scholarly journals An R toolbox for score-based measurement invariance tests in IRT models

2021 ◽  
Author(s):  
Lennart Schneider ◽  
Carolin Strobl ◽  
Achim Zeileis ◽  
Rudolf Debelak
Keyword(s):  
Measurement Invariance ◽  
Statistical Computing ◽  
Partial Credit Model ◽  
Generalized Partial Credit Model ◽  
The Past ◽  
New Family ◽  
Irt Models ◽  
Item Functioning ◽  
Generalized Partial Credit ◽  
R Packages

AbstractThe detection of differential item functioning (DIF) is a central topic in psychometrics and educational measurement. In the past few years, a new family of score-based tests of measurement invariance has been proposed, which allows the detection of DIF along arbitrary person covariates in a variety of item response theory (IRT) models. This paper illustrates the application of these tests within the R system for statistical computing, making them accessible to a broad range of users. This presentation also includes IRT models for which these tests have not previously been investigated, such as the generalized partial credit model. The paper has three goals: First, we review the ideas behind score-based tests of measurement invariance. Second, we describe the implementation of these tests within the R system for statistical computing, which is based on the interaction of the R packages mirt, psychotools and strucchange. Third, we illustrate the application of this software and the interpretation of its output in two empirical datasets. The complete R code for reproducing our results is reported in the paper.

scholarly journals An R Toolbox for Score-Based Measurement Invariance Tests in IRT Models

2020 ◽  
Author(s):  
Lennart Schneider ◽  
Carolin Strobl ◽  
Achim Zeileis ◽  
Rudolf Debelak
Keyword(s):  
Measurement Invariance ◽  
Statistical Computing ◽  
Partial Credit Model ◽  
Generalized Partial Credit Model ◽  
New Family ◽  
Irt Models ◽  
Item Functioning ◽  
Generalized Partial Credit ◽  
R Packages ◽  
Power Analyses

The detection of differential item functioning (DIF) is a central topic in psychometrics and educational measurement. In the past few years, a new family of score-based tests of measurement invariance has been proposed that allows the detection of DIF along arbitrary person covariates in a variety of item response theory (IRT) models. This paper illustrates the application of these tests within the R system for statistical computing, making them accessible to a broad range of users. This presentation also includes IRT models for which these tests have not previously been investigated, such as the generalized partial credit model. The paper has three goals: First, we review the ideas behind score-based tests of measurement invariance. Second, we describe the implementation of these tests within the R system for statistical computing, which is based on the interaction of the R packages mirt, psychotools and strucchange. Third, we illustrate the application of this software and the interpretation of its output in two empirical datasets, and show how to conduct simulation studies, such as IRT-based power analyses, in the context of DIF investigations. The complete R code for reproducing our results is reported in the paper and its appendix.

scholarly journals Contribuições ao estudo do Modelo de Crédito Parcial Generalizado

2017 ◽  
Vol 28 (67) ◽  
pp. 236
Author(s):  
Eduardo Vargas Ferreira ◽  
Caio Lucidius Naberezny Azevedo
Keyword(s):  
Item Response Theory ◽  
Item Response ◽  
Model Parameters ◽  
Partial Credit Model ◽  
Partial Credit ◽  
Response Theory ◽  
Generalized Partial Credit Model ◽  
Irt Models ◽  
Palabras Clave ◽  
Generalized Partial Credit

<p>Este artigo aborda os mais importantes aspectos inferenciais do Modelo de Crédito Parcial Generalizado (MCPG), da Teoria da Resposta ao Item (TRI). É mostrado um estudo sobre uma das principais dificuldades encontradas no processo de estimação e inferência dos modelos da TRI, que é a falta de identificabilidade. Além disso, apresenta-se a interpretação dos parâmetros do modelo e da função de informação do item e do teste.</p><p><strong>Palavras-chave:</strong> Teoria da Resposta ao Item; Modelos Politômicos; Modelo de Crédito Parcial Generalizado; Psicometria.</p><p> </p><p><strong>Contribuciones al estudio del Modelo de Crédito Parcial Generalizado</strong></p><p>Este artículo aborda los más importantes aspectos inferenciales del Modelo de Crédito Parcial Generalizado (MCPG), de la Teoría de la Respuesta al Ítem (TRI). Se presenta un estudio sobre una de las principales dificultades encontradas en el proceso de estimación e inferencia de los modelos de la TRI, que es la falta de identificabilidad. Por otra parte, se expone la interpretación de los parámetros del modelo y de la función de información del ítem y el test.</p><p><strong>Palabras clave:</strong> Teoría de la Respuesta al Ítem; Modelos Politómicos; Modelo de Crédito Parcial Generalizado; Psicometría.</p><p> </p><p><strong>Contributions to the study of Generalized Partial Credit Model</strong></p><p>This article covers the most important inferential aspects of the Generalized Partial Credit Model (GPCM) of the Item Response Theory (IRT). It presents a study on one of the main difficulties encountered in the process of estimation and inference of the IRT models, which is the lack of identifiability. In addition, it presents the interpretation of the model parameters and the information function of the item and the test.</p><p><strong>Keywords:</strong> Item Response Theory; Polytomous Models; Generalized Partial Credit Model; Psychometrics.</p>

scholarly journals Translation and Psychometric Evaluation of the Dutch Families Importance in Nursing Care: Nurses’ Attitudes Scale Based on the Generalized Partial Credit Model

2018 ◽  
Vol 24 (4) ◽  
pp. 538-562 ◽  
Author(s):  
E. I. Hagedoorn ◽  
W. Paans ◽  
T. Jaarsma ◽  
J. C. Keers ◽  
C. P. van der Schans ◽  
...  
Keyword(s):  
Psychometric Properties ◽  
Nursing Care ◽  
Latent Trait ◽  
Psychometric Evaluation ◽  
Partial Credit Model ◽  
Partial Credit ◽  
Cross Sectional ◽  
Online Application ◽  
Generalized Partial Credit Model ◽  
Generalized Partial Credit

The instrument called Families Importance in Nursing Care–Nurses’ Attitudes (FINC-NA) is used to measure nurses’ attitudes toward involving families in their nursing care. The aim of this study is to evaluate the FINC-NA scale in a population of Dutch nurses and add new psychometric information to existing knowledge about this instrument. Using a cross-sectional design, 1,211 nurses received an online application in 2015. Psychometric properties were based on polychoric correlations and the Generalized Partial Credit Model. A total of 597 (49%) nurses responded to the online application. Results confirmed a four-subscale structure. All response categories were utilized, although some ceiling effects occurred. Most items increase monotonically, and the majority of items discriminate well between different latent trait scores of nurses with some items providing more information than others. This study reports the psychometric properties of the Dutch language FINC-NA instrument. New insights into the construct and content of items enable the possibility of a more generic instrument that could be valid across several cultures.

scholarly journals Extreme Response Style: A Simulation Study Comparison of Three Multidimensional Item Response Models

2018 ◽  
Vol 43 (4) ◽  
pp. 322-335 ◽  
Author(s):  
Brian C. Leventhal
Keyword(s):  
Item Response ◽  
Partial Credit Model ◽  
Response Model ◽  
Multidimensional Item Response ◽  
Generalized Partial Credit Model ◽  
Response Models ◽  
Item Response Models ◽  
Nominal Response Model ◽  
Extreme Response ◽  
Generalized Partial Credit

Several multidimensional item response models have been proposed for survey responses affected by response styles. Through simulation, this study compares three models designed to account for extreme response tendencies: the IRTree Model, the multidimensional nominal response model, and the modified generalized partial credit model. The modified generalized partial credit model results in the lowest item mean squared error (MSE) across simulation conditions of sample size (500, 1,000), survey length (10, 20), and number of response options (4, 6). The multidimensional nominal response model is equally suitable for surveys measuring one substantive trait using responses to 10 four-option, forced-choice Likert-type items. Based on data validation, comparison of item MSE, and posterior predictive model checking, the IRTree Model is hypothesized to account for additional sources of construct-irrelevant variance.

Sign in / Sign up

Export Citation Format

Share Document