Clustering Of Ukrainian Regions Based On Value Orientations And Political Choice Of The Populations: Methodological Rationale And Analysis Using Combining Data Sources

The aim of this study is clustering of administrative-territorial units of Ukraine on the basis of value orientations and the electoral choice of the population of these units. The k-means method is used. Creation of macroregions based on the political orientations of the population is quite widespread, but such approaches have a number of limitations, primarily due to the fact that the list of political leaders or political parties can change significantly in rather short periods of time and because of difficulties with using of several political parties/leaders simultaneously in the analysis. The «value» in this article is defined within Schwartz's theory as desirable goals that go beyond specific situations, differ in importance from each other and are guiding principles in human life. The analysis uses the ten Schwartz's values, which are grouped into four dimensions: «Conservation», «Self-Enhancement», «SelfTranscendence» and «Openness to Change». The data set for this study is a combination of two sources of data – sample survey and electoral statistics. Thus, the data set in this study is formed by a combination of the results of the Ukrainian vote in the Parliamentary elections in 2012 and sample survey – European Social Survey – the latest wave of which was held in Ukraine in 2012. The European Social Survey is the most actual source of data on the value orientations of Ukrainians which is in free access. After 2012 this study in Ukraine was no longer conducted. The main result of this study is the creation of clusters of administrative-territorial units based on the similarity of the results of voting and value orientations of population in these units. The first cluster includes administrative-territorial units, where population has more expressed values of Self-transcendence than in Ukraine as a whole. In the second cluster there are units where population has more expressed values of Self-enhancement and Openness to change. The third cluster is characterized by more expressive values of Self-transcendence and Conservation. Except of different levels of expression values, clusters differ by the level of support of political parties that participated in Parliamentary elections. This approach allows evaluate the received cluster structure in dynamics, use in analysis results of national and local elections in different years. Also it makes clustering space two-dimensional, which enables not only to discover similar administrative-territorial units, but also, for example, to identify groups of parties whose supporters share similar values. Although the article uses data from 2012, the successful application of this approach to the clustering of administrative-territorial units opens up the ways for such clustering on more recent data