Many models in social psychology, which have been developed to explain behavior, postulate interaction effects between explanatory latent variables. In the last years, there have been many new developments for estimating interactions between latent variables in structural equation modeling. However, there have been very few applications with real data from theory-driven studies. This paper provides an empirical test with real data from an ongoing research project about travel mode choice in Frankfurt, using the theory of planned behavior. We apply three statistical approaches for the estimation of interaction effects between the latent variables perceived behavioral control (PBC) and intention for predicting travel mode choice (behavior): latent variable scores, maximum likelihood and robust maximum likelihood. We compare the strengths and weaknesses of the approaches from an applied point of view. In a metaanalytic review we summarize the results of 14 articles, which estimated the interaction between intention and PBC for predicting behavior, and discuss the problems associated with such a meta-analysis.
Stored only when both papers are in the corpus — 51 citations exist upstream.
Imported from OpenAlex (CC0) — nothing here is generated.