This special issue includes seven articles concerning strategies for overcoming barriers in standardized testing as it pertains to African Americans. Most of the articles focus on the effects of testing in educational settings or on related educational experiences of test takers. Three of the articles focus on alternatives to standardized testing for students with special or exceptional needs (gifted college students or students with learning disabilities or emotional-behavior disorders). Two focus on historical and recent efforts to overcome the limiting effects of testing on employment opportunities for military personnel and potential teachers, respectively. The remaining two articles search for explanations for why policy mandates concerning testing have had such negative implications for the educational experiences of African Americans. Collectively, one message threads through the articles: testing innovation does not seem to have changed the nature of test usage, and African Americans of all ages in educational and vocational settings continue to be harmed by assessments. Debates about test bias and fairness are on-going, particularly in this era of high-stakes testing. At the same time, the testing industry continues to assert that newly created tests and revisions of old assessment tools are culturally sensitive. Why then do supposedly 'unbiased' tests remain unfair?
Keywords: standardized testing, African Americans, test bias, test fairness
As we write this introduction to the special issue of The Journal of Negro Education on "Testing and Assessing of African Americans: Past, Present, and Future Problems and Promises," the Teachers' Union in the Chicago public school district, the third largest district in the United States, is on strike. The primary catalyst for the strike was city administrators' insistence that teachers' merit pay and continued employment be tied to evaluations of their performance, defined in large part as improving students' scores on (presumably) standardized tests of cognitive abilities, intelligence, and academic skills. To date, school districts across the country are embroiled in controversy over the appropriateness of using students' test scores as an indicator of the quality of teachers' performance, a clear violation of good testing practices (Rothstein, Ladd, Ravitch, Baker, et al., 2010). Also questionable and objectionable is the use of test scores to determine African Americans' potential when issues of test bias and unfairness abound.
Although the circumstance of using someone else's test scores (e.g., students) to evaluate another person's (e.g., teachers) knowledge, skills, or abilities is peculiar to public school educational systems, the use of people's own tests to determine life options is pervasive in U.S. society. Therefore, to cite just a few examples, some of which the contributors address, test scores are used to determine whether: (a) students will be promoted to the next grade level; (b) students receive high school diplomas; (c) students receive special education or gifted education services; (d) applicants are admitted to institutions of higher education; (e) potential workers are selected for certain occupations or job openings, including those who have successfully graduated from college with teaching degrees; (f) employees are promoted once they enter the workforce; and (g) university graduates are permitted to practice in their areas of expertise, even if they have successfully completed the requisite education.
To the layperson, use of test scores in any of these individual situations might seem reasonable. That is, the widespread acceptance of tests in such diverse roles is predicated on the assumption that tests objectively measure something of importance about the person, such as cognitive abilities, academic abilities, and employment skills that generalize to or will be useful in other situations. This presumption is particularly problematic for African Americans across the lifespan because historically, as a group, they virtually never have obtained scores as high as their White counterparts in the same settings (see Helms, 2008). …