from Women's Caucus for Political Science Quarterly, Vol. 3, No. 1 (October 1976), 12-15.
Who could possibly be against "quality education?" What we can and should oppose, however, is a definition of quality that rests on the racist and elitist notion that quality means selectivity, the exclusion of a large proportion of the population from the colleges. The recent and impending cutbacks in higher education are justified by rhetoric which claims that by eliminating students who do poorly on standardized tests, the worth of the colleges will improve. The SAT is the most widely used measure of so-called college "aptitude." It is important, then, to examine some assumptions behind these tests and some issues they raise.
The validity of the SAT should first be questioned. What are the tests supposed to measure and to what extent do they do so? After asserting that the SAT is an indicator merely of a "potential skills deficiency," a clearly tentative phrase, Fullilove (1976) in a recent report of the New Jersey Department of Higher Education (DHE) goes on to claim that the SAT is a reliable measure of the talents, skills and abilities which students acquire by virtue of attending elementary and secondary schools in this country..." (p. 4) Donlong and Angoff (1971) disagree and assume that the tests measure skills which ". . . grow slowly over the years through interaction with the student's total environment. They are assumed to be part of every student's equipment, and are relatively independent of what he is currently learning in the normal classroom curriculum." (p. 15) Whatever the tests ostensibly measure, the scores reflect the inequities that exist in our school systems and in our society.
It is often argued that the test scores are useful for their predictive value; that is, it is claimed that they do a good job of predicting success in college.
"The SAT has been demonstrated to be an effective tool for measuring the skills required for college study, and its widespread use as a predictor of college success (along with other measures of a student's high school performance) is well documented." (Fullilove, 1976, p. 4)So where is the documentation? The fact that the SAT is used to predict college success does not prove that it is an "effective tool" for that purpose. Jencks and Riesman (1969) bury the low correlation in a footnote:
"The multiple correlation of high school grade and aptitude scores with college grades usually runs between 0.5 and 0.6, which means that these two predictors can account for about a third of the variation in students' college grades. This is not very good, but it is much better than any other known forecasting device." (p. 124n, emphasis added)
One may wonder why a forecasting device is needed at all. The scores may even have some predictive value for the students who attend college, but how do we know how those millions who are eliminated by the tests might have done in college? Astin (1971) reports a study from the University Research Corporation which found that those "disadvantaged" students admitted in special programs (that is, in spite of low SAT scores) had a higher retention rate than college students in general. The glowing reports of success in New Jersey's Educational Opportunity Fund program argue against reliance on SAT scores: "In spite of the fact that E.O.F. students come to college with lower SAT scores than their regularly admitted counterparts, they quickly close the gap." (The Educational Opportunity Fund Fourth Annual Report, 73-74, 1974, p. 8)
The SAT (as is the case with all standardized tests) is based on the assumption that so-called "scholastic aptitude" is something that is distributed normally in the test-taking population. The raw test scores are standardized so that most scores cluster around the mean with few exceptionally high or low scores. This assumption serves to perpetuate a stratified society. Only a few students are rated good enough for Harvard, for high-paying jobs. The tests exist to tell most of the student population that they do not measure up. It could be assumed that most students are perfectly capable of benefitting from higher education. Lane (1971) quotes Bloom:
"We are expressing the view that, given sufficient time (and appropriate types of help), 95% can learn a subject up to a high level of mastery. We are convinced that the grade of 'A' as an index of mastery of a subject can, under appropriate conditions , be achieved by up to 95% of the students in a class." (p. 103-104)But, if all can be educated, then how can one justify the limited availability of higher education? The SAT provides the rationale for rationing access, and those students who are poor or from minority groups are those who are denied admission in increasing proportions.
When the class and racial biases of the tests are taken into account, the myth of equal opportunity is even more clearly exposed.
"The lower one's family income, the lower the SAT score is likely to be. The mean score of New Jersey students for families earning less than $6,000 annually in 1974-1975 is 79 points lower than the mean of students from families earning $15,000 or more per year." (Fullilove, 1976, p. 13)
Minorities and women also have what the DHE report calls "a propensity for earning low scores". (p. 11) It must be recognized that lower average scores for poor, minority and female students persist because of the assumptions underlying the tests. In standardizing a test, the test-maker often eliminates items which discriminate between subgroups of the test-taking population who are assumed to be equally able. So, for example, when it was determined that women scored higher on other notorious standardized tests, the IQ tests, the tests were restandardized to eliminate sex differences in performance on the tests. There the assumption was that men were as "intelligent" as women and so should score equally on the average. (One might well wonder whether the same restandardization would have taken place had men scored higher than women!)
Jencks and Riesman (1969), again in a footnote, dismiss the idea of a "culture-free" test because they assert, first, that such a test is "virtually inconceivable" (They proceed by outlining exactly how such a test might be constructed), and second, that it would be "virtually useless."
"Life is not culture-free, and tests restricted to those aspects of life that are culture-free would measure little of importance. The point can be illustrated by the following example. Suppose we took a conventional test of academic aptitude and scored it separately for various subcultures. Suppose we then standardized the scores so that there were no differences in group averages. Both Negroes and whites, e.g., would be assigned median IQ's of 100; two-thirds of both groups would be assigned scores between 85 and 115; and so forth. The same could be done for urban and rural children, for rich and poor children, and so on. When a child took the test his raw score would then be translated into a standard score, with the formula varying with his race, parental income, geographic origin, or whatever. The result would be a 'culture-free' system for ranking students from various backgrounds without making invidious group comparisons..,, The point is, of course, that a test that suggests that youngsters from poor families are as competent as youngsters from affluent families is misleading and of little value to anyone." (p. 123n, last emphasis added)
Of little value to anyone? Such a test would be of great value to most working people in this country. A test that suggests that youngsters from poor families and minority families are as competent as youngsters from affluent families would draw the blame for inequities in our society away from the victims of those inequities and place it squarely on those who control our institutions. An oppressed population that believes itself to be as capable as those who occupy the limited positions of privilege may well turn the world upside down.
Astin, H.S. Response to Edmund W. Gordon's paper. In Barriers to Higher Education. New York: College Entrance Examination Board, 1971.
Clark, K.B. Response to Julian C. Stanley's paper. In Barriers to Higher Education. New York: College Entrance Examination Board, 1971.
Crossland, F.E. Minority Access to College. New York: Schocken Books, 1971.
Donlong, T.F. & Angoff, W.H. The scholastic aptitude test. In The College Board Admissions Testing Program. Princeton, 1971.
Education Opportunity Fund Fourth Annual Report, 73-74. Trenton: New Jersey Department of Higher Education, 1974.
Fullilove, R.E. III. The SAT Score Decline and Its Impact on Admissions in New Jersey Public Colleges. Trenton: New Jersey Department of Higher Education, 1976.
Jencks, C. & Riesman, D. The Academic Revolution. New York: Doubleday Anchor Books, 1969.
Lane, H.W. Response to Winton H. Manning's paper. In Barriers to Higher Education. New York: College Entrance Examination Board, 1971.
Manning, W.H. Personal and Institutional Assessment: Alternatives to Tests of Scholastic Aptitude and Achievement in the Admissions Process. In Barriers to Higher Education. New York: College Entrance Examination Board, 1971.
Stanley, J.C. Predicting College Success of Educationally Disadvantaged Students. In Barriers to Higher Education, New York: College Entrance Examination Board, 1971.
http://www.shss.montclair.edu/english/furr/weilandart.html | email@example.com | last modified 27 Jan 97