Effects of Anchor Item Methods on Differential Item Functioning Detection with the Likelihood Ratio Test

Wang, Wen Chung; Yeh, Ya-Li

doi:10.1177/0146621603259902

Cited by 110 publications

(189 citation statements)

References 29 publications

(35 reference statements)

Supporting

Mentioning

169

Contrasting

Order By: Relevance

“…This procedure has been implemented on several popular computer programs, such as BILOG-MG [37], WINSTEPS [38], and ConQuest. There are other procedures to solve scale indeterminacy [39], for example, if a set of items are believed to have no DIF, they can served as the anchors so that the other items can be detected for the evidence of DIF. Unfortunately, this procedure was not applicable in this study, because no prior knowledge was available to claim which items were indeed DIF-free.…”

Section: Resultsmentioning

confidence: 99%

Validating, Improving Reliability, and Estimating Correlation of the Four Subscales in the WHOQOL-BREF using Multidimensional Rasch Analysis

et al. 2006

Self Cite

View full text Add to dashboard Cite

Objective: This study examined the construct validity, and improved the test reliability and the estimation accuracy for the correlation between domains of the WHOQOL-BREF using multidimensional Rasch analysis. Method: A total of 13,083 adults were administered the 28-item WHOQOL-BREF Taiwan version, which consists of 4 subscales (domains). The multidimensional form of the partial credit model was used to examine the fit of the 4 subscales. For comparison, each subscale individually was also fitted to the unidimensional partial credit model. Standard item fit statistics and analysis of differential item functioning (DIF) were used to check model-data fit. Results: After excluding 2 overall items and deleting 7 DIF items, the remaining items of each subscale in the WHOQOL-BREF constituted a single construct. The test reliabilities and correlations between domains obtained from the multidimensional approach, (0.82-0.86) and (0.79-0.89), respectively, were much higher than those obtained from the unidimensional approach, (0.67-0.75) and (0.53-0.65), respectively. Conclusion: The 19-item WHOQOL-BREF measures more succinct latent traits than the original design. The multidimensional approach yields not only more accurate estimates for the correlation between domains but also substantially higher reliabilities, than the standard unidimensional approach.

show abstract

Section: Resultsmentioning

confidence: 99%

Validating, Improving Reliability, and Estimating Correlation of the Four Subscales in the WHOQOL-BREF using Multidimensional Rasch Analysis

et al. 2006

Self Cite

View full text Add to dashboard Cite

show abstract

“…Several commonly used models and associated test statistics have been developed to detect DIF as a function of membership in observed groups (like gender or ethnicity), including the Mantel-Hansel test (Holland & Thayer, 1988), the standardization method (Dorans & Kulick, 1986), the logistic regression model (for example, Swaminathan & Rogers, 1990), the IRT-based chi-square test (Lord, 1980;Wright & Stone, 1979), the likelihood ratio test (IRT-LRT; Thissen, Steinberg, & Wainer, 1988;Wang & Yeh, 2003), and the multiple indicators multiple causes (MIMIC : Muthén, 1985;1989) model.…”

Section: Chapter 1: Introductionmentioning

confidence: 99%

Evaluation of Two Types of Differential Item Functioning in Factor Mixture Models With Binary Outcomes

Lee

Beretvas

2014

Educational and Psychological Measurement

View full text Add to dashboard Cite

Conventional differential item functioning (DIF) detection methods (e.g., the Mantel–Haenszel test) can be used to detect DIF only across observed groups, such as gender or ethnicity. However, research has found that DIF is not typically fully explained by an observed variable. True sources of DIF may include unobserved, latent variables, such as personality or response patterns. The factor mixture model (FMM) is designed to detect unobserved sources of heterogeneity in factor models. The current study investigated use of the FMM for detecting between-class latent DIF and class-specific observed DIF. Factors that were manipulated included the DIF effect size and the latent class probabilities. The performance of model fit indices (Akaike information criterion [AIC], Bayesian information criterion [BIC], sample size–adjusted BIC, and consistent AIC) were assessed for their detection of the correct DIF model. The recovery of DIF parameters was also assessed. Results indicated that use of FMMs with binary outcomes performed well in terms of the DIF detection and for recovery of large DIF effects. When class probabilities were unequal with small DIF effects, performance decreased for fit indices, power, and the recovery of DIF effects compared with equal class probability conditions. Inflated Type I errors were found for non-DIF items across simulation conditions. Results and future research directions for applied and methodological are discussed.

show abstract

“…In addition, as Wang and Yeh (2003) stated, if LR DIF can detect non-uniform DIF better than the MH DIF method, and is as powerful at detecting uniform DIF as the MH DIF method, then the inclusion of an effect size would make LR DIF a very attractive choice as a DIF detection method.…”

Section: Discussionmentioning

confidence: 99%

A Psychometric Analysis of an Inventory for Evaluating Students' Leadership Skills

Ibrahim¹

2017

JTTE

View full text Add to dashboard Cite

Abstract:The study classified individual items that function differentially according to the magnitude of DIF in dichotomous test and compared the power of Generalized Mantel Haenszel (GMH), Simultaneous Item Bias Test (SIBTEST), and Logistic Discriminant Function Analysis (LDFA) methods in detecting DIF in dichotomous test items. It also determined the relationship between the proportions of test items that function differentially in dichotomous test items when the different methods were used. These were with a view to improving the quality of dichotomous test items construction. The study adopted a survey design. The population consisted of all undergraduate students who registered for EFC 303 (Tests and Measurement) at Obafemi Awolowo University during 2011/2012 Harmattan Semester. The sample consisted of an intact class of 457 Part 3 undergraduate students who registered for the course. A 50-item multiple -choice achievement test designated "Undergraduate Students' Achievement Test (USAT)" was developed by the researcher for the study. A total of 445 scripts were found properly completed. Data collected were analysed using Generalized Mantel Haenszel, Simultaneous Item Bias Test, and Logistic Discriminant Function Analysis. The results showed that there was a significant difference in the performance of the GMH, SIBTEST, and LDFA methods in detecting DIF in dichotomous test items. Further, a non significant relationship existed between the proportions of test items that functioned differentially in the dichotomous test items when the different methods were used. The study concluded that GMH, SIBTEST and LDFA were effective in detecting DIF across dichotomous test items but complement one other in their ability to detect DIF in dichotomous test items.

show abstract

Effects of Anchor Item Methods on Differential Item Functioning Detection with the Likelihood Ratio Test

Cited by 110 publications

References 29 publications

Validating, Improving Reliability, and Estimating Correlation of the Four Subscales in the WHOQOL-BREF using Multidimensional Rasch Analysis

Validating, Improving Reliability, and Estimating Correlation of the Four Subscales in the WHOQOL-BREF using Multidimensional Rasch Analysis

Evaluation of Two Types of Differential Item Functioning in Factor Mixture Models With Binary Outcomes

A Psychometric Analysis of an Inventory for Evaluating Students' Leadership Skills

Contact Info

Product

Resources

About