METHODS FOR ASSESSING UNIDIMENSIONALITY IN DICHOTOMOUS IRT MODELS

Q: How can I download an article?

To download an article from SID, first log in to the site, search for the article title, and click on the 'Download Article' option.

Q: How can I download an ISI article?

To download an ISI article on SID, enter the keyword or article title in the search bar, view the relevant results, click on the desired article, and select the 'Download Article' option.

Q: How can I access the SID database?

To access the SID database, visit SID.ir, create an account, and log in to access scientific resources.

Q: Is downloading articles from SID free?

Some articles on SID are available for free, while others require payment. Details are specified on the article's page.

MINAEI A.; FALSAFINEJAD M.R.

مرکز اطلاعات علمی Scientific Information Database (SID) - Trusted Source for Research and Academic Resources

Journal Paper

Paper Information

Journal: Educational Measurement Year:2010 | Volume:1 | Issue:3 Page(s): 71-99

Download Full-Text

Persian Verion

View:

4,879

Download:

Cites:

Information Journal Paper

Title

METHODS FOR ASSESSING UNIDIMENSIONALITY IN DICHOTOMOUS IRT MODELS

Author(s)

MINAEI A. | FALSAFINEJAD M.R. | Issue Writer Certificate

Keywords

ITEM RESPONSE THEORY (IRT)Q2

UNIDIMENSIONALITYQ2

ITEMQ3

Abstract

Most commonly employed IRT models are based on three assumptions of monotonicity, UNIDIMENSIONALITY and local/conditional independence. Monotonicity asserts that the higher the ability the higher the correct response probability. Local independence means that the response to any ITEM is unrelated to any other ITEM when trait level is controlled; and UNIDIMENSIONALITY means that all ITEMs of test, measure only one trait or latent variable. Violating UNIDIMENSIONALITY assumption may lead to a serious misestimation of ITEMs and persons parameters or standard errors. In addition, UNIDIMENSIONALITY is necessary and important for clear interpreting of test scores and comparing persons, as well as for some psychometrics works such as test equating. Despite of importance of UNIDIMENSIONALITY assumption in IRT models, there is no agreement among psychometricians concerning the definition and assessing of UNIDIMENSIONALITY. Given the role and importance of local independence in defining and determining new methods for assessing test dimensionality, in this paper in addition doing a literature review on local independence and introducing several methods for assessing UNIDIMENSIONALITY in dichotomous IRT models, we used NOHARM and DIMTEST methods to assess UNIDIMENSIONALITY of a set of test data and both methods resulted in similar results. Given the present literature and results from NOHARM and DIMTEST, we maintain that researchers should now be starting to move from reporting heuristic indices such as "variance accounted for by the first factor" or "ration of first to second eigenvalue" and start implementing the new procedures that tackle these issues in a much more sophisticated manner such as, NOHARM, TESTFACT, DIMTEST, DETECT, H&R, and MSP.

Cites

No record.

References

No record.

Cite

APA: Copy

MINAEI, A., & FALSAFINEJAD, M.R.. (2010). METHODS FOR ASSESSING UNIDIMENSIONALITY IN DICHOTOMOUS IRT MODELS. TRAINING MEASUREMENT, 1(3), 71-99. SID. https://sid.ir/paper/214519/en

Vancouver: Copy

MINAEI A., FALSAFINEJAD M.R.. METHODS FOR ASSESSING UNIDIMENSIONALITY IN DICHOTOMOUS IRT MODELS. TRAINING MEASUREMENT[Internet]. 2010;1(3):71-99. Available from: https://sid.ir/paper/214519/en

IEEE: Copy

A. MINAEI, and M.R. FALSAFINEJAD, “METHODS FOR ASSESSING UNIDIMENSIONALITY IN DICHOTOMOUS IRT MODELS,” TRAINING MEASUREMENT, vol. 1, no. 3, pp. 71–99, 2010, [Online]. Available: https://sid.ir/paper/214519/en

Related Journal Papers