Iterative Purification and Effect Size Use With Logistic Regression for Differential Item Functioning Detection

French, Brian F.; Maller, Susan J.

doi:10.1177/0013164406294781

Cited by 85 publications

(126 citation statements)

References 33 publications

Supporting

Mentioning

107

Contrasting

Unclassified

Order By: Relevance

“…However, a carefully designed purification procedure needs to be the first step for identifying potential DIF items when conducting DIF analyses with real data. In the literature, different anchor purification methods have been suggested to select DIF-free items for different DIF detection approaches (e.g., French and Maller, 2007;Wang et al, 2009;Woods, 2009b;Gonzalez-Betanzos and Abad, 2012). Depending on the selection of DIF-free items (i.e., purification), the DIF detection methods may provide different results regarding the number and type of detected DIF items.…”

Section: Limitations and Future Researchmentioning

confidence: 99%

Detecting Multidimensional Differential Item Functioning with the Multiple Indicators Multiple Causes Model, the Item Response Theory Likelihood Ratio Test, and Logistic Regression

Bulut

Suh

2017

Front. Educ.

View full text Add to dashboard Cite

Differential item functioning (DIF) is typically evaluated in educational and psychological assessments with a simple structure in which items are associated with a single latent trait. This study aims to extend the investigation of DIF for multidimensional assessments with a non-simple structure in which items can be associated with two or more latent traits. A simulation study was conducted with the multidimensional extensions of the item response theory likelihood ratio (IRT-LR) test, the multiple indicators multiple causes (MIMIC) model, and logistic regression for detecting uniform and non-uniform DIF in multidimensional assessments. The results indicated that the IRT-LR test outperformed the MIMIC and logistic regression approaches in detecting non-uniform DIF. When detecting uniform DIF, the MIMIC and logistic regression approaches appeared to perform better than the IRT-LR test in short tests, while the performances of all three approaches were very similar in longer tests. Type I error rates for logistic regression were severely inflated compared with the other two approaches. The IRT-LR test appears to be a more balanced and powerful method than the MIMIC and logistic regression approaches in detecting DIF in multidimensional assessments with a non-simple structure.

show abstract

Section: Limitations and Future Researchmentioning

confidence: 99%

Detecting Multidimensional Differential Item Functioning with the Multiple Indicators Multiple Causes Model, the Item Response Theory Likelihood Ratio Test, and Logistic Regression

Bulut

Suh

2017

Front. Educ.

View full text Add to dashboard Cite

show abstract

“…Scale levels were estimated by the sum score, including also the item being tested [26,30,31]. Other items with DIF were excluded from the sum score, i.e., it was purified [26,32].…”

Section: Methodsmentioning

confidence: 99%

ThyPROgr: the Greek edition of the ThyPRO questionnaires for patients with benign thyroid diseases

et al. 2018

View full text Add to dashboard Cite

The thyroid-related patient-reported outcome measure ThyPRO is currently the most reliable and valid instrument for the measurement of thyroid-related quality of life. The objective of the current study was to translate the original (85 items) and short (39 items) versions of ThyPRO into the Greek language as well as to validate and culturally adapt ThyPRO among the Greek population. Translation of patient-reported outcomes was done according to standard methodology. Following the translation process, the Greek version of ThyPRO (ThyPROgr) was validated in consecutive patients with thyroid diseases who visited the outpatient clinics of the Department of Endocrinology, "Hippokration" General Hospital, Thessaloniki, Greece, between September and December 2016. To test cross-cultural validity, analysis of differential item functioning (DIF) between the original Danish and the new Greek version, using ordinal logistic regression, was applied. Cross-cultural validity of the Greek translation was evaluated for both versions of ThyPRO. The ThyPRO questionnaire was validated in 143 consecutive patients (131 women and 12 men) with thyroid diseases, with a mean age of 62 ± 6 (SD) years. In the 85-item ThyPRO, 17 instances of DIF were detected, most of which were minor, this accounting for the~5% of the variation in scale score. Two instances of DIF among positively worded items were larger, with 8-10% of variance explained. For ThyPRO-39, five minor instances of DIF were found. As judged by the linguistic validation as well as the DIF analyses, ThyPROgr has good cross-cultural validity when compared with the original Danish version. The two versions of ThyPROgr can now be used to assess thyroid-specific quality of life in the Greek population.

show abstract

“…to unrefined results both in terms of Type I and Type II error. Recent reviews of the refinement literature (Colvin & Randall, 2011;French & Maller, 2007) concluded that refinement was typically found to have a favorable effect on the accuracy of DIF procedures.…”

Section: Criterion Refinement In Differential Item Functioning Analysesmentioning

confidence: 99%

A Review of Ets Differential Item Functioning Assessment Procedures: Flagging Rules, Minimum Sample Size Requirements, and Criterion Refinement

Zwick

2012

ETS Research Report Series

121

104

View full text Add to dashboard Cite

Differential item functioning (DIF) analysis is a key component in the evaluation of the fairness and validity of educational tests. The goal of this project was to review the status of ETS DIF analysis procedures, focusing on three aspects: (a) the nature and stringency of the statistical rules used to flag items, (b) the minimum sample size requirements that are currently in place for DIF analysis, and (c) the efficacy of criterion refinement. The main findings of the review are as follows:• The ETS C rule often displays low DIF detection rates even when samples are large.• With improved flagging rules in place, minimum sample size requirements could probably be relaxed. In addition, updated rules for combining data across administrations could allow DIF analyses to be performed in a broader range of situations.• Refinement of the matching criterion improves detection rates when DIF is primarily in one direction but can depress detection rates when DIF is balanced. If nothing is known about the likely pattern of DIF, refinement is advisable.Each of these findings is discussed in detail, focusing on the case of dichotomous items.

show abstract

Iterative Purification and Effect Size Use With Logistic Regression for Differential Item Functioning Detection

Cited by 85 publications

References 33 publications

Detecting Multidimensional Differential Item Functioning with the Multiple Indicators Multiple Causes Model, the Item Response Theory Likelihood Ratio Test, and Logistic Regression

Detecting Multidimensional Differential Item Functioning with the Multiple Indicators Multiple Causes Model, the Item Response Theory Likelihood Ratio Test, and Logistic Regression

ThyPROgr: the Greek edition of the ThyPRO questionnaires for patients with benign thyroid diseases

A Review of Ets Differential Item Functioning Assessment Procedures: Flagging Rules, Minimum Sample Size Requirements, and Criterion Refinement

Contact Info

Product

Resources

About