Detecting answer copying using the kappa statistic

Leonardo Sotaridona, Willem J. van der Linden, R.R. Meijer

Research output: Contribution to journalArticleProfessional

15 Citations (Scopus)

Abstract

A statistical test for detecting answer copying on multiple-choice tests based on Cohen's kappa is proposed. The test is free of any assumptions on the response processes of the examinees suspected of copying and having served as the source, except for the usual assumption that these processes are probabilistic. Because the asymptotic null and alternative distributions of the kappa statistic are derived under the assumption of common marginal probabilities for all items, a recoding of the item alternatives is proposed to approximate this case. The results from a simulation study in this article show that under this recoding, the test approximates its nominal Type I error rates and has promising power functions.
Original languageUndefined
Pages (from-to)412-431
Number of pages19
JournalApplied psychological measurement
Volume30
Issue number5
DOIs
Publication statusPublished - 2006

Keywords

  • METIS-235080
  • Type I error
  • Kappa statistic
  • power analysis
  • Answer Copying
  • IR-60069

Cite this

Sotaridona, Leonardo ; van der Linden, Willem J. ; Meijer, R.R. / Detecting answer copying using the kappa statistic. In: Applied psychological measurement. 2006 ; Vol. 30, No. 5. pp. 412-431.
@article{12da1582a2b7480ea521424bdc2bdc69,
title = "Detecting answer copying using the kappa statistic",
abstract = "A statistical test for detecting answer copying on multiple-choice tests based on Cohen's kappa is proposed. The test is free of any assumptions on the response processes of the examinees suspected of copying and having served as the source, except for the usual assumption that these processes are probabilistic. Because the asymptotic null and alternative distributions of the kappa statistic are derived under the assumption of common marginal probabilities for all items, a recoding of the item alternatives is proposed to approximate this case. The results from a simulation study in this article show that under this recoding, the test approximates its nominal Type I error rates and has promising power functions.",
keywords = "METIS-235080, Type I error, Kappa statistic, power analysis, Answer Copying, IR-60069",
author = "Leonardo Sotaridona and {van der Linden}, {Willem J.} and R.R. Meijer",
year = "2006",
doi = "10.1177/0146621606288891",
language = "Undefined",
volume = "30",
pages = "412--431",
journal = "Applied psychological measurement",
issn = "0146-6216",
publisher = "SAGE Publications",
number = "5",

}

Detecting answer copying using the kappa statistic. / Sotaridona, Leonardo; van der Linden, Willem J.; Meijer, R.R.

In: Applied psychological measurement, Vol. 30, No. 5, 2006, p. 412-431.

Research output: Contribution to journalArticleProfessional

TY - JOUR

T1 - Detecting answer copying using the kappa statistic

AU - Sotaridona, Leonardo

AU - van der Linden, Willem J.

AU - Meijer, R.R.

PY - 2006

Y1 - 2006

N2 - A statistical test for detecting answer copying on multiple-choice tests based on Cohen's kappa is proposed. The test is free of any assumptions on the response processes of the examinees suspected of copying and having served as the source, except for the usual assumption that these processes are probabilistic. Because the asymptotic null and alternative distributions of the kappa statistic are derived under the assumption of common marginal probabilities for all items, a recoding of the item alternatives is proposed to approximate this case. The results from a simulation study in this article show that under this recoding, the test approximates its nominal Type I error rates and has promising power functions.

AB - A statistical test for detecting answer copying on multiple-choice tests based on Cohen's kappa is proposed. The test is free of any assumptions on the response processes of the examinees suspected of copying and having served as the source, except for the usual assumption that these processes are probabilistic. Because the asymptotic null and alternative distributions of the kappa statistic are derived under the assumption of common marginal probabilities for all items, a recoding of the item alternatives is proposed to approximate this case. The results from a simulation study in this article show that under this recoding, the test approximates its nominal Type I error rates and has promising power functions.

KW - METIS-235080

KW - Type I error

KW - Kappa statistic

KW - power analysis

KW - Answer Copying

KW - IR-60069

U2 - 10.1177/0146621606288891

DO - 10.1177/0146621606288891

M3 - Article

VL - 30

SP - 412

EP - 431

JO - Applied psychological measurement

JF - Applied psychological measurement

SN - 0146-6216

IS - 5

ER -