ERIC Number: ED177210
Record Type: RIE
Publication Date: 1979-Apr
Instrumentation for Assessing Culturally Different: Some Conceptual Issues.
Rosenbach, John H.
Since their beginning, intelligence tests have favored the higher social classes. Despite federal mandates to the contrary, bias in assessment is likely to remain a problem. Claims of test bias can be categorized as popular (naive); clinical (intuitive); statistical (predictive); and psychometric (construct and content). A literature review has led the author to believe that most claims of test bias are unfounded, especially for intelligence and aptitude tests which are used to predict success in school. Since most of these tests are valid and remarkably accurate in their predictions, the bias must be associated with the nature of what is being predicted (the Criteria). Although tests are accurate predictors of school achievement, success in school is not very predictive of success outside of school. Alternative solutions to this issue are to explore aptitude-treatment-interaction models which may be used to improve the test performance of minority group students, or to modify our expectations of what should be learned in school. Before the problems of test bias can be resolved, the sources of bias must be objectively determined and the purposes of testing must be understood. (GDC)
Descriptors: Academic Achievement, Aptitude Tests, Cultural Differences, Culture Fair Tests, Elementary Secondary Education, Higher Education, Intelligence Differences, Intelligence Tests, Predictive Validity, Racial Differences, Relevance (Education), Social Differences, Test Bias, Testing Problems
Publication Type: Speeches/Meeting Papers; Opinion Papers
Education Level: 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 9-11, 1979)