ВАЛІДНІСТЬ ІНДИВІДУАЛЬНОГО ЗВАЖУВАННЯ ЗМІННИХ ПРИ ФОРМУВАННІ ІНДЕКСУ ЗАДОВОЛЕНОСТІ ДЕМОКРАТІЄЮ

Abstract

The common approach to measuring satisfaction with life or other complex phenomenon is to calculate a simple or weighted sum of satisfaction scores for its individual domains or aspects. There is a widespread perception among researchers that the validity of a composite score can be increased when domain satisfaction scores are weighted by the importance that respondents assign to the respective domains, but empirical studies do not confirm this assumption. The article is devoted to finding of conditions under which individual importance weighting of satisfaction with the state of democracy in Ukraine demonstrates higher validity for predicting overall satisfaction compared to unit-weighted and normatively weighted composites. In the present study, importance weighting was investigated on the data of sixth round of European Social Survey collected in Ukraine (N = 2178). It has been shown that the validity of an individual importance weighting of satisfaction with democracy depends on intra-individual variability of the aspects importance ratings. Using hierarchical linear regression, a significant (up to 11%) statistically significant increase in the percentage of explained variance of overall satisfaction with democracy was found when adding to the model an individual weighted indicator in a sample of respondents with differentiated perceptions of 14 aspects of democracy. The article also demonstrates an interesting trend: in the regression equations, individual importance weighting can play a role of a suppressor variable.