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.
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 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|
|Forum||Rasch 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, www.rasch.org.
|Coming Rasch-related Events|
|June 29 - July 27, 2018, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Further Topics (E. Smith, Winsteps), www.statistics.com|
|July 25 - July 27, 2018, Wed.-Fri.||Pacific-Rim Objective Measurement Symposium (PROMS), (Preconference workshops July 23-24, 2018) Fudan University, Shanghai, China "Applying Rasch Measurement in Language Assessment and across the Human Sciences", www.promsociety.org|
|July 29 - August 4, 2018||Vth International Summer School `Applied Psychometrics in Psychology and Education`, Institute of Education at the Higher School of Economics, St. Petersburg, Russia, https://ioe.hse.ru/en/announcements/215681182.html|
|July 30 - Nov., 2018||Online Introduction to Classical and Rasch Measurement Theories (D.Andrich), University of Western Australia, Perth, Australia, http://www.education.uwa.edu.au/ppl/courses|
|Aug. 10 - Sept. 7, 2018, Fri.-Fri.||On-line workshop: Many-Facet Rasch Measurement (E. Smith, Facets), www.statistics.com|
|August 25 - 28, 2018, Sat.-Tue.||Análisis de Rasch introductorio (en español). (Agustín Tristán), Instituto de Evaluación e Ingeniería Avanzada. San Luis Potosí, México. www.ieia.com.mx|
|Sept. 3 - 6, 2018, Mon.-Thurs.||IMEKO World Congress, Belfast, Northern Ireland, www.imeko2018.org|
|Oct. 12 - Nov. 9, 2018, Fri.-Fri.||On-line workshop: Practical Rasch Measurement - Core Topics (E. Smith, Winsteps), www.statistics.com|
The URL of this page is www.rasch.org/rmt/rmt252d.htm