Person-Fit Statistics for Joint Models for Accuracy and Speed

Jean Paul Fox, Sukaesi Marianti

Research output: Contribution to journalArticleAcademicpeer-review

10 Citations (Scopus)
1 Downloads (Pure)

Abstract

Response accuracy and response time data can be analyzed with a joint model to measure ability and speed of working, while accounting for relationships between item and person characteristics. In this study, person-fit statistics are proposed for joint models to detect aberrant response accuracy and/or response time patterns. The person-fit tests take the correlation between ability and speed into account, as well as the correlation between item characteristics. They are posited as Bayesian significance tests, which have the advantage that the extremeness of a test statistic value is quantified by a posterior probability. The person-fit tests can be computed as by-products of a Markov chain Monte Carlo algorithm. Simulation studies were conducted in order to evaluate their performance. For all person-fit tests, the simulation studies showed good detection rates in identifying aberrant patterns. A real data example is given to illustrate the person-fit statistics for the evaluation of the joint model.

Original languageEnglish
Pages (from-to)243-262
Number of pages20
JournalJournal of educational measurement
Volume54
Issue number2
DOIs
Publication statusPublished - 1 Jun 2017

Fingerprint

Joints
statistics
human being
Aptitude
Reaction Time
significance test
Markov Chains
simulation
ability
evaluation
performance
Values
time

Cite this

@article{16f0e6cf3120434fae2e21374268469a,
title = "Person-Fit Statistics for Joint Models for Accuracy and Speed",
abstract = "Response accuracy and response time data can be analyzed with a joint model to measure ability and speed of working, while accounting for relationships between item and person characteristics. In this study, person-fit statistics are proposed for joint models to detect aberrant response accuracy and/or response time patterns. The person-fit tests take the correlation between ability and speed into account, as well as the correlation between item characteristics. They are posited as Bayesian significance tests, which have the advantage that the extremeness of a test statistic value is quantified by a posterior probability. The person-fit tests can be computed as by-products of a Markov chain Monte Carlo algorithm. Simulation studies were conducted in order to evaluate their performance. For all person-fit tests, the simulation studies showed good detection rates in identifying aberrant patterns. A real data example is given to illustrate the person-fit statistics for the evaluation of the joint model.",
author = "Fox, {Jean Paul} and Sukaesi Marianti",
year = "2017",
month = "6",
day = "1",
doi = "10.1111/jedm.12143",
language = "English",
volume = "54",
pages = "243--262",
journal = "Journal of educational measurement",
issn = "0022-0655",
publisher = "Wiley-Blackwell",
number = "2",

}

Person-Fit Statistics for Joint Models for Accuracy and Speed. / Fox, Jean Paul; Marianti, Sukaesi.

In: Journal of educational measurement, Vol. 54, No. 2, 01.06.2017, p. 243-262.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Person-Fit Statistics for Joint Models for Accuracy and Speed

AU - Fox, Jean Paul

AU - Marianti, Sukaesi

PY - 2017/6/1

Y1 - 2017/6/1

N2 - Response accuracy and response time data can be analyzed with a joint model to measure ability and speed of working, while accounting for relationships between item and person characteristics. In this study, person-fit statistics are proposed for joint models to detect aberrant response accuracy and/or response time patterns. The person-fit tests take the correlation between ability and speed into account, as well as the correlation between item characteristics. They are posited as Bayesian significance tests, which have the advantage that the extremeness of a test statistic value is quantified by a posterior probability. The person-fit tests can be computed as by-products of a Markov chain Monte Carlo algorithm. Simulation studies were conducted in order to evaluate their performance. For all person-fit tests, the simulation studies showed good detection rates in identifying aberrant patterns. A real data example is given to illustrate the person-fit statistics for the evaluation of the joint model.

AB - Response accuracy and response time data can be analyzed with a joint model to measure ability and speed of working, while accounting for relationships between item and person characteristics. In this study, person-fit statistics are proposed for joint models to detect aberrant response accuracy and/or response time patterns. The person-fit tests take the correlation between ability and speed into account, as well as the correlation between item characteristics. They are posited as Bayesian significance tests, which have the advantage that the extremeness of a test statistic value is quantified by a posterior probability. The person-fit tests can be computed as by-products of a Markov chain Monte Carlo algorithm. Simulation studies were conducted in order to evaluate their performance. For all person-fit tests, the simulation studies showed good detection rates in identifying aberrant patterns. A real data example is given to illustrate the person-fit statistics for the evaluation of the joint model.

UR - http://www.scopus.com/inward/record.url?scp=85020222032&partnerID=8YFLogxK

U2 - 10.1111/jedm.12143

DO - 10.1111/jedm.12143

M3 - Article

VL - 54

SP - 243

EP - 262

JO - Journal of educational measurement

JF - Journal of educational measurement

SN - 0022-0655

IS - 2

ER -