Testing Hypotheses About the Person Response Function in Person-Fit Analysis

Wilco H.M. Emons; Klaas Sijtsma; Rob R. Meijer

doi:10.1207/s15327906mbr3901_1

Testing Hypotheses About the Person Response Function in Person-Fit Analysis

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

Research output: Contribution to journal › Article › Academic › peer-review

26 Citations (Scopus)

26 Downloads (Pure)

Abstract

The person-response function (PRF) relates the probability of an individual's correct answer to the difficulty of items measuring the same latent trait. Local deviations of the observed PRF from the expected PRF indicate person misfit. We discuss two new approaches to investigate person fit. The first approach uses kernel smoothing to estimate continuous PRF estimates. Graphical displays of PRFs were used to localize and diagnose misfit. The second approach approximates the PRF by a logistic regression model. Hypothesis tests on the regression parameters were used to detect certain types of misfit. A simulation study was conducted to investigate the Type I error rates and the detection rates of the regression approach.

Original language	English
Pages (from-to)	1-35
Number of pages	35
Journal	Multivariate behavioral research
Volume	39
Issue number	1
DOIs	https://doi.org/10.1207/s15327906mbr3901_1
Publication status	Published - 2004

Access to Document

10.1207/s15327906mbr3901_1

Cite this

@article{095c0693ebd648fab817db2f3d299115,

title = "Testing Hypotheses About the Person Response Function in Person-Fit Analysis",

abstract = "The person-response function (PRF) relates the probability of an individual's correct answer to the difficulty of items measuring the same latent trait. Local deviations of the observed PRF from the expected PRF indicate person misfit. We discuss two new approaches to investigate person fit. The first approach uses kernel smoothing to estimate continuous PRF estimates. Graphical displays of PRFs were used to localize and diagnose misfit. The second approach approximates the PRF by a logistic regression model. Hypothesis tests on the regression parameters were used to detect certain types of misfit. A simulation study was conducted to investigate the Type I error rates and the detection rates of the regression approach.",

author = "Emons, {Wilco H.M.} and Klaas Sijtsma and Meijer, {Rob R.}",

year = "2004",

doi = "10.1207/s15327906mbr3901_1",

language = "English",

volume = "39",

pages = "1--35",

journal = "Multivariate behavioral research",

issn = "0027-3171",

publisher = "Psychology Press Ltd",

number = "1",

}

TY - JOUR

T1 - Testing Hypotheses About the Person Response Function in Person-Fit Analysis

AU - Emons, Wilco H.M.

AU - Sijtsma, Klaas

AU - Meijer, Rob R.

PY - 2004

Y1 - 2004

N2 - The person-response function (PRF) relates the probability of an individual's correct answer to the difficulty of items measuring the same latent trait. Local deviations of the observed PRF from the expected PRF indicate person misfit. We discuss two new approaches to investigate person fit. The first approach uses kernel smoothing to estimate continuous PRF estimates. Graphical displays of PRFs were used to localize and diagnose misfit. The second approach approximates the PRF by a logistic regression model. Hypothesis tests on the regression parameters were used to detect certain types of misfit. A simulation study was conducted to investigate the Type I error rates and the detection rates of the regression approach.

AB - The person-response function (PRF) relates the probability of an individual's correct answer to the difficulty of items measuring the same latent trait. Local deviations of the observed PRF from the expected PRF indicate person misfit. We discuss two new approaches to investigate person fit. The first approach uses kernel smoothing to estimate continuous PRF estimates. Graphical displays of PRFs were used to localize and diagnose misfit. The second approach approximates the PRF by a logistic regression model. Hypothesis tests on the regression parameters were used to detect certain types of misfit. A simulation study was conducted to investigate the Type I error rates and the detection rates of the regression approach.

U2 - 10.1207/s15327906mbr3901_1

DO - 10.1207/s15327906mbr3901_1

M3 - Article

SN - 0027-3171

VL - 39

SP - 1

EP - 35

JO - Multivariate behavioral research

JF - Multivariate behavioral research

IS - 1

ER -

Testing Hypotheses About the Person Response Function in Person-Fit Analysis

Abstract

Access to Document

Fingerprint

Cite this