Implied Probabilities of Polytomous Response Functions for Model-Based Prediction and Comparison

• In Event: Evaluating Fit

Abstract

Polytomous response models are typically motivated by assumptions about specific dichotomizations. We build on this by considering all possible dichotomizations. We enumerate such dichotomizations and show that many of these dichotomizations have dramatically different forms when computed for different models. These differences can be used to evaluate model fit.

Authors

• Ben Domingue, STANFORD UNIVERSITY

• Klint Kanopka, NEW YORK UNIVERSITY

• Esther Ulitzsch, University of Oslo

• Lijin Zhang, Stanford Graduate School of Education