Statistical properties of the K-index for detecting answer copying

Leonardo Sotaridona, R.R. Meijer

Research output: Contribution to journalArticleAcademic

25 Citations (Scopus)

Abstract

We investigated the statistical properties of the K-index (Holland, 1996) that can be used to detect copying behavior on a test. A simulation study was conducted to investigate the applicability of the K-index for small, medium, and large datasets. Furthermore, the Type I error rate and the detection rate of this index were compared with the copying index, o (Wollack, 1997). Several approximations were used to calculate the K-index. Results showed that all approximations were able to hold the Type I error rates below the nominal level. Results further showed that using o resulted in higher detection rates than the K-indices for small and medium sample sizes (100 and 500 simulees).
Original languageEnglish
Pages (from-to)115-132
Number of pages17
JournalJournal of educational measurement
Volume39
Issue number2
DOIs
Publication statusPublished - 2002

Fingerprint

Sample Size
Datasets
simulation

Keywords

  • IR-58268
  • METIS-210624

Cite this

Sotaridona, Leonardo ; Meijer, R.R. / Statistical properties of the K-index for detecting answer copying. In: Journal of educational measurement. 2002 ; Vol. 39, No. 2. pp. 115-132.
@article{21036e906eea4017b15e989e8d78aa2f,
title = "Statistical properties of the K-index for detecting answer copying",
abstract = "We investigated the statistical properties of the K-index (Holland, 1996) that can be used to detect copying behavior on a test. A simulation study was conducted to investigate the applicability of the K-index for small, medium, and large datasets. Furthermore, the Type I error rate and the detection rate of this index were compared with the copying index, o (Wollack, 1997). Several approximations were used to calculate the K-index. Results showed that all approximations were able to hold the Type I error rates below the nominal level. Results further showed that using o resulted in higher detection rates than the K-indices for small and medium sample sizes (100 and 500 simulees).",
keywords = "IR-58268, METIS-210624",
author = "Leonardo Sotaridona and R.R. Meijer",
year = "2002",
doi = "10.1111/j.1745-3984.2002.tb01138.x",
language = "English",
volume = "39",
pages = "115--132",
journal = "Journal of educational measurement",
issn = "0022-0655",
publisher = "Wiley-Blackwell",
number = "2",

}

Statistical properties of the K-index for detecting answer copying. / Sotaridona, Leonardo; Meijer, R.R.

In: Journal of educational measurement, Vol. 39, No. 2, 2002, p. 115-132.

Research output: Contribution to journalArticleAcademic

TY - JOUR

T1 - Statistical properties of the K-index for detecting answer copying

AU - Sotaridona, Leonardo

AU - Meijer, R.R.

PY - 2002

Y1 - 2002

N2 - We investigated the statistical properties of the K-index (Holland, 1996) that can be used to detect copying behavior on a test. A simulation study was conducted to investigate the applicability of the K-index for small, medium, and large datasets. Furthermore, the Type I error rate and the detection rate of this index were compared with the copying index, o (Wollack, 1997). Several approximations were used to calculate the K-index. Results showed that all approximations were able to hold the Type I error rates below the nominal level. Results further showed that using o resulted in higher detection rates than the K-indices for small and medium sample sizes (100 and 500 simulees).

AB - We investigated the statistical properties of the K-index (Holland, 1996) that can be used to detect copying behavior on a test. A simulation study was conducted to investigate the applicability of the K-index for small, medium, and large datasets. Furthermore, the Type I error rate and the detection rate of this index were compared with the copying index, o (Wollack, 1997). Several approximations were used to calculate the K-index. Results showed that all approximations were able to hold the Type I error rates below the nominal level. Results further showed that using o resulted in higher detection rates than the K-indices for small and medium sample sizes (100 and 500 simulees).

KW - IR-58268

KW - METIS-210624

U2 - 10.1111/j.1745-3984.2002.tb01138.x

DO - 10.1111/j.1745-3984.2002.tb01138.x

M3 - Article

VL - 39

SP - 115

EP - 132

JO - Journal of educational measurement

JF - Journal of educational measurement

SN - 0022-0655

IS - 2

ER -