Considering Large Group Differences in Ability in DIF Analysis

Differential item functioning (DIF) occurs when an item has a different probability of endorsement for different groups of respondents who are equivalent on the measure. When groups are not equivalent, matching of respondents on ability or other method of controlling for group differences is necessary. Nevertheless, there is evidence that some DIF procedures may result in increased false positive DIF results when groups differ in ability (DeMars, 2010; Li & Stout, 1996). Previous studies have generally focused on the impact of small to moderate (e.g., 0.5 to 1.0 logits) group differences. What effect do larger group differences have on DIF results?

A series of Monte Carlo simulations were performed to answer this question. Dichotomous item responses to a 25-item instrument were simulated according to the Rasch model for two groups of simulees (n = 250 or 500 per group). In each dataset, a common set of item calibrations derived from a uniform random distribution were used for both groups (i.e., no DIF was simulated). Group means for both reference and focal groups ranged from -1.5 to 1.5 logits in increments of 0.5 logits (group standard deviations = 1.0). For each combination of sample size, focal group mean, and reference group mean, 100 datasets were generated. False-positive DIF using both t-test comparisons of item calibrations and the Mantel-Haenszel (M-H) test was based on statistical significance at the .05 level. Simulations were performed using R (version 2.13), the rWinsteps package (version 1.01) and Winsteps (version 3.72.3).

The figure shows the false positive DIF rates for each method by focal group - reference group differences in ability and sample size. Both methods resulted in false positive DIF rates that are generally within the nominal .05 level, with M-H having lower error rates, particularly with absolute group differences = 2 logits. Only with the t-test procedure, when the number of respondents per group equaled 500 and when absolute group differences were = 2.5 logits did the false positive rate exceed 5 percent.

Conclusions: These results suggest that DIF methods commonly used in conjunction with Rasch measurement are robust against large differences in group ability. The Mantel-Haenszel procedure resulted in a lower false positive rate and may be the more appropriate method when group differences and sample sizes are large. A limitation of the analysis is that power in detecting true DIF was not assessed.

Barth Riley

DeMars, C. E. (2010). Type I error inflation for detecting DIF in the presence of impact. Educational and Psychological Measurement, 70(6), 961-972.

Li, H.-H., & Stout, W. F. (1996). A new procedure for detection of crossing DIF. Psychometrika, 61(4), 647-677.

Figure. False Positive DIF Rate by Group Mean Difference and Sample Size.

Considering Large Group Differences in Ability in DIF Analysis, Barth Riley ... Rasch Measurement Transactions, 2011, 251:2, 1326

Rasch Publications
Rasch Measurement Transactions (free, online) Rasch Measurement research papers (free, online) Probabilistic Models for Some Intelligence and Attainment Tests, Georg Rasch Applying the Rasch Model 3rd. Ed., Bond & Fox Best Test Design, Wright & Stone
Rating Scale Analysis, Wright & Masters Introduction to Rasch Measurement, E. Smith & R. Smith Introduction to Many-Facet Rasch Measurement, Thomas Eckes Invariant Measurement: Using Rasch Models in the Social, Behavioral, and Health Sciences, George Engelhard, Jr. Statistical Analyses for Language Testers, Rita Green
Rasch Models: Foundations, Recent Developments, and Applications, Fischer & Molenaar Journal of Applied Measurement Rasch models for measurement, David Andrich Constructing Measures, Mark Wilson Rasch Analysis in the Human Sciences, Boone, Stave, Yale
in Spanish: Análisis de Rasch para todos, Agustín Tristán Mediciones, Posicionamientos y Diagnósticos Competitivos, Juan Ramón Oreja Rodríguez

To be emailed about new material on
please enter your email address here:

I want to Subscribe: & click below
I want to Unsubscribe: & click below

Please set your SPAM filter to accept emails from welcomes your comments:

Your email address (if you want us to reply):


ForumRasch Measurement Forum to discuss any Rasch-related topic

Go to Top of Page
Go to index of all Rasch Measurement Transactions
AERA members: Join the Rasch Measurement SIG and receive the printed version of RMT
Some back issues of RMT are available as bound volumes
Subscribe to Journal of Applied Measurement

Go to Institute for Objective Measurement Home Page. The Rasch Measurement SIG (AERA) thanks the Institute for Objective Measurement for inviting the publication of Rasch Measurement Transactions on the Institute's website,

Coming Rasch-related Events
Oct. 6 - Nov. 3, 2023, Fri.-Fri. On-line workshop: Rasch Measurement - Core Topics (E. Smith, Facets),
Oct. 12, 2023, Thursday 5 to 7 pm Colombian timeOn-line workshop: Deconstruyendo el concepto de validez y Discusiones sobre estimaciones de confiabilidad SICAPSI (J. Escobar, C.Pardo)
June 12 - 14, 2024, Wed.-Fri. 1st Scandinavian Applied Measurement Conference, Kristianstad University, Kristianstad, Sweden
Aug. 9 - Sept. 6, 2024, Fri.-Fri. On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets),


The URL of this page is