English title: A Cross-Country and Cross-Time Comparison of the Human Values Measurements with the Second Round of the European Social Survey
Author(s): Eldad Davidov -
Type: Journal article
The study reported in this paper assesses the fit of a 21-item instrument measuring values in the second round of the European Social Survey (ESS) to the theory of 10 basic values on which it was based (Schwartz 1992). In particular, the measurement invariance of this instrument for studying value priorities across nations and over time was investigated. In the first part of the study, using multi-group confirmatory factor analysis (MGCFA) of data from the second ESS round, configural, metric, and scalar invariance of the values are assessed across 25 countries. Metric invariance is a necessary condition to insure equivalence of the meaning of factors and a precondition for comparing values’ correlates. Scalar invariance is a precondition for comparing value means. The MGCFA did not support configural and metric invariance across 25 countries. After reducing the number of countries to 14, the MGCFA supported metric invariance of a model with seven distinct values, the same values identified with data from the first ESS round. These value measurements may now be used by researchers to study relationships among values, attitudes, behavior, and sociodemographic characteristics across the 14 nations. Comparing national value means may be possible only across a smaller set of countries where scalar invariance holds. In the second part of the study, metric and scalar invariance were established between the first and the second rounds of the ESS in each of 19 countries separately. Value means may be compared for each of the countries between the first and second ESS rounds (2002-2003 and 2004-2005, respectively).
From page no: 33
To page no: 46
Journal: Survey Research Methods