Bayesian Psychometric Modeling

Preface

Context and Aims

Psychometrics concerns the principles and practices of reasoning in assessment. As a discipline, it came of age at a time marked by the dominance of frequentist approaches to probability, statistical modeling, and inference. Despite this backdrop, Bayesian methods have enjoyed widespread acceptance as solutions to certain problems in psychometrics. Nevertheless, as of this writing, conventional psychometric practice—fitting models, estimating parameters, evaluating model-data fit, using results to make or equate tests, evaluate fairness, and so on—employs mainly frequentist approaches. The vast majority of psychometric texts couch most if not all of their presentations in a frequentist paradigm (Bartholomew, Knott, ...