To study the composition of political cultures, it is necessary to classify citizens according to a theoretical typology of political subcultures. Different methods of classification have been used for confronting this challenge, but the choice of method is rarely discussed in any detail because most studies apply a method without considering implications or possible alternatives. This is unfortunate because the choice of method has important consequences for the ensuing results. With this article, we aim to determine the implications of different methods for classification and hereby call attention to the importance of this choice for comparative research on political culture. We compare three commonly used methods of classification: critical thresholds, factor analysis and cluster analysis. These methods are used for classifying respondents from the 2008 European Social Survey according to a typology of political subcultures. Based on empirical analyses, we conclude that: (a) the choice of method of classification affects the outcome of the analysis, (b) cluster analysis and factor analysis may result in classifications that do not adequately reflect the theoretical typology, and (c) cluster analysis and factor analysis provide classifications that differ depending on analytical level. While the results do not show that either method is inherently superior, they clearly demonstrate that the choice of method should be recognized as a critical part of the research process.
SAMHÄLLSVETENSKAP -- Statsvetenskap (hsv//swe)
SOCIAL SCIENCES -- Political Science (hsv//eng)
Methods of classification; Political culture; Cluster analysis; Factor analysis; Critical thresholds; Comparative survey analysis