Studying value heterogenity in Europe by means of multi-group latent class analysis
Vladimir Magun, Maksim Rudnev
Building: Law Building
Room: Breakout 2 - Law Building, Room 026
Date: 2012-07-10 01:30 PM – 03:00 PM
Last modified: 2011-12-22
Abstract
Cross-national value differences have been studied all over the world mostly on the country level. Heterogeneity of within-country value differences have been ignored. Recent developments in methodology allowed researchers to obtain universal types of respondents, who has the similar pattern of value preferences and can be compared across countries. Current study uses multi-group latent class analysis to explore universal value types measured with Sh. Schwartz instrument in European Social Survey and with R. Inglehart instruments in European Values Study. Raw items despite the value indices were used to produce classes. First, we obtained heterogeneous 3- and 6- class solutions based on Schwartz data and 3-class solution based on Inglehart's values. Every European country has a share in each class. This eneables researchers to interpret the national differences in a very new way - not like traditional and rather vague \"level of values\" statements, but in terms of certain shares of people in each country tending to share the same values. Then we turned to homogeneous solutions and found that there are no appropriate solution if using Inglehart's instument, so we obtained partial equivalence of value classes (Kankaras, Moors, Vermunt, 2011). A value fractionalization index was calculated as analogue of ethnolinguistic fractionalization (Alesina et al., 2001). Surprisingly, the rank of country's value fractionalization doesn't depend on the number of classes, that is it's free of specific LCA solution. In Schwartz terms, the most homogeneous counties in Europe are less developed - Romania, Turkey, Greece, Russia, and the least homogeneous are more developed ones, like Germany, Finland, Switzerland. Differences in heterogeneity in Inglehart's terms are not correlated with economic development. Different interpretations of the results are discussed.