Status Inconsistency: A Body of Negative Evidence or a Statistical Artifact?

Publisher Website
Google Scholar
Abstract
The disappearance of status inconsistency effects after controlling for rank dimensions may be artifactual. Thus, the failure of inconsistency models to explain a greater proportion of the variance in measures of stress does not necessarily demonstrate the absence of inconsistency effects. One may also obtain significant status inconsistency effects even if none actually exist. Inconsistency models violate the assumptions of multiple regression in such a way that the inconsistency-stress hypothesis has yet to receive a fair test. The nature of the violations is discussed and their implications demonstrated with simulated data. None of four other models tested proves adequate for all types of status inconsistency hypotheses, but the multiple classification analysis specification of Hope's diamond model of status inconsistency generally yields correct results. Either this model or cross-tabular alternatives such as those proposed by Goodman should be used in future status inconsistency research.

This publication has 0 references indexed in Scilit: