 elections. The data at the municipal level from the Census 2001 and 2011 and election data are used. Data are analyzed using GWR, analysis of variance and multiple linear regression. It is shown that the global decline in turnout in 2010 was greater in the left-wing than in the right-wing regions. The existence of regional differences in the relations between local characteristics and the results of the election of the parties is demonstrated. These observed differences appear to be relatively stable - results in 2010 for each party are similar to results from 2006, especially for the ODS and the KSČM. Explanatory variables for voting for TOP 09 are very similar to those of ODS, with a larger positive effect of..