ABSTRACT
Large-scale educational assessments have been criticized for a variety of reasons over the years, and one of the most enduring and oft-repeated criticisms is that there are consistent and substantial score differences between minority and nonminority test takers (Bronner, 1997; Jencks & Phillips, 1998; Sacks, 1997). Persistent score differences among racial groups have been very troubling and have led to charges of test bias. However, it should not be surprising that, when individuals are grouped in various ways that are related to differential educational opportunities, these groups score differently on tests such as the SATI: Reasoning Test (SAT I; Widaman, 1998).
