Diagnostic Assessment: Methods for the Reliable Measurement of Multidimensional Abilities

Jonathan Templin

doi:10.4324/9781315871493-16

Back

Diagnostic Assessment: Methods for the Reliable Measurement of Multidimensional Abilities

Book chapter

Open access

Diagnostic Assessment: Methods for the Reliable Measurement of Multidimensional Abilities

Jonathan Templin

Technology and Testing, pp.285-304

Routledge

2015

DOI: 10.4324/9781315871493-16

Files and links (1)

url

https://doi.org/10.4324/9781315871493-16View

Published (Version of record) Open Access

Abstract

This chapter describes the current state of diagnostic models in educational measurement. Educational measurement is in a constant cycle of rediscovering, refining, and advancing theories to meet the challenges of the current times. The diagnostic classification models (DCMs) are very similar to other latent-variable models that are currently used by specialists in educational and psychological measurement. The modeling frameworks that unify a large variety of modern DCMs are typically labelled generalized linear mixed models (GLMMs) and nonlinear mixed models (NLMM). The log-linear cognitive diagnosis model (LCDM) subsumes many core DCMs in ways that allow for a greater understanding of the response processes that generate the item responses. The current state of the art in the field of DCMs should predominantly be comforting for researchers, because the field of educational and psychological measurement has amassed a rich resource library and practical expertise for tackling the challenges in other areas such as M-IRT and M-CFA specifically as well as GLMMs more generally.

DINA Model

Attribute Profiles

USMLE Step

General Diagnostic Model

GLMM Framework

Latent Variables

IRT Model

Continuous Latent Variables

Discrete Latent Variables

Automated Test Assembly

Item Parameters

Item Response Probability

Latent Class Models

Item Response

Automated Essay Scoring

Predicted Log Odds

Linear Predictor

Categorical Latent Variables

Latent Variable Models

DCMs

Marginal Maximum Likelihood Estimators

Multivariate Bernoulli Distribution

Latent Predictor Variables

Shadow Test

Details

Title: Subtitle: Diagnostic Assessment: Methods for the Reliable Measurement of Multidimensional Abilities
Creators: Jonathan Templin - University of Iowa, Psychological and Quantitative Foundations
Resource Type: Book chapter
Publication Details: Technology and Testing, pp.285-304
Publisher: Routledge; New York
DOI: 10.4324/9781315871493-16
Alternative title: Diagnostic Assessment
Language: English
Date published: 2015
Academic Unit: Psychological and Quantitative Foundations
Record Identifier: 9984380160102771

Metrics

24 Record Views