Public Release: 

'Why test?'--study assesses student intelligence testing

American Sociological Association

Chicago, IL - Assessing the intelligence of students for the purpose of college admissions is difficult and controversial. On the one hand, the higher overall performance by whites on tests raises questions about whether they are fair or not. Tests in the early grades, for example, often advantage whites and magnify racial differences at later ages. For these and other reasons, including differences in the quality of schools, blacks generally have lower scores on assessment tests.

Sociologists Judith Blau and Stephanie Moller of the University of North Carolina at Chapel Hill will present these and other findings from their research on student testing in a panel at the Annual Meeting of the American Sociological Association in Chicago on Monday August 19th. Their study examines racial divisions in assessment test scores and student outcomes using data from the National Education Longitudinal Study (NELS).

Blau and Moller assert that current assessment practices are unfair to both black and white students. Blacks are disadvantaged because testing is part of a system that has been and remains discriminatory and because test scores are used unfairly. There is also evidence that blacks view testing differently than whites--and are therefore less motivated than white students to perform well on them.

The authors, however, also suggest that tests are not fair to white youth either. Research studies suggest that, in white culture, there is a particular seriousness about individual performance and competition, and an exaggerated belief in the validity of competitive tests and the rankings they produce. Current assessment practices reinforce the notion that talent can be measured along a single continuous dimension, with the consequence that many qualified white students do not apply to college.

Test scores analyzed in context of other variables (educational aspirations, probabilities of attending a post secondary institution, different college programs, socioeconomic status) reveal that:

    · High-scoring blacks are less likely than high-scoring whites to pursue higher education, in part because blacks attend less good schools;
    · Low-scoring whites are less likely than low-scoring blacks to pursue higher education;
    · Blacks are more likely to pursue higher education, with controls for SES and test scores; and
    · Blacks' educational aspirations exceed those of whites.

Blau and Moller, therefore, pose the question: If tests are unfair to white youth and to youth of color, why test?

They conclude that using high-school class rank, instead of test score, yields fairer race and gender outcomes when assessing students for college admissions. Both blacks and males benefit if class rank is used. The pool of black students is larger when class rank is used as a measure of talent, and educational returns for blacks are fairer when based on class rank than test scores. Males also benefit as their pool is enlarged by class ranks. Since the mid-1980s, male enrollment rates in degree-granting institutions have been dropping as institutions increasingly favored women over men.


The Annual Meeting of the American Sociological Association is being held from August 16-19 at the Chicago Hilton and Hilton Palmer House Hotels in Chicago, IL. The American Sociological Association is a non-profit membership organization dedicated to advancing sociology as a scientific discipline and profession, serving the public good. The purpose of the Annual Meeting is to meet the scholarly, teaching, training and practice needs of sociologists and social scientists at every career stage.

Journalists are invited to attend any of the more than 500 presentations, book panels, poster sessions and symposia featured on a wide range of topics at the Annual Meeting. Press facilities are located in Room PDR 1 of the Hilton Chicago Hotel.

Additional Contact Information: Lee Herring (202) 383-9005 ext. 332 or (312) 294-6783 (Chicago)

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.