You are invited to join the upcoming Mellenbergh lecture on Tuesday, August 26th from 11:00 - 12:00!
Wes Bonifay (University of Missouri) will talk about corroborative factor analysis.
Speaker: Wes Bonifay (University of Missouri)
Title : Corroborative Factor Analysis

Abstract: Confirmatory factor analysis (CFA; Jöreskog, 1969) was originally intended as a method by which researchers can test an a priori theory about the latent variable structure underlying a set of indicators. In typical CFA applications, “confirmation” of such a test primarily entails inspecting the goodness of fit between the model and the observed data. In this talk, I argue that (a) goodness-of-fit tests are too easy to pass (Roberts & Pashler, 2000; 2002); (b) riskier tests are needed (Meehl, 1978; Mayo, 1991); and (c) confirmation is the wrong goal (Meehl, 1990). To bring factor analysis into closer alignment with these important philosophical-scientific principles, I introduce corroborative factor analysis: an inferential framework for (dis)corroborating factor analytic theories via constrained maximum likelihood estimation and/or Bayesian prior predictive model checking.

Where and when? REC A1.03 at 11:00