NotesFAQContact Us
Search Tips
ERIC Number: ED171801
Record Type: RIE
Publication Date: 1979-Apr-9
Pages: 14
Abstractor: N/A
Reference Count: 0
Empirical Methods for Studying Sex Bias in Achievement Test Items.
Brown, F. G.; Moss, Jacque
Two studies were conducted of the effects of sex biased items on achievement test scores obtained by male and female college students. Sex bias was defined as a disproportionate number of references to one sex. Study I used 1,008 classroom test items. Items were classified as male sex referent, female sex referent, no sex referent (neutral), or ambiguous sex referent (other). Data showed a small, but statistically significant, sex bias effect, whereby students performed better on items containing referents to their own sex than on other classes of items. Study II was experimental and controlled for sex bias by using two forms of a 3,000-word passage on defense mechanisms and two tests: one with predominantly male sex referents and the other with predominantly female sex referents. Half of the students took a test having the same predominant sex referents as the passage they had read; the other half took a test with the opposite sex referents predominating. There were no statistically significant effects. The results of the two studies suggest that sex bias in test items has in actuality, minimal effects on test scores. (Author/MH)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A
Note: Paper presented at the Annual Meeting of the National Council on Measurement in Education (San Francisco, California, April, 1979)