Data Reduction in Item Response Theory Models

Authors

Keywords:

IRT models, simulation, reduction of categories

Abstract

One of the main classes of statistical models established in psychometric literature is Item Response Theory (IRT), which plays a crucial role in constructing scales of latent traits and allows for the creation of measurement instruments that are more precise and adapted to respondents' characteristics. The primary objective of this paper is to analyze, through simulation studies, the possible consequences of reducing response categories on the parameter estimates of IRT models, as this process can simplify the estimation phase and is used in some research. After analyzing the obtained results, it is concluded that this reduction mechanism provides satisfactory computational optimization. However, it also results in the loss of information, as the quality of IRT parameter estimates is negatively affected, directly interfering with the results of applied studies. Finally, this analysis aims to contribute significantly to psychometric literature, becoming capable of promoting new research in this statistical domain, as well as providing a solid foundation for future methodological developments.

References

ANDRADE, D. F.; TAVARES, H. R.; VALLE, R. C. Teoria da Resposta ao Item: conceitos e aplicações. ABE, São Paulo, 2000.

BIRNBAUM, A. Some latent trait models and their use in inferring an examinee's ability. Statistical theories of mental test scores, 1968.

BUENO, R. B. et al. Analysis of environmental sustainability perception in Paraná Basin III residents via Item Response Theory. Brazilian Journal of Biometrics, v. 42, n. 2, p. 158-170, 2024.

CHALMERS, R. P. mirt: A multidimensional item response theory package for the R environment. Journal of statistical Software, v. 48, p. 1-29, 2012.

MOREIRA, M.d.R et al. The Perception of Brazilian Livestock Regarding the Use of Precision Livestock Farming for Animal Welfare. Agriculture, v. 14, n. 8, p. 1-17, 2024.

DA SILVA, M. A. et al. Estimating the DINA model parameters using the No-U-Turn Sampler. Biometrical Journal, v. 60, n. 2, p. 352-368, 2018.

EHRLICH, P. J. Pesquisa operacional: curso introdutório. Atlas, 1991.

FRAGOSO, T. M.; CÚRI, M. Improving psychometric assessment of the Beck Depression Inventory using multidimensional item response theory. Biometrical Journal, v. 55, n. 4, p. 527-540, 2013.

LORD, Frederic. A theory of test scores. Psychometric monographs, 1952.

PEREIRA, R. H. M. et al. geobr: download official spatial data sets of Brazil. R package version 1.6.1, 2021.

R CORE TEAM. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria, 2023.

RAO, C. R.; SINHARAY, S. (Ed.). Psychometrics. Elsevier, 2007.

SAMEJIMA, F. Estimation of latent ability using a response pattern of graded scores. Psychometrika monograph supplement, 1969.

TEAM, RStudio. RStudio: integrated development for R 2022. RStudio, PBC: Boston, MA, USA, 2022.

THISSEN, D.; STEINBERG, L. A taxonomy of item response models. Psychometrika, v. 51, n. 4, p. 567-577, 1986.

VINCENZI, S. L. et al. Assessment of environmental sustainability perception through item response theory: A case study in Brazil. Journal of Cleaner Production, v. 170, p. 1369-1386, 2018.

WICKHAM, H. et al. Welcome to the Tidyverse. Journal of open source software, v. 4, n. 43, p. 1686, 2019.

Published

17-12-2024

How to Cite

Barufaldi Bueno, R., & Andrade da Silva, M. (2024). Data Reduction in Item Response Theory Models. Sigmae, 13(4), 293–307. Retrieved from https://publicacoes.unifal-mg.edu.br/revistas/index.php/sigmae/article/view/2527

Issue

Section

Applied Statistics