Applying the Estimation Using Pooled Data Approach to the Multitrait-Multimethod Experiments of the European Social Survey (Rounds 1 to 7)

Abstract

ABSTRACT A common way of estimating measurement quality is the split-ballot multitrait-multimethod (SB-MTMM) approach. However, it often leads to non-convergence or improper solutions, especially when using a 2-group design. Thus, the “Estimation Using Pooled Data” (EUPD) has been proposed to reduce these problems when cross-national data are available. To evaluate the performance of the EUPD procedure, we apply it to 30 SB-MTMM experiments conducted in the European Social Survey in a total of 34 countries. We find that the EUPD approach decreases the number of non-convergence and improper solutions compared to expected outcomes based on previous findings using a country-by-country approach. However, especially the problems of improper solutions are still not negligible.