Comparing Relations among Psychological Constructs Across Groups
Many important research questions in psychology seek to uncover relations among psychological constructs that are not directly observable. Rather, these constructs are known to be latent and are measured indirectly via questionnaire items. Structural equation modelling (SEM) is the state-of-the-art technique to test the relations among these psychological constructs and questionnaire items. When comparing the relations between the latent constructs across many groups (e.g., countries in a cross-national study), they likely differ among the groups (e.g., they may be culturally dependent). However, it is equally possible that some of the groups share the same relations, such that ‘clusters’ of groups emerge with respect to these relations (e.g., clusters of countries with a similar culture). This PhD-project aims to develop a novel mixture SEM method for finding such clusters when the measurement model is not known a priori (i.e., it is exploratory rather than confirmatory). To do so, differences in how the psychological constructs are measured by the questionnaire items (i.e., the measurement model) should be distinguished from the differences in the psychological relations of interest (i.e., the structural model), while at the same time capturing clusters of groups with equivalent relations among the psychological constructs.
Dr. K. De Roover
Prof. dr. Y. Rosseel
1 October 2023 – 1 October 2027