The Recovery of Correlation Between Latent Abilities Using Compensatory and Noncompensatory Multidimensional IRT Models. Issue 2 (2nd April 2020)

• Record Type: Journal Article
• Authors: Fu, Yanyan; Strachan, Tyler; Ip, Edward H.; Willse, John T.; Chen, Shyh-Huei; Ackerman, Terry
• Is Part Of: International journal of testing. Volume 20:Issue 2(2020)
• Page Start: 169
• Page End: 186
• Publication Date: 2020-04-02
• Subjects: compensatory and noncompensatory models -- latent correlation bias -- item response theory -- multidimensional item response theory -- simple structure; Psychological tests -- Periodicals; Educational tests and measurements -- Periodicals; 150.28705
• DOI: 10.1080/15305058.2019.1692212
• Languages: English
• ISSNs: 1530-5058

Abstract

Abstract : This research examined correlation estimates between latent abilities when using the two-dimensional and three-dimensional compensatory and noncompensatory item response theory models. Simulation study results showed that the recovery of the latent correlation was best when the test contained 100% of simple structure items for all models and conditions. When a test measured weakly discriminated dimensions, it became harder to recover the latent correlation. Results also showed that increasing the sample size, test length, or using simpler models (i.e., two-parameter logistic rather than three-parameter logistic, compensatory rather than noncompensatory) could improve the recovery of latent correlation.