手法を比較

	Differential Distractor Functioning	Differential Item Functioning in Educational Testing
分野	Education	Education
系統	Latent structure	Latent structure
提唱年	2008	1993
提唱者	Item-bias methodology (Green, Crone & Folk; Penfield)	Educational measurement / test-fairness tradition (Holland, Wainer, Dorans, Thissen)
種類	Group-difference analysis of the incorrect options (distractors) of multiple-choice items	Test-fairness analysis detecting items that function differently across groups
原典	Penfield, R. D. (2008). An odds ratio approach for assessing differential distractor functioning effects under the nominal response model. Journal of Educational Measurement, 45(3), 247–269.	Holland, P. W., & Wainer, H. (Eds.). (1993). Differential Item Functioning. Lawrence Erlbaum Associates. ISBN: 9780805809725
別名	DDF, Distractor-Level DIF, Differential Option Functioning, Distractor Functioning Analysis	Educational DIF Analysis, Item Bias Detection in Tests, Test Fairness DIF, Mantel-Haenszel DIF
関連	4	4
概要	Differential distractor functioning (DDF) extends test-fairness analysis from the correct answer to the wrong ones. It asks whether examinees of equal ability but different group membership are differentially attracted to particular distractors (incorrect options) of a multiple-choice item. By analyzing option-level rather than just right/wrong responses, DDF can detect bias that ordinary differential item functioning misses and, crucially, help explain why an item functions differently — pointing to the specific wrong option luring one group. Penfield's odds-ratio approach under the nominal response model is a standard tool.	Differential item functioning (DIF) analysis is the central statistical tool for evaluating the fairness of test items in education. An item shows DIF when examinees of equal ability but different group membership — for example by gender, race/ethnicity, or language background — have unequal probabilities of answering it correctly. By conditioning on ability before comparing groups, DIF analysis separates genuine item bias from real group differences in proficiency, and flags items for expert review before they affect high-stakes decisions.