The standardized tests used by schools and employers for placement, admissions, and hiring are not intrinsically biased against minority groups, a panel of experts has reported after studying such tests over a four-year period.
The tests are “important predictive tools” that are equally valid for whites and blacks, according to the panel, whose analysis was conducted under the auspices of the National Academy of Sciences (nas).
Nonetheless, panelists warned at a symposium held last week by the academy to announce results of the study, test scores alone should not be used to decide whether a child is placed in a special-education class, a high-school senior receives a diploma, or an applicant gets into college. And they warned that standardized tests have “inherent limitations and some potential for misuse.”
The latter point is also made in a new Ford Foundation-sponsored study of the experience of minority groups in higher education over the past 15 years.
‘Opportunity to Achieve’
Released earlier this month, that study argues that educators have “produced testing procedures and used test criteria for admission to college without asking whether all our citizens had been equally provided with the opportunity to achieve.” Colleges should refrain from using such tests to rank students against each other, the study argues, and should instead evaluate them “on the basis of their potential for learning and growth.”
The National Academy of Sciences panelists acknowledged that test scores may, in fact, reflect social inequities and discrimination. That, however, is a problem that will not be corrected by writing new tests, the panel said. “Tests should not be required to do things they cannot do,” the report says, “such as guarantee that distributions of scores will not differ for different racial or ethnic groups.”
“Research evidence does not support the notion that tests systematically underpredict the performances of minority-group members,” said Lyle V. Jones, a psychologist from the University of North Carolina who served on the panel. But he noted that while studies have shown that the tests predict performance equally well for blacks and whites, not enough data are yet available on members of other minority groups to judge whether the same holds true for these groups.
Echoing the authors of the Ford Foundation report, James McGhee, director of research for the National Urban League and a symposium participant, said that the key problem with standardized tests is the use to which the scores are put.
“We need more emphasis on how the tests are used,” he said. “There are a number of problems in the standards of selection used after the tests,” he said, but “at least tests provide us with an objective measure. The alternative to testing in many cases may be worse.”
New Methods Suggested
Mr. McGhee suggested developing new methods to combat some of the problems.
Mr. Jones noted that the panel made a “strong distinction” between the characteristics of tests and the ways that the scores are used. “No matter how scientifically valid a test may be,” he said in a prepared statement presented at the symposium, “it may have effects that our society finds unacceptable.”
The panel seemed to differ with the Ford Foundation researchers on the issue of how important the tests are in the college admissions process.
The nas group suggested that since standardized tests such as the Scholastic Aptitude Test frequently play only a minor role in deciding whether a student is admitted to college, higher-education officials in all but the most selective institutions should “re-examine whether those tests are actually necessary.”
The nas report, “Ability Testing: Uses, Consequences, and Controversies,” was described by virtually all of the speakers at the symposium as balanced, thorough, and fair.
Gregory R. Anrig, the president of the Educational Testing Service, called it a “constructive and scholarly resource for those concerned with fairness in testing and proper test use in education, government, and the business world.”
The panel made note of two types of test use in elementary and secondary education that have potentially serious effects on the “life chances” of individuals: in special education and in assessments of “competency.”
In neither case should test scores be the sole basis for a decision on placement or to withhold a diploma, according to the report. In special education, “Patterning of test scores can be suggestive to the sophisticated interpreter, but it should not be the basis for an automatic decision formula,” according to the report.
Competency tests, particularly when used as a criterion for granting a high-school diploma, must also be used judiciously, the panel warns.
When used as an “instrument of accountability,” such testing may involve sanctions if the test score reveals inadequacies.
However, the report notes, “Where sanctions are involved, they are generally imposed, not on those who control the quality of instruction--teachers, principals, school district administrators, state legislators, and education officials--but on students, who are denied a high school diploma.”
Consequently, the panel recommends, competency tests should be given early, in plenty of time so that remedial instruction can be provided if necessary. Otherwise, “the end result of the movement could well be to make those who fail the tests less able to make their way in the world than they otherwise would be,” the report says. (The panel’s report does not cover the testing of teachers.)
“The primary message of our report is a call for balance,” Mr. Jones said. “Ability tests should not be viewed as a panacea for deep-seated social ills or as a scapegoat for society’s ills.”
The report is available in two volumes, Part I: Report of the Committee ($13.95) and Part II: Documentation Section ($24.95) from the National Academy Press, 2101 Constitution Ave. N.W., Washington, D.C.