skip to main content skip to footer

An Investigation of the Simultaneous Moderation of Average Gender and African-American Score Differences on a Test of Mathematical Reasoning

Author(s):
Stocking, Martha L.; Jirele, Thomas J.; Lewis, Charles; Swanson, Len
Publication Year:
1998
Report Number:
RR-98-46
Source:
ETS Research Report
Document Type:
Report
Page Count:
66
Subject/Key Words:
Test Construction, Expert Systems, Sex Bias, Automation, Scores, Blacks, Test Bias, Mathematics Tests

Abstract

A pool of items from operational tests of mathematical reasoning was constructed to investigate the feasibility of using automated test assembly methods to simultaneously moderate possibly irrelevant differences between performance of women and men and African-American and White test takers. None of artificial tests investigated exhibited substantial impact moderation, although the estimated mean scaled score differences for the relevant population indicated a modest move into intended direction: the difference between scaled score means was reduced by about 20% for women and men and about 9% for African-American and White test takers. Although many issues and implementation of this methodology remains to be solved, the consideration of impact in automated test assembly along with the maintenance of the detailed test plan appears to be a potential method of moderating possibly irrelevant mean test score differences.

Read More