Key points are not available for this paper at this time.
The concomitant proliferation of causal modeling and hypotheses of multiplicative effects has brought about a tremendous need for procedures that allow the testing of moderated structural equation models (MSEMs). The seminal work of Kenny and Judd and Hayduk has been drawn on by several authors in the past 10 years, thus producing procedures that allow for such tests. Yet, utilization of MSEMs in empirical research has been quite rare. The purposes of this article are twofold. First, the authors discuss general issues with respect to multivariate normality, indicators of latent products, the nature of latent products, and identification problems in MSEM. Second, they review and illustrate techniques that are available for the testing of interaction effects in structural equation models.
Cortina et al. (Mon,) studied this question.