Evaluating Type I Error and Power Rates Using an Effect Size Measure With the Logistic Regression Procedure for DIF Detection

By M. Jodoin, Mark J. Gierl. 2001

496 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Information Systems, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/2e6b125046a9efcd914aeef32caeb881f1105bcb

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Information Systems

In Management Science and Operations Research

Citations in the corpus

Effect of Multiple Testing Adjustment in Differential Item Functioning Detection (2013, unknown): “For the test length, lengths of 20 items and 40 items were chosen as common assessments are constructed with equal or fewer than 40 items (Jodoin & Gierl, 2001; Narayanan & Swaminathan, 1994; Raju et al., 1995; Rogers & Swaminathan, 1993; Roussos & Stout, 1996).”
Facing the situation: Testing a biosocial contingency model of leadership in intergroup relations using masculine and feminine faces (2012, background): “2 Effects sizes for the logistic regression procedures were not reported as the dichotomous nature of the dependent variable only provides limited information— particularly for relatively small sample sizes (Jodoin & Gierl, 2001).”
Effects of Anchor Item Methods on Differential Item Functioning Detection with the Likelihood Ratio Test (2003, unknown): “…other DIF detection methods with or without iterative purification procedures (e.g., Candell & Drasgow, 1988; Hidalgo-Montesinos & Lopez-Pina, 2002; Jodoin & Gierl, 2001; Kim & Cohen, 1995; Li & Stout, 1996; Lord, 1980; Miller & Oshima, 1992; Swaminathan & Rogers, 1990; van der Flier, Mellenbergh,…”
lordif: An R Package for Detecting Differential Item Functioning Using Iterative Hybrid Ordinal Logistic Regression/Item Response Theory and Monte Carlo Simulations (2011, background): “Subsequent studies (Jodoin and Gierl 2001; Kim et al. 2007), however, found the proposed thresholds to be too large, resulting in under-identification of DIF.”
Iterative Purification and Effect Size Use With Logistic Regression for Differential Item Functioning Detection (2007, methodology): “The use of purification and the combination of the statistical test and effect size may improve the LR method by (a) enhancement of the validity of the matching criterion (Jodoin & Gierl, 2001), (b) improvement of statistical power, and (c) reduction of the Type I error rate.”
Factors Influencing the Mantel and Generalized Mantel-Haenszel Methods for the Assessment of Differential Item Functioning in Polytomous Items (2004, methodology): “Although the logistic regression procedure seems to be effective in detecting nonuniform DIF in dichotomous items (Jodoin & Gierl, 2001; Rogers & Swaminathan, 1993; Swaminathan & Rogers, 1990), extending the logistic regression procedure to assess DIF in polytomous items when tests contain multiple…”
Differential Item Functioning Detection and Effect Size: A Comparison between Logistic Regression and Mantel-Haenszel Procedures (2004, methodology): “In contrast, when the Jodoin and Gierl (2001) criteria were used, 15.”
Efficiency of the Mantel, Generalized Mantel–Haenszel, and Logistic Discriminant Function Analysis Methods in Detecting Differential Item Functioning for Polytomous Items (2005, methodology): “H method (Dorans & Holland, 1993) and the difference in R2 between two nested models for the logistic regression methods (Jodoin & Gierl, 2001; Zumbo, 1999).”
Effects of Anchor Item Methods on the Detection of Differential Item Functioning Within the Family of Rasch Models (2004, methodology): “…studies may be conducted to compare these three anchor item methods and investigate the effects of MIDD (or ASA) using non-IRT procedures (e.g., Jodoin & Gierl, 2001; Miller & Oshima, 1992; Narayanan & Swaminathan, 1994, 1996; Rogers & Swaminathan, 1993; Roussos & Stout, 1996; Shealy & Stout,…”
Effects of Average Signed Area Between Two Item Characteristic Curves and Test Purification Procedures on the DIF Detection via the Mantel-Haenszel Method (2004, methodology): “…IRT-basedornonparametric DIF detection methods in which nonuniform DIF, multiple groups, or polytomous items are compared (Jodoin & Gierl, 2001; Kim & Cohen, 1995; Li & Stout, 1996; Mazor&Clauser,1994;Penfield,2001;Rogers&Swaminathan,1993;Swaminathan & Rogers, 1990; Wang…”
An Approach for Categorizing DIF in Polytomous Items (2007, methodology): “For example, categories of small, medium, and large levels of DIF effect have been proposed using several different DIF effect estimators ( Dorans & Holland, 1993 ; Jodoin & Gierl, 2001 ; Roussos & Stout, 1996 ; Zumbo & Thomas, 1996 ).”
Odds Ratio, Delta, ETS Classification, and Standardization Measures of DIF Magnitude for Binary Logistic Regression (2007, background): “Jodoin and Gierl (2001) suggested that R2-like indices are preferable to effect sizes based on b̂2 because the latter would depend on the coding of the group variable [i.e., reference cell (0/1) vs. deviations-from-means method (−1/1)].”
Does the EDI Measure School Readiness in the Same Way Across Different Groups of Children? (2007, methodology): “For the interpretation of effect sizes from ordinal logistic regression DIF analyses, Jodoin and Gierl (2001) have suggested guidelines.”
Evaluating Guidelines For Test Adaptations (2006, background): “Jodoin and Gierl (2001) stated that R2 effect sizes of .035 and .07 roughly corresponded to the Educational Testing Service’s categories of moderate and large DIF, respectively.”
Differential Item Functioning Results May Change Depending On How An Item Is Scored: An Illustration With The Center For Epidemiologic Studies Depression Scale (2003, methodology): “Building on recommendations made by Zumbo (1999) and Jodoin and Gierl (2001), two criteria must be met for an item to be classified as displaying DIF. First, the two degrees of freedom chi-squared test for DIF (i.e., testing for the gender and interaction effects simultaneously) must have a p value ”