Search

This paper discusses two basic issues related to the assessment of the goodness-of-fit of sociological models to behavioral data: How simplifying assumptions contained in models should be evaluated and whether analysis of data should be carried out at the individual group or aggregate level. The discussion is conducted in the context of the fit of a model of participation rates by Fisek, Berger, and Norman (1991) to data collected by Smith-Lovin, Skvoretz, and Hudson (1986) in a Bales-type situation. The conclusions are that simplifying assumptions should be evaluated from the point of view of robustness of model to violations of such assumptions, and that the assessment of the adequacy of minimal actor models should be in terms of aggregated data. The fit of the model and data under consideration is assessed to be adequate. [ABSTRACT FROM AUTHOR]