Bringing values back in the adequacy of the European Social Survey to measure values in 20 countries

Abstract

Values are prominent in public discourse today. Theorists have long considered values central to understanding attitudes and behavior. The Schwartz (1992) theory of basic human values has promoted a revival of empirical research on values. The semi-annual European Social Survey (ESS) includes a new 21-item instrument to measure the importance of the ten basic values of the theory. Representative national samples in 20 countries responded to the instrument in 2002-3. We briefly describe the theory and the ESS instrument and assess its adequacy for measuring values across countries. Using multiple group confirmatory factor analyses, augmented with mean-structure information, we assess the configural and measurement (metric) invariance of the values—necessary conditions for equivalence of the meaning of constructs, and scalar invariance—a precondition for comparing value means across countries. Only if such equivalence is established can researchers make meaningful and clearly interpretable cross- national comparisons of value priorities and their correlates. The ESS values scale demonstrates configural and metric invariance, allowing researchers to use it to study relationships among values, attitudes, behavior and socio-demographic characteristics across countries. Comparing the mean importance of values across countries is possible only for subsets of countries where scalar invariance holds.