STATISTICAL-METHODS FOR COMPARING REGRESSION-COEFFICIENTS BETWEEN MODELS

Statistical methods are developed for comparing regression coefficient s between models in the setting where one of the models is nested in t he other. Comparisons of this kind are of interest whenever two explan ations of a given phenomenon are specified as linear models. In this c ase, researchers should ask whether the coefficients associated with a given set of predictors change in a significant way when other predic tors or covariates are added as controls. Simple calculations based on quantities provided by routines for regression analysis can be used t o obtain the standard errors and other statistics that are required. R esults are also given for the class of generalized linear models (e.g. , logistic regression, log-linear models, etc.). We recommend fundamen tal change in strategies for model comparisons in social research as w ell as modifications in the presentation of results from regression or regression-type models.