ERIC Number: ED395998
Record Type: RIE
Publication Date: 1996-Apr
Reference Count: N/A
Detecting DIF on Mathematics Items: The Case for Gender and Calculator Sensitivity.
Ryan, Katherine E.; Chiu, Shuwan
The use of scientific calculators on standardized mathematics tests is becoming more common, and how their use affects the standardized testing process continues to be a topic of study. This study extends previous work by examining test item characteristics and equity issues, asking whether items designed to be "calculator neutral" function as intended, and whether such items function the same for males and females. Data came from trials of the Midwestern Mathematics Placement Examination for precalculus and calculus classes for over 1,000 college students. A logistic regression analysis was conducted to detect differential item functioning (DIF) or nonuniform DIF. Findings suggest that calculator neutral items can be constructed, and that the items that were constructed to be calculator neutral were largely free from gender DIF. Results also showed that the logistic regression approach was a useful addition to studying DIF, although larger sample sizes would have been highly preferable. Using the rule-space approach as a basis for the content analyses was also quite effective. An appendix contains two sample problems and a description of the mathematical challenge items. (Contains 4 tables, 4 figures, and 17 references.) (SLD)
Publication Type: Reports - Research; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A
Identifiers: Item Bias Detection; Rule Space Model
Note: Paper presented at the Annual Meeting of the American Educational Research Association (New York, NY, April 8-12, 1996).