Modification indices for the 2-pl and the nominal response model

Research output: Contribution to journalArticleAcademicpeer-review

84 Citations (Scopus)

Abstract

In this paper, it is shown that various violations of the 2-PL model and the nominal response model can be evaluated using the Lagrange multiplier test or the equivalent efficient score test. The tests presented here focus on violation of local stochastic independence and insufficient capture of the form of the item characteristic curves. Primarily, the tests are item-oriented diagnostic tools, but taken together, they also serve the purpose of evaluation of global model fit. A useful feature of Lagrange multiplier statistics is that they are evaluated using maximum likelihood estimates of the null-model only, that is, the parameters of alternative models need not be estimated. As numerical examples, an application to real data and some power studies are presented.
Original languageEnglish
Pages (from-to)273-294
Number of pages21
JournalPsychometrika
Volume64
Issue number3
DOIs
Publication statusPublished - 1999

Fingerprint

Likelihood Functions
Categorical or nominal
Lagrange multipliers
Lagrange multiplier Test
Characteristic Curve
Model
Score Test
Maximum Likelihood Estimate
Maximum likelihood
Null
Diagnostics
Statistics
Numerical Examples
Alternatives
Evaluation

Keywords

  • Efficient score test - generalized partial credit model - item response theory - model fit - modification indices - 2-parameter logistic model - nominal response model - Lagrange multiplier test
  • IR-60251
  • METIS-135401

Cite this

@article{ff26e5c62898411b88b195cd60cd631b,
title = "Modification indices for the 2-pl and the nominal response model",
abstract = "In this paper, it is shown that various violations of the 2-PL model and the nominal response model can be evaluated using the Lagrange multiplier test or the equivalent efficient score test. The tests presented here focus on violation of local stochastic independence and insufficient capture of the form of the item characteristic curves. Primarily, the tests are item-oriented diagnostic tools, but taken together, they also serve the purpose of evaluation of global model fit. A useful feature of Lagrange multiplier statistics is that they are evaluated using maximum likelihood estimates of the null-model only, that is, the parameters of alternative models need not be estimated. As numerical examples, an application to real data and some power studies are presented.",
keywords = "Efficient score test - generalized partial credit model - item response theory - model fit - modification indices - 2-parameter logistic model - nominal response model - Lagrange multiplier test, IR-60251, METIS-135401",
author = "Glas, {Cornelis A.W.}",
year = "1999",
doi = "10.1007/BF02294296",
language = "English",
volume = "64",
pages = "273--294",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer",
number = "3",

}

Modification indices for the 2-pl and the nominal response model. / Glas, Cornelis A.W.

In: Psychometrika, Vol. 64, No. 3, 1999, p. 273-294.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Modification indices for the 2-pl and the nominal response model

AU - Glas, Cornelis A.W.

PY - 1999

Y1 - 1999

N2 - In this paper, it is shown that various violations of the 2-PL model and the nominal response model can be evaluated using the Lagrange multiplier test or the equivalent efficient score test. The tests presented here focus on violation of local stochastic independence and insufficient capture of the form of the item characteristic curves. Primarily, the tests are item-oriented diagnostic tools, but taken together, they also serve the purpose of evaluation of global model fit. A useful feature of Lagrange multiplier statistics is that they are evaluated using maximum likelihood estimates of the null-model only, that is, the parameters of alternative models need not be estimated. As numerical examples, an application to real data and some power studies are presented.

AB - In this paper, it is shown that various violations of the 2-PL model and the nominal response model can be evaluated using the Lagrange multiplier test or the equivalent efficient score test. The tests presented here focus on violation of local stochastic independence and insufficient capture of the form of the item characteristic curves. Primarily, the tests are item-oriented diagnostic tools, but taken together, they also serve the purpose of evaluation of global model fit. A useful feature of Lagrange multiplier statistics is that they are evaluated using maximum likelihood estimates of the null-model only, that is, the parameters of alternative models need not be estimated. As numerical examples, an application to real data and some power studies are presented.

KW - Efficient score test - generalized partial credit model - item response theory - model fit - modification indices - 2-parameter logistic model - nominal response model - Lagrange multiplier test

KW - IR-60251

KW - METIS-135401

U2 - 10.1007/BF02294296

DO - 10.1007/BF02294296

M3 - Article

VL - 64

SP - 273

EP - 294

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 3

ER -