 |
|
 |
|
 |
|
 |
|||||||||||||||||||||||||||
|
In effect, Harper argues that the College Board has successfully outfoxed Atkinson by making the reformed test even tougher than the original. Atkinson called for the elimination of the traditional SAT test, and its replacement by an achievement test. Alright, said the College Board, you want an achievement test, we'll give you one. The new test demands higher math skills and includes a writing component that will not only be challenging, but can also act as a check on college-application essays, which nowadays are thoroughly vetted (and perhaps even written) by parents and guidance counselors. In short, according to Harper, standards have been toughened. On top of that, Harper assures us that parents and teachers will now be motivated by an achievement test to push for rigorous schooling, instead of wasting time reviewing old-style SAT analogies of dubious relevance to the high-school curriculum. Harper makes a good argument, but I don't buy it. All that is good in the new test could have been adopted without eliminating traditional measures of aptitude. For example, the test's length has now been extended to accommodate the new writing section. So why not simply add a writing section to the old test? Will an achievement test encourage schools to raise academic standards and devote more time to the traditional curriculum? Then why not encourage more colleges to require the achievement tests that already exist, and that are already used by many colleges? There's simply no getting around the fact that Atkinson has succeeded in eliminating critical information from the college-admissions process — information on a student's academic aptitude. That is a major loss, a terrible precedent, and an invitation to serious mischief. Atkinson bases his opposition to the SAT on studies of the University of California system that show that achievement tests, in combination with high-school grades, are the best predictors of success in college. The traditional SAT's measure of academic aptitude adds something to our ability to predict a student's success in college, but not much. So Atkinson reasons (with Harper's approval), why not get rid of the SAT altogether and simply rely on achievement tests and high-school grades to predict academic success? What we have not been told, however, is that there are unresolved problems in the UC studies — problems which suggest that the real reason for the SAT's allegedly weak predictive power rests, not with the test itself, but with the already well-advanced corruption of our academic standards. For example, an important study of UC Riverside students showed that the combined predictive power of the traditional SAT, the newer SAT II achievement test, and high-school grade-point average has fallen precipitously over the past several years. (See Table 3.) Why should that be? The authors of the Riverside study suggest that grade inflation may be the answer. In other words, as creeping grade inflation makes it more and more difficult to identify real academic excellence, the increment of information on ability provided by all measures, including the traditional SAT test, grows more and more difficult to verify. Indeed, according to the table provided, the predictive power of the traditional SAT test has fallen much more markedly than the predictive power of the SAT II achievement test. In fact the table indicates that, although the traditional SAT test is now slightly less predictive than the SAT II achievement test, it actually used to be more predictive. With college grades now lumping nearly all students in a relatively high range, it's no wonder that the finely tuned SAT measure is tough to "confirm." But the problem isn't the test, it's the gutted grading system by which we now measure academic achievement. And it's hardly surprising that sometimes inflated high-school grading predicts inflated college grading. What the SAT used to give us was a check on all that. Note also that the Riverside study is using freshman grades to measure the usefulness of SAT scores, when grades in later years of college, after students have mastered the system, may be even more reflective of SAT performance. In a sense, however, it's true that the problems with the SAT's predictive power may go beyond grade inflation to encompass the test itself. That's because the SAT was "re-centered" in the mid-Nineties — a move that inflated the scores of poor and mediocre students and flattened out differences among the best students. With the test recalibrated to bring out minor differences in the middle of the pack, predictive power was inevitably lost. Given that both grades and the "reformed" SAT have been prevented from identifying either real excellence or significant deficits, a large share of the information that testing can reveal has already been (intentionally) suppressed. The re-centered SAT may even have driven much of the grade inflation of the Nineties, simply by admitting less-qualified students (thus increasing pressure for inflated grades). And now grade inflation has prompted an even more radical gutting of the test. Last year, in "Grades and Race," I showed that a study purporting to prove racial bias in the Law School Admissions Test was better explained as proof of affirmative-action-fueled grade inflation in some of our finest colleges and universities. The study in question showed that minority students from prestigious universities scored well below white counterparts who had received similar grades in similar majors from the same universities. What this suggests is not that academic-aptitude tests are racially biased, but that the tests act as an important check on our inflated college-grading system. Yet this is exactly the sort of information that the College Board has now made permanently inaccessible. Instead of using academic-aptitude tests to call our corrupted grading system into question, Atkinson has tricked us into accepting inflated freshman grades as proof that college-aptitude tests are meaningless. Harper argues that, contrary to the claims of the College Board, private tutors can now raise SAT scores by somewhere between 100 and 300 points. Yet his figures are vague and his evidence appears to be anecdotal. Harper asks if it's believable that wealthy parents in New York City could spend thousands of dollars on test tutors without real results. I find this utterly believable. Just look at the millions of dollars spent on phony medical remedies every year. Hope springs eternal, and when people are wealthy and/or desperate, they will pay a heavy price just to cover the chance that some remedy (be it medical or educational) might do the trick. Harper's strongest argument is that an achievement test will create pressure from parents for rigorous education, instead of making students waste time drilling on analogies and other problems that have little to do with the actual high school curriculum. But Harper assumes the best sort of response to the new test — acceptance of its standards, and pressure placed on schools and students, not the College Board. No doubt, some of what he envisions will take place. But the shift to an achievement test will also bring a very different sort of pressure — from the very same people who have been opposed to testing all along. And this time, the SAT test will be far more vulnerable to attack. When the SAT was aptitude based, it was relatively insulated from tinkering or dilution. Bogus attempts to prove racial bias in the test were made, but any questions that conceivably carried racial bias were long ago eliminated, with very little change to the test itself. But an achievement test will be subject to death by a thousand cuts. Just wait until the accusations of racial bias on grading of the written section begin, with accompanying arguments about Black English, Ebonics, etc. And this only scratches the surface of potential problems involved in the relatively subjective grading of a written test. The addition of a writing section may well be a good idea (if done right), but it could easily have been adopted without eliminating aptitude testing, which remains invaluable, not only for its distinctive information, but also for its resistance to pressures for dilution. You can't easily recruit parents for an attempt to dilute a test filled with verbal analogies and mathematical-concept questions. There's no real target, and no easy way of proposing a definable dilution. But just watch parents and political groups sign on to an effort to lower the level of math achievement demanded of their children. Are preparations for analogies a waste of time? Actually, the best sort of school-based prep for the SAT concentrates on expanding vocabulary — teaching the meaning of the words, and not simply facility at running analogies. And as noted, even if some recalibration in favor of emphasis on training for higher math is in order, that can be achieved by encouraging more schools to require achievement tests — in addition to, but not instead of, the traditional SAT. The real losers in this "reform" are the famous "diamonds in the rough" — the students of high aptitude who may not yet have settled down, discovered their potential, or had access to the best secondary schools. Harper argues that these students are now victimized by those who have the advantage of test prep, but many "diamonds in the rough" are no doubt the very ones who actually benefit most from the test prep. The biggest jumps in scores after test prep are probably coming from the students who have not been high achievers and who do not feel secure on tests but who, with a little confidence, find that they have far more natural ability than they realized. And even undiscovered "diamonds" with no prep at all will do far better on an aptitude test than an achievement test. In short, a measure of aptitude, balanced with a measure of achievement, provides a college with the information it requires to draw up a total picture of a young and still developing student. Hiding half of that information turns the admissions process into little more than a series of half-truths. Harper himself notes that "aptitude test" is now a dirty word on the academic Left. This hostility to differences in ability is the real reason for the change in the SAT. Surrendering to Atkinson's threat to withdraw the California system from the SAT, instead of taking him on in a principled public fight, was an act of cowardice that cannot be separated from the substantive meaning of this so-called reform. Having seen the College Board surrender to relatively mild political and financial pressure, the opponents of testing will be emboldened to pick away at an SAT which will now be far more vulnerable to calls for dilution than the old. In the meantime, the most important tool we have for exposing the corruption of our inflated grading system has been completely destroyed. This is a famous victory for opponents of academic excellence — all the greater because some conservatives have convinced themselves that it's all for the best. Richard Atkinson has pronounced himself "delighted" with the College Board's reforms. It's no wonder. By caving in to political pressure and eliminating a critical and irreplaceable measure of academic ability, the College Board hasn't outfoxed its critics, it has outsmarted itself. — Stanley Kurtz is a research fellow at the Hoover Institution at Stanford University. |
|
|
|
|||||||||||||||||||||||
<% dim printurl
printurl = Request.ServerVariables("URL")%>
he
