Standard

Why ordinal variables can (almost) always be treated as continuous variables : Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods. / Robitzsch, Alexander.

In: Frontiers in Education, Vol. 5, 589965, 10.2020.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Robitzsch, A 2020, 'Why ordinal variables can (almost) always be treated as continuous variables: Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods', Frontiers in Education, vol. 5, 589965. https://doi.org/10.3389/feduc.2020.589965

APA

Robitzsch, A. (2020). Why ordinal variables can (almost) always be treated as continuous variables: Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods. Frontiers in Education, 5, [589965]. https://doi.org/10.3389/feduc.2020.589965

Vancouver

Robitzsch A. Why ordinal variables can (almost) always be treated as continuous variables: Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods. Frontiers in Education. 2020 Oct;5. 589965. https://doi.org/10.3389/feduc.2020.589965

Author

Robitzsch, Alexander. / Why ordinal variables can (almost) always be treated as continuous variables : Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods. In: Frontiers in Education. 2020 ; Vol. 5.

BibTeX

@article{c4db49e83cce421bb57665e4288ded03,
title = "Why ordinal variables can (almost) always be treated as continuous variables: Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods",
abstract = "The analysis of factor structures is one of the most critical psychometric applications. Frequently, variables (i.e., items or indicators) resulting from questionnaires using ordinal items with 2–7 categories are used. There are plenty of articles that recommend treating ordinal variables in a factor analysis by default as ordinal and not as continuous imposing a multivariate normal distribution assumption. In this article, we exhibit that the reasoning behind such suggestions is flawed. In our view, findings from simulation studies cannot tell about the right modeling strategy of ordinal variables in factor analysis. Moreover, it is argued that ordinal factor models impose a normality assumption for underlying continuous variables, which might also often be incorrect in empirical applications. However, researchers seldom opt for more flexible modeling strategies that involve correctly specified distributions. Finally, the consequences of modeling choices for validity, reliability, measurement invariance, handling of missing data, and the assessment of global model fit are discussed.",
keywords = "Methodological research and development, factor analysis, ordinal variable, polychoric correlations, structural equation modeling, Gaussian copula model",
author = "Alexander Robitzsch",
year = "2020",
month = oct,
doi = "10.3389/feduc.2020.589965",
language = "English",
volume = "5",
journal = "Frontiers in Education",
issn = "2504-284X",
publisher = "Frontiers Media S.A.",

}

RIS

TY - JOUR

T1 - Why ordinal variables can (almost) always be treated as continuous variables

T2 - Clarifying assumptions of robust continuous and ordinal factor analysis estimation methods

AU - Robitzsch, Alexander

PY - 2020/10

Y1 - 2020/10

N2 - The analysis of factor structures is one of the most critical psychometric applications. Frequently, variables (i.e., items or indicators) resulting from questionnaires using ordinal items with 2–7 categories are used. There are plenty of articles that recommend treating ordinal variables in a factor analysis by default as ordinal and not as continuous imposing a multivariate normal distribution assumption. In this article, we exhibit that the reasoning behind such suggestions is flawed. In our view, findings from simulation studies cannot tell about the right modeling strategy of ordinal variables in factor analysis. Moreover, it is argued that ordinal factor models impose a normality assumption for underlying continuous variables, which might also often be incorrect in empirical applications. However, researchers seldom opt for more flexible modeling strategies that involve correctly specified distributions. Finally, the consequences of modeling choices for validity, reliability, measurement invariance, handling of missing data, and the assessment of global model fit are discussed.

AB - The analysis of factor structures is one of the most critical psychometric applications. Frequently, variables (i.e., items or indicators) resulting from questionnaires using ordinal items with 2–7 categories are used. There are plenty of articles that recommend treating ordinal variables in a factor analysis by default as ordinal and not as continuous imposing a multivariate normal distribution assumption. In this article, we exhibit that the reasoning behind such suggestions is flawed. In our view, findings from simulation studies cannot tell about the right modeling strategy of ordinal variables in factor analysis. Moreover, it is argued that ordinal factor models impose a normality assumption for underlying continuous variables, which might also often be incorrect in empirical applications. However, researchers seldom opt for more flexible modeling strategies that involve correctly specified distributions. Finally, the consequences of modeling choices for validity, reliability, measurement invariance, handling of missing data, and the assessment of global model fit are discussed.

KW - Methodological research and development

KW - factor analysis

KW - ordinal variable

KW - polychoric correlations

KW - structural equation modeling

KW - Gaussian copula model

U2 - 10.3389/feduc.2020.589965

DO - 10.3389/feduc.2020.589965

M3 - Journal article

VL - 5

JO - Frontiers in Education

JF - Frontiers in Education

SN - 2504-284X

M1 - 589965

ER -

ID: 1419559