Item Bias in Individually Administered Ability Tests: Implications for Nondiscriminatory Assessment

Abstract

The study reviewed recent research investigating test-item bias in frequently used individually administered ability tests. An examination of relevant literature suggested that only a limited number of studies in recent years have examined the item-bias issue involving the use of individually administered ability tests. Six methodologies for identifying biased items are presented: (1) measures of item difficulty, (2) analysis of variance procedures, (3) item response theory, (4) chi-square procedures, (5) log-linear models, and (6) judgmental analysis. Implications of the research for assessment practice are examined.