Abstract
For every IRT model, a mathematical function is used to specify the probability of item responses as a function of the ability (latent trait). The degree to which the observed data fit the mathematical function needs to be examined since valid results can only be drawn if the data fit the model.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Colvin KF, Randall J (2011) A review of recent findings on DIF analysis techniques (Center for Educational Assessment Research Report No. 795). University of Massachusetts, Amherst, Center for Educational Assessment, Amherst, MA
Dorans NJ, Holland PW (1993) DIF detection and description: Mantel-Haenszel and standardization. In: Holland PW, Wainer H (eds) Differential item functioning. Lawrence Erlbaum Associates, Hillsdale, NJ, pp 35–66
Holland PW, Thayer DT (1988) Differential item performance and the Mantel-Haenszel procedure. In: Wainer H, Braun HI (eds) Test validity. Lawrence Erlbaum Associates, Hillsdale, NJ, pp 129–145
Holland PW, Wainer H (1993) Differential item functioning. Lawrence Erlbaum, Hillsdale, NJ
Mantel N, Haenszel W (1959) Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22:719–748
Osterlind SJ, Everson HT (2009) Differential item functioning. Sage Publishing, Thousand Oaks, CA
Wu ML (2010) Measurement, sampling and equating errors in large-scale assessments. Educ Measur Issues Pract 29(4):15–27
Zumbo BD (2007) Three generations of differential item functioning (DIF) analyses: considering where it has been, where it is now, and where it is going. Lang Assess Q 4:223–233
Zwick R (2012) A review of ETS differential item functioning assessment procedures: flagging rules, minimum sample size requirements, and criterion refinement (ETS Research Report No. RR-86-31). ETS
Author information
Authors and Affiliations
Corresponding author
Appendices
Hands on Practise
-
(1)
The following table shows the percentages correct of PISA 2009 mathematics items for Shanghai and Australia (Table 11.4).
The following shows a plot of these percentages correct.
Given the information provided in the table and the graph, discuss the relative performances of Shanghai and Australian students, and give your impressions of the presence/absence of DIF.
Discussion Points
-
(1)
Why is it that the existence of DIF items would violate the assumptions of the IRT model applied to the test?
-
(2)
Do you agree that the ability estimates will not change a great deal when DIF items are retained? Why or why not?
Exercises
-
Q1.
Indicate whether you agree or disagree with each of the following statements
A statistical DIF analysis did not detect any DIF item in a test. We can be assured then the test is a fair test for the groups of respondents for which the DIF analysis was performed | Agree/disagree |
To determine if a test is a fair test, count the number of DIF items in a test using a statistical DIF detection procedure. If there are considerably more items favouring one group than the other group, then the test is biased | Agree/disagree |
A test is biased if the average performance of one group of respondents is considerably higher than for other groups | Agree/disagree |
When sample size increases, the magnitudes of DIF (difference in item difficulties for two groups of respondents) will tend to increase | Agree/disagree |
When sample size increases, more items will tend to show statistically significant DIF estimates | Agree/disagree |
Rights and permissions
Copyright information
© 2016 Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Wu, M., Tam, H.P., Jen, TH. (2016). Differential Item Function. In: Educational Measurement for Applied Researchers. Springer, Singapore. https://doi.org/10.1007/978-981-10-3302-5_11
Download citation
DOI: https://doi.org/10.1007/978-981-10-3302-5_11
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-3300-1
Online ISBN: 978-981-10-3302-5
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)