Comparing simulated and theoretical sampling distributions of the U3 person-fit statistic

Wilco H.M. Emons, R.R. Meijer, Klaas Sijtsma

Research output: Contribution to journalArticleProfessional

16 Citations (Scopus)

Abstract

The accuracy with which the theoretical sampling distribution of van der Flier’s person-fit statistic U3 approaches the empirical U3 sampling distribution is affected by the item discrimination. A simulation study showed that for tests with a moderate or a strong mean item discrimination, the Type I error rates were either too high or too low to be used in practice. It was concluded that the use of standard normal deviates for the standardized version of the U3 statistic may be problematic. Nevertheless, the U3 statistic is suitable for evaluating the relative likelihood of item score patterns, for example, if one wishes to select a fixed percentage of the most improbable patterns.
Original languageUndefined
Pages (from-to)88-108
Number of pages20
JournalApplied psychological measurement
Volume26
Issue number1
DOIs
Publication statusPublished - 2002

Keywords

  • METIS-209479
  • IR-60184

Cite this

Emons, Wilco H.M. ; Meijer, R.R. ; Sijtsma, Klaas. / Comparing simulated and theoretical sampling distributions of the U3 person-fit statistic. In: Applied psychological measurement. 2002 ; Vol. 26, No. 1. pp. 88-108.
@article{27c0b6ca4f4746f383123fa662fb099b,
title = "Comparing simulated and theoretical sampling distributions of the U3 person-fit statistic",
abstract = "The accuracy with which the theoretical sampling distribution of van der Flier’s person-fit statistic U3 approaches the empirical U3 sampling distribution is affected by the item discrimination. A simulation study showed that for tests with a moderate or a strong mean item discrimination, the Type I error rates were either too high or too low to be used in practice. It was concluded that the use of standard normal deviates for the standardized version of the U3 statistic may be problematic. Nevertheless, the U3 statistic is suitable for evaluating the relative likelihood of item score patterns, for example, if one wishes to select a fixed percentage of the most improbable patterns.",
keywords = "METIS-209479, IR-60184",
author = "Emons, {Wilco H.M.} and R.R. Meijer and Klaas Sijtsma",
year = "2002",
doi = "10.1177/0146621602026001006",
language = "Undefined",
volume = "26",
pages = "88--108",
journal = "Applied psychological measurement",
issn = "0146-6216",
publisher = "SAGE Publications",
number = "1",

}

Comparing simulated and theoretical sampling distributions of the U3 person-fit statistic. / Emons, Wilco H.M.; Meijer, R.R.; Sijtsma, Klaas.

In: Applied psychological measurement, Vol. 26, No. 1, 2002, p. 88-108.

Research output: Contribution to journalArticleProfessional

TY - JOUR

T1 - Comparing simulated and theoretical sampling distributions of the U3 person-fit statistic

AU - Emons, Wilco H.M.

AU - Meijer, R.R.

AU - Sijtsma, Klaas

PY - 2002

Y1 - 2002

N2 - The accuracy with which the theoretical sampling distribution of van der Flier’s person-fit statistic U3 approaches the empirical U3 sampling distribution is affected by the item discrimination. A simulation study showed that for tests with a moderate or a strong mean item discrimination, the Type I error rates were either too high or too low to be used in practice. It was concluded that the use of standard normal deviates for the standardized version of the U3 statistic may be problematic. Nevertheless, the U3 statistic is suitable for evaluating the relative likelihood of item score patterns, for example, if one wishes to select a fixed percentage of the most improbable patterns.

AB - The accuracy with which the theoretical sampling distribution of van der Flier’s person-fit statistic U3 approaches the empirical U3 sampling distribution is affected by the item discrimination. A simulation study showed that for tests with a moderate or a strong mean item discrimination, the Type I error rates were either too high or too low to be used in practice. It was concluded that the use of standard normal deviates for the standardized version of the U3 statistic may be problematic. Nevertheless, the U3 statistic is suitable for evaluating the relative likelihood of item score patterns, for example, if one wishes to select a fixed percentage of the most improbable patterns.

KW - METIS-209479

KW - IR-60184

U2 - 10.1177/0146621602026001006

DO - 10.1177/0146621602026001006

M3 - Article

VL - 26

SP - 88

EP - 108

JO - Applied psychological measurement

JF - Applied psychological measurement

SN - 0146-6216

IS - 1

ER -