The Calico Cat
A blog about business, economics, law, politics, current events, and political sex scandals

click here to return to the Calico Cat homepage

A new SAT? Why fix what works?

I take a break from writing about oil to write about the SAT. As you may be aware, the SAT was the subject of a cover article from Time magazine two weeks ago (yes, I know I'm a little late on this, but better late than never).

Reading this article was the first time I discovered that a new SAT is being intorduced in 2005 (link to college board page about the new SAT).

It seems to me that the old SAT was working just fine, why do we need a new one? Of course there do seem to be a lot of people out there who dislike the SAT, but their reasons have never seemed very easy to fathom. For example, the University of California has been talking about not using the SAT (see The Tech: California Colleges May Drop SAT Use).

The turning point, according to Atkinson [the President of the University of California], was a trip to an upscale private school, where he learned "that they spend hours each month -- directly and indirectly -- preparing for the SAT."

"The time involved was not aimed at developing the students' reading and writing abilities but rather their test-taking skills," he continued. "I have concluded what many others have concluded -- that America's overemphasis on the SAT is compromising our educational system."

The article in Time mentioned the University of California as one of the motivations behind the creation of the new SAT. If UC dropped the SAT, some think that this would be a pretty big blow to the SAT. I would point out, however, that Harvard Business School dropped the use of the GMAT, but then several years later HBS started using it again. This might very well have been the situation in California. Could the UC system really stop using the SAT without harming its academic reputation?

With the SAT being an important part of college admission, it shouldn't come as a surprise that upscale private schools are aiming to help their students do better on it. But surely the same will apply to the new SAT. So what's the point of changing it?

We need standardized testing, otherwise there would be no way to compare students from different high schools. If student A has a 93.4 average at high school A, student B has a 4.65 average at high school B and student C has a 3.3 average at high school C, which is the better student? There is absolutely no way of knowing. Without standardized tests, elite colleges would wind up only taking students from elite high schools that they were familiar with, and students from high schools that were unfamiliar to the elite colleges would be at a severe disadvantage. Thanks to the SAT, a kid at a mediocre public high school can demonstrate that he's as equally qualified as a kid at a private prep school.

The new SAT is going to be less of an aptitude test and more of an achievement test. The IQ-type test questions that aren't normally seen on high school tests, such as analogies and quantitative comparisons, are being eliminated. Questions covering 11th grade math and grammar are being added. A writing test is being added. The new SAT will look less like an IQ test and more like a regular high school test. But there's no evidence that this will make the test more useful for selecting college students. The elite colleges already require that students take Achievement Tests in addition to the SAT (nowadays the Achievement Tests are called SAT-II tests, but I remember what they used to be called), and these tests do cover specific course material. And the colleges also have high school grades to use to evaluate the applicant. It's not as if people with high SAT scores and no other qualifications are sneaking undeservedly into Harvard and Yale. In fact, the opposite is true. A kid with a high SAT score but nothing else going for him (no extra-curricular activities and mediocre high school grades) will probably be rejected by Harvard and Yale.

I find the graded essay to be the most worrisome part of the new SAT. Essays are much more difficult to grade than multiple choice questions. The cost of grading the SAT will increase, and the reliability will decrease. And college bound kids, instead of spending countless hours practicing analogies, will spend countless hours learning how to write in a way that the SAT graders will appreciate.

Let me share my own personal example about graded essays. If you've read my bio, you know that I went to law school. In the UCC class I took in third year of law school, I got the highest grade in the class based on an essay exam test on contract law. So I am pretty confident I know how to write an essay exam answer on contract law. Two years later I took the CPA exam, and on the law part I was delighted to see a contract law essay question. I was certain that I smoked that essay and got the highest possible score. Well it turns out, I got the lowest possible score on the essay question. The same writing style that I had learned to use in law school to get high grades actually got me the lowest possible grade on the CPA exam essay! This demonstrates how essay grading can be highly subjective. Obviously, the non-lawyer CPA exam graders were unfamiliar with the law school writing style.

Kids who want to do well on the new SAT will have to spend countless hours learning the "correct" way to write an SAT essay. I suspect that kids at non-college prep oriented high schools will be at a bigger disadvantage on the new SAT than on the old SAT. And the Time article suggests the same thing.

I suspect that the real motivation for the new SAT is the statistic that shows blacks do poorly on the SAT compared to whites. According to Time, in 2003 the average black SAT score was 857 and the average white SAT score was 1063. (The average Asian SAT score was 1083, which shows that there's nothing special about being white.)

Liberal angst over the black/white SAT score difference is probably causing the SAT to be redesigned. But there's no evidence that blacks will do any better on the new SAT. In fact, Time says that test questions involving specific knowledge learned in high school can make the gap even worse, because many blacks go to high schools with less academically challenging programs.

Time also points out that SAT scores actually overpredict the college grades of blacks. In other words, if a black student and white student both have the same SAT score, the white student will probably do a little bit better in college than the black student. This is the exact opposite of what people probably think--that the black student with the lower SAT score will do just fine in college because the SAT is biased. College grades are based mostly on tests. Black students don't do as well as whites on academically oriented tests. There are no standardized tests that I've heard about where blacks score better than whites. The SAT isn't causing the situation, it is merely reporting it. People who don't like the SAT because there is a black/white score difference are wasting their energy blaming the messenger.

Because colleges use affirmative action when making admissions decisions, black applicants aren't even being hurt by the fact that they score lower on the SAT. Most colleges seem to be adding back most of the 206 point score difference.

I feel that the proverb "if it ain't broke, don't fix it" applies to the SAT. The SAT has done a good job for many decades. Messing around with the SAT seems more likely to make things worse than to make anything better.

posted Sunday, November 02, 2003

0 Comments:

Post a Comment

Previous Posts

  • The end of the Oil Age, Part II
  • The end of the Oil Age, Part I
  • More about student loans and the value of a colleg...
  • A cool Census link
  • Google and "blog noise"
  • Statehood for Washington, D.C.? No thanks
  • Student loan rip-off
  • Kodak and the future of film
  • 10,000 is just a number
  • Higher corporate profits through lower salaries
  • This page is powered by Blogger. Isn't yours?