Abstract
This article illustrates the role of group-agreement indices in the process of multilevel construct validation. A two-step process is outlined that includes multitrait-multimethod confirmatory factor analysis (CFA) and the use of a dispersion model to examine correlates of consensus of multilevel constructs. The role of dispersion models for providing discriminant validity evidence for similar constructs, as well as providing insight as to why some teams share perceptions more strongly than others, is illustrated using three efficacy-related constructs.
Get full access to this article
View all access options for this article.
