In this review of research studies on differences between HispanicAmerican and White, non-Hispanic groups on the Scholastic Aptitude Test (SAT), four questions are addressed: (1) What factors are associated with ethnic group mean differences on the SAT? (2) What types of item format or content are identified as being differentially easier or more difficult for Hispanic vs. White, non-Hispanic students? (3) How accurately do selective admissions tests predict the performance of Hispanic students in college? and (4) Do Hispanic students have equal access to information necessary for long-term and short-term preparation for selective admissions tests?
The Status of Research on Selective Admissions Tests and Hispanic Students in Postsecondary EducationIn this paper, studies evaluating the validity of the Scholastic Aptitude Test (SAT) for use in college admissions of Hispanic students will be reviewed. Other tests for admissions to graduate or professional schools are not considered here because small sample sizes and other practical problems limit the number of studies on Hispanic vs. non-Hispanic White group differences in validity. Before beginning, it is worthwhile to repeat the cautions in Linn's (1982) preface to his review of group differences in test validity.The controversies over testing are neither created by, nor will they be resolved by, the results of investigations of test validity (Cronbach, 1975) ... Justification of test use obviously depends upon much more than [how well an ability test predicts academic or professional performance]. Potential benefits and losses for the individual, the institution, and the society at large need to be considered, and the relative importance of the benefits and losses can be expected to vary greatly in the eyes of these various interests. Nonetheless, information about the degree of relationship of test scores to particular criterion measures and about the degree to which the observed relationship is generalizable across situations and from one situation to another is an important component in the evaluation of the use of tests ... (pp. 335-336).In this presentation, I have taken a broad view of test validity, going beyond considerations of how well tests predict undergraduate grades.Group differences are reviewed in terms of the relationship between test scores and other educational and demographic variables, evaluations of testitem content and format, and the availability of college-admissions counseling and guidance information. Discussion of the studies will be organized around the following questions.(1) Mean Differences. What factors are associated with ethnic group mean differences on the SAT? What are the implications of these differences?