Generating dichotomous item scores with the four-parameter beta compound binomial model

Patrick Monahan, Won Chan Lee, Robert D. Ankenmann

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A Monte Carlo simulation technique for generating dichotomous item scores is presented that implements (a) a psychometric model with different explicit assumptions than traditional parametric item response theory (IRT) models, and (b) item characteristic curves without restrictive assumptions concerning mathematical form. The four-parameter beta compound-binomial (4PBCB) strong true score model (with two-term approximation to the compound binomial) is used to estimate and generate the true score distribution. The nonparametric item-true score step functions are estimated by classical item difficulties conditional on proportion-correct total score. The technique performed very well in replicating inter-item correlations, item statistics (point-biserial correlation coefficients and item proportion-correct difficulties), first four moments of total score distribution, and coefficient alpha of three real data sets consisting of educational achievement test scores. The technique replicated real data (including subsamples of differing proficiency) as well as the three-parameter logistic (3PL) IRT model (and much better than the 1PL model) and is therefore a promising alternative simulation technique. This 4PBCB technique may be particularly useful as a more neutral simulation procedure for comparing methods that use different IRT models.

Original languageEnglish
Pages (from-to)211-225
Number of pages15
JournalJournal of Educational Measurement
Volume44
Issue number3
DOIs
StatePublished - Sep 2007

Fingerprint

Statistical Models
Educational Status
model theory
Psychometrics
simulation
educational achievement test
psychometrics
logistics
statistics
Datasets

ASJC Scopus subject areas

  • Psychology(all)
  • Applied Psychology
  • Developmental and Educational Psychology
  • Psychology (miscellaneous)

Cite this

Generating dichotomous item scores with the four-parameter beta compound binomial model. / Monahan, Patrick; Lee, Won Chan; Ankenmann, Robert D.

In: Journal of Educational Measurement, Vol. 44, No. 3, 09.2007, p. 211-225.

Research output: Contribution to journalArticle

@article{2e6a28d732af4aaca5fe28950be591ce,
title = "Generating dichotomous item scores with the four-parameter beta compound binomial model",
abstract = "A Monte Carlo simulation technique for generating dichotomous item scores is presented that implements (a) a psychometric model with different explicit assumptions than traditional parametric item response theory (IRT) models, and (b) item characteristic curves without restrictive assumptions concerning mathematical form. The four-parameter beta compound-binomial (4PBCB) strong true score model (with two-term approximation to the compound binomial) is used to estimate and generate the true score distribution. The nonparametric item-true score step functions are estimated by classical item difficulties conditional on proportion-correct total score. The technique performed very well in replicating inter-item correlations, item statistics (point-biserial correlation coefficients and item proportion-correct difficulties), first four moments of total score distribution, and coefficient alpha of three real data sets consisting of educational achievement test scores. The technique replicated real data (including subsamples of differing proficiency) as well as the three-parameter logistic (3PL) IRT model (and much better than the 1PL model) and is therefore a promising alternative simulation technique. This 4PBCB technique may be particularly useful as a more neutral simulation procedure for comparing methods that use different IRT models.",
author = "Patrick Monahan and Lee, {Won Chan} and Ankenmann, {Robert D.}",
year = "2007",
month = "9",
doi = "10.1111/j.1745-3984.2007.00035.x",
language = "English",
volume = "44",
pages = "211--225",
journal = "Journal of Educational Measurement",
issn = "0022-0655",
publisher = "Wiley-Blackwell",
number = "3",

}

TY - JOUR

T1 - Generating dichotomous item scores with the four-parameter beta compound binomial model

AU - Monahan, Patrick

AU - Lee, Won Chan

AU - Ankenmann, Robert D.

PY - 2007/9

Y1 - 2007/9

N2 - A Monte Carlo simulation technique for generating dichotomous item scores is presented that implements (a) a psychometric model with different explicit assumptions than traditional parametric item response theory (IRT) models, and (b) item characteristic curves without restrictive assumptions concerning mathematical form. The four-parameter beta compound-binomial (4PBCB) strong true score model (with two-term approximation to the compound binomial) is used to estimate and generate the true score distribution. The nonparametric item-true score step functions are estimated by classical item difficulties conditional on proportion-correct total score. The technique performed very well in replicating inter-item correlations, item statistics (point-biserial correlation coefficients and item proportion-correct difficulties), first four moments of total score distribution, and coefficient alpha of three real data sets consisting of educational achievement test scores. The technique replicated real data (including subsamples of differing proficiency) as well as the three-parameter logistic (3PL) IRT model (and much better than the 1PL model) and is therefore a promising alternative simulation technique. This 4PBCB technique may be particularly useful as a more neutral simulation procedure for comparing methods that use different IRT models.

AB - A Monte Carlo simulation technique for generating dichotomous item scores is presented that implements (a) a psychometric model with different explicit assumptions than traditional parametric item response theory (IRT) models, and (b) item characteristic curves without restrictive assumptions concerning mathematical form. The four-parameter beta compound-binomial (4PBCB) strong true score model (with two-term approximation to the compound binomial) is used to estimate and generate the true score distribution. The nonparametric item-true score step functions are estimated by classical item difficulties conditional on proportion-correct total score. The technique performed very well in replicating inter-item correlations, item statistics (point-biserial correlation coefficients and item proportion-correct difficulties), first four moments of total score distribution, and coefficient alpha of three real data sets consisting of educational achievement test scores. The technique replicated real data (including subsamples of differing proficiency) as well as the three-parameter logistic (3PL) IRT model (and much better than the 1PL model) and is therefore a promising alternative simulation technique. This 4PBCB technique may be particularly useful as a more neutral simulation procedure for comparing methods that use different IRT models.

UR - http://www.scopus.com/inward/record.url?scp=34547971664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34547971664&partnerID=8YFLogxK

U2 - 10.1111/j.1745-3984.2007.00035.x

DO - 10.1111/j.1745-3984.2007.00035.x

M3 - Article

AN - SCOPUS:34547971664

VL - 44

SP - 211

EP - 225

JO - Journal of Educational Measurement

JF - Journal of Educational Measurement

SN - 0022-0655

IS - 3

ER -