A flexible latent class approach to estimating test-score reliability

Daniel W. van der Palm, L. Andries van der Ark, K. Sijtsma

Research output: Contribution to journalArticleScientificpeer-review

Abstract

The latent class reliability coefficient (LCRC) is improved by using the divisive latent class model instead of the unrestricted latent class model. This results in the divisive latent class reliability coefficient (DLCRC), which unlike LCRC avoids making subjective decisions about the best solution and thus avoids judgment error. A computational study using large numbers of items shows that DLCRC also is faster than LCRC and fast enough for practical purposes. Speed and objectivity render DLCRC superior to LCRC. A decisive feature of DLCRC is that it aims at closely approximating the multivariate distribution of item scores, which might render the method suited when test data are multidimensional. A simulation study focusing on multidimensionality shows that DLCRC in general has little bias relative to the true reliability and is relatively accurate compared to LCRC and classical lower bound methods coefficients α and λ2 and the greatest lower bound.
Original languageEnglish
Pages (from-to)339-357
JournalJournal of Educational Measurement
Volume51
Issue number4
DOIs
Publication statusPublished - 2014

Cite this

van der Palm, Daniel W. ; van der Ark, L. Andries ; Sijtsma, K. / A flexible latent class approach to estimating test-score reliability. In: Journal of Educational Measurement. 2014 ; Vol. 51, No. 4. pp. 339-357.
@article{f0cc1b02e17d4ec390c58c40f9173df9,
title = "A flexible latent class approach to estimating test-score reliability",
abstract = "The latent class reliability coefficient (LCRC) is improved by using the divisive latent class model instead of the unrestricted latent class model. This results in the divisive latent class reliability coefficient (DLCRC), which unlike LCRC avoids making subjective decisions about the best solution and thus avoids judgment error. A computational study using large numbers of items shows that DLCRC also is faster than LCRC and fast enough for practical purposes. Speed and objectivity render DLCRC superior to LCRC. A decisive feature of DLCRC is that it aims at closely approximating the multivariate distribution of item scores, which might render the method suited when test data are multidimensional. A simulation study focusing on multidimensionality shows that DLCRC in general has little bias relative to the true reliability and is relatively accurate compared to LCRC and classical lower bound methods coefficients α and λ2 and the greatest lower bound.",
author = "{van der Palm}, {Daniel W.} and {van der Ark}, {L. Andries} and K. Sijtsma",
year = "2014",
doi = "10.1111/jedm.12053",
language = "English",
volume = "51",
pages = "339--357",
journal = "Journal of Educational Measurement",
issn = "0022-0655",
publisher = "Wiley-Blackwell",
number = "4",

}

A flexible latent class approach to estimating test-score reliability. / van der Palm, Daniel W.; van der Ark, L. Andries; Sijtsma, K.

In: Journal of Educational Measurement, Vol. 51, No. 4, 2014, p. 339-357.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - A flexible latent class approach to estimating test-score reliability

AU - van der Palm, Daniel W.

AU - van der Ark, L. Andries

AU - Sijtsma, K.

PY - 2014

Y1 - 2014

N2 - The latent class reliability coefficient (LCRC) is improved by using the divisive latent class model instead of the unrestricted latent class model. This results in the divisive latent class reliability coefficient (DLCRC), which unlike LCRC avoids making subjective decisions about the best solution and thus avoids judgment error. A computational study using large numbers of items shows that DLCRC also is faster than LCRC and fast enough for practical purposes. Speed and objectivity render DLCRC superior to LCRC. A decisive feature of DLCRC is that it aims at closely approximating the multivariate distribution of item scores, which might render the method suited when test data are multidimensional. A simulation study focusing on multidimensionality shows that DLCRC in general has little bias relative to the true reliability and is relatively accurate compared to LCRC and classical lower bound methods coefficients α and λ2 and the greatest lower bound.

AB - The latent class reliability coefficient (LCRC) is improved by using the divisive latent class model instead of the unrestricted latent class model. This results in the divisive latent class reliability coefficient (DLCRC), which unlike LCRC avoids making subjective decisions about the best solution and thus avoids judgment error. A computational study using large numbers of items shows that DLCRC also is faster than LCRC and fast enough for practical purposes. Speed and objectivity render DLCRC superior to LCRC. A decisive feature of DLCRC is that it aims at closely approximating the multivariate distribution of item scores, which might render the method suited when test data are multidimensional. A simulation study focusing on multidimensionality shows that DLCRC in general has little bias relative to the true reliability and is relatively accurate compared to LCRC and classical lower bound methods coefficients α and λ2 and the greatest lower bound.

U2 - 10.1111/jedm.12053

DO - 10.1111/jedm.12053

M3 - Article

VL - 51

SP - 339

EP - 357

JO - Journal of Educational Measurement

JF - Journal of Educational Measurement

SN - 0022-0655

IS - 4

ER -