ERIC Number: ED161930
Record Type: RIE
Publication Date: 1978-Mar
Reference Count: 0
A Methodology for Identifying Biased Achievement Test Items that Removes the Confounding in an Items by Groups Interaction Due to Possible Group Differences in Instructional Level.
Plake, Barbara S.; Hoover, H. D.
A method of investigating for possible bias in test items is proposed that uses analysis of variance for item data based on groups that have been selected to have identical test score distributions. The item data used are arcsin transformations of item difficulties. The methodological procedure has the following advantages: (1) The arcsin transformation stabilizes the variance within replications. (2) Since the groups are formed to have equal distributions of raw scores, there is no achievement difference between the groups. (3) Since the selection procedure allows for consideration of the entire raw score distribution, a range of raw scores is represented within each group. (4) The use of variance components allows for an investigation of the practical significance of the variance components; and (5) The follow-up procedure is in the same units as the analysis of variance, provides Type I error control, and investigates the items by contributing to a significant items-by-group interaction. (Author/CTM)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the American Educational Research Association (62nd, Toronto, Ontario, Canada, March 27-31, 1978)