Educating for Democracy: The SAT Controversy: What Do We Really Learn from Tests?

As the head of the honors program at my college for a number of years I had to determine how much validity I would put into the SAT scores in weighing the admissions criteria for a candidate. In making my decision I also included such criteria as gpa, course choice--some students would choose "easy A" courses in their senior year to beef up their average--and the personal essay the student would write. All of the students selected had among the highest SAT scores and gpa's of that freshman class. Not more than half graduated with an honors degree four or five years later.
Of course, there were rigorous standards that included a minimum "B" in any honors course and a "Senior Thesis" supervised by a faculty member in the candidate's area of study (This was one of the factors that discouraged many of the students from completing their honors requirements.). But from what I experienced, there was little correlation between SAT scores and student success in the program. Some of those with the highest SAT scores, in fact, not only dropped out of the program after the first semester but actually flunked out of the college.
In a recent (8/6/10) episode of "The Brian Lehrer Show" on WNYC, Laurence Bunin, senior vice president of the SAT tests and the College Board, defended the charges of a recent study that found the test "racially biased." He was rebutted by several callers and other discussants questioning the validity of the tests and their objectivity. I found the discussions more than vaguely familiar; it seems as if they have been going on for decades and each side has the "data" to support its view. But the issue of the SAT controversy as to whether or not the test is "racially biased" or "culturally biased" is, to me, beside the point. I don't believe that any standardized test should be a significant factor in determining how well a student is learning. I just don't think the instrument of testing which is, of necessity, a generic one, can accurately account for the many factors that go into measuring what, how and why students learn.
One of my colleagues, a German psychologist with whom I regularly correspond and who is very well acquainted with the "metrics mania" of Americans, describes the use of statistical data to determine social and educational policy "KISS"--"Keep It Simple. . . and Stupid." That's what I believe the SAT's and any of the other knowledge or intelligence tests consists of: reducing human complexity into the simplicity of numbers.
Since I began my career as an educator forty-five years ago, I have heard innumerable controversies over the SAT's, the Stanford-Binet "intelligence tests," and my favorite: the "high-stakes testing" that is doing to the educational system in this country what wild kudzu does to gardens. I have heard from many sources that the SAT's are only accurate in predicting the performance of students in their first semester of college. After that, they are no longer valid. To use a sports metaphor, that would be like determining the quality of a starting pitcher solely on the basis of his performance in the first inning.
My own explanation for the unreliability in predicting college student performance on the basis of standardized tests is that once students become acclimated to the environment of an academic setting, their own personalities and basic work ethic will determine how much they will study, how many classes they will actually attend, and how much they will learn. That is why I believe that if colleges really want to know how well students are likely to do they will rely more heavily on interviews, submitted projects and essays (hopefully written by the student and not "edited" by someone else) than tests. Some schools do this. I believe that all should. But that takes a lot of time and money and effort and in a mass society in which privilege and wealth seem to be more and more the determining factors in having a successful life--however that's defined--it is much easier to give out a standardized test and claim that the "objective" scores prove students' intellectual value and, more significantly, provide a likely estimate of the future of their academic and occupational careers.
That about half of college freshmen never receive a degree and that many who do need to be retrained when they get a job (or, in the present economic climate IF they get a job) should point out that there is something profoundly wrong with our educational system that test scores, whether "fair" or "biased," will not be able to address. That is the failure of an economic system that can no longer generate the good-paying jobs that used to be the expectation of a high school graduate let alone a college graduate. Unless people begin to look outside the narrow way in which education is conventionally viewed and see it as becoming the product of a rapidly developing elitist society, no tests will find an answer to our problems of improving education which can only be done by addressing many factors: economically, culturally, socially and intellectually.