On the Use of Mixed -Effects Models for the Analysis of Probability Judgments

In this dissertation I propose a general statistical modeling framework for the purpose of making inferences concerning the external correspondence (i.e., calibration and discrimination) of probability judgments. The statistical model is based on a new stochastic judgment model which is expressed as a mixed-effects ordinal probit regression model. This model can be used to derive several new model-based measures of external correspondence as well as establishing a means of deriving the sampling/posterior distributions of such measures. Issues of model specification and inference for applied research are discussed in detail. Several detailed examples are given.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Psychology

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Johnson, Timothy Robin

Contributors dc:contributor

• Budescu, David V.

• Statistics

Language dc:language

eng

Identifier

(MiAaPQ)AAI3030442

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/82016