Using response times to detect aberrant responses in computerized adaptive testing

Willem J. van der Linden (Corresponding Author), Edith van Krimpen-Stoop

Research output: Contribution to journalArticleAcademic

44 Citations (Scopus)

Abstract

A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.
Original languageEnglish
Pages (from-to)251-266
JournalPsychometrika
Volume68
Issue number2
DOIs
Publication statusPublished - 2003

Fingerprint

Adaptive Testing
Response Time
Reaction Time
Testing
Adaptive Test
False Alarm Rate
Guidelines

Keywords

  • Aberrant response patterns
  • Computerized Adaptive Testing
  • Posterior predictive checks
  • Person misfit
  • Response times
  • Residual analysis

Cite this

@article{11703fcb18ce46b1a4389fdb145bfe00,
title = "Using response times to detect aberrant responses in computerized adaptive testing",
abstract = "A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.",
keywords = "Aberrant response patterns, Computerized Adaptive Testing, Posterior predictive checks, Person misfit, Response times, Residual analysis",
author = "{van der Linden}, {Willem J.} and {van Krimpen-Stoop}, Edith",
year = "2003",
doi = "10.1007/BF02294800",
language = "English",
volume = "68",
pages = "251--266",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer",
number = "2",

}

Using response times to detect aberrant responses in computerized adaptive testing. / van der Linden, Willem J. (Corresponding Author); van Krimpen-Stoop, Edith.

In: Psychometrika, Vol. 68, No. 2, 2003, p. 251-266.

Research output: Contribution to journalArticleAcademic

TY - JOUR

T1 - Using response times to detect aberrant responses in computerized adaptive testing

AU - van der Linden, Willem J.

AU - van Krimpen-Stoop, Edith

PY - 2003

Y1 - 2003

N2 - A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.

AB - A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.

KW - Aberrant response patterns

KW - Computerized Adaptive Testing

KW - Posterior predictive checks

KW - Person misfit

KW - Response times

KW - Residual analysis

U2 - 10.1007/BF02294800

DO - 10.1007/BF02294800

M3 - Article

VL - 68

SP - 251

EP - 266

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 2

ER -