Testing measurement invariance in the International Social Survey Program Health 2011 – the mental well-being scale

Purpose In the present contribution we address the measurement invariance of a new mental well-being scale of three items that was applied in the International Social Survey Program (ISSP) Health 2011 module. Our aim is to establish if and for how many countries (partial) scalar invariance is accepted and meaningful comparisons between countries can be made. Methods We used the data release version 2.0.0 that was made public on the 27th August 2013 and that covers 29 countries. We employed multi-group confirmatory factor analysis (MGCFA) and judged our models using four goodness of fit indices: Chi-Square index, the Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI), and the Root Mean Squared Error of Approximation (RMSEA). Results We found that the mental well-being scale is partial scalar invariant for 28 out of the 29 countries in the sample. Conclusion Our results support the idea of measurement invariance of the ISSP Health 2011 mental well-being scale. However, our analyses showed an unusual ranking of the countries that was not associated with rankings based on other mental well-being measures. It is not clear what aspects of mental well-being are reflected by the ISSP Health 2011 mental well-being scale.