Wednesday, March 05, 2014

The Perfect Score (or not)

I've been fascinated to read the media coverage of Debbie Stier's new book, The Perfect Score. As her teenage son prepared to apply to college, Stier decided to uncover the secrets of the SAT, and took it 7 times herself in the course of a year. She went through lots of coaching and SAT prep too in her quest for (as the title notes) the perfect score.

What's most fascinating to me, though, in light of the debate about the role of the SAT in college admissions, and hence society, is that she didn't actually succeed.

The SAT began to be used, many decades ago, as a way to compare students from various backgrounds. In some ways, such a standardized test was supposed to be an equalizer. A kid from a rough background could still score well on a test of intelligence, and hence could open up elite colleges to students not from the Andovers and Exeters of the world.

Of course, the idea of testing intelligence has gone in and out of favor over the years. The SAT has broadly been changed to test more of the material covered in high school. In theory, it can still be a way to compare kids from different backgrounds. Some high schools are much harder than others. Straight A's at one school may mean little in terms of how prepared for college you are, whereas straight A's at another school may mean a great deal. I experienced this myself in my two different high schools. A good college admissions test should be able to show this.

But people can prepare for tests. And so, one widespread criticism of the SAT is that well-to-do kids can spend thousands of dollars on test prep. They can be coached to higher scores, and take the test numerous times, and hence appear more prepared than they are.

Which may be true. But stories like Stier's also show that the SAT may still mean something. After her year of coaching and prep, she did manage to get a perfect score on the writing component. And guess what? Having worked in publishing for years, and as a published author of a book, she probably is quite competent at writing!

On the other hand, she never managed to boost her math SAT score higher than 560 (out of 800). This is after a year of studying the high school math covered on the SAT and taking the test numerous times. Given that, isn't it possible to believe that a high schooler scoring in the 700s on the math section is, in fact, showing serious mathematical promise? Whether she's been coached or not?

If the SAT were perfectly "coachable," you'd see a lot more perfect scores. As it is, only a few hundred students per year score perfect 2400s. It's fashionable to trash the SAT, but it may mean something despite its flaws.


nicoleandmaggie said...

Unlike you, I did not get a perfect SAT, so you can see that my bias lies in a different direction. However, I did max out my school's math options (with straight As) and took substantial university math in high school (my first college math class was DiffyQ, having already taken number theory, linear algebra, and calc 3, among other less challenging math electives). I tutored math and taught math summer school.

A problem with the SAT and GRE is that if you make one silly mistake, your score drops precipitously. A perfect score isn't much different from an imperfect score.

On top of that, we're not worried about the people like me who are stuck in 700-land. We're going to do fine with or without coaching. We're worried about the people whose scores are not actually reflective of their ability on the other end. It doesn't matter if the SAT can't be coached perfectly for the people whose scores are artificially low because of cultural biases in the test. They're not the ones getting coaching (and they're the ones whose scores are most likely to be improved by such coaching if offered). There are documented racial differences in scores not because of aptitude but because of the cultural differences in examples used. If you've never seen a gazebo, it is going to throw you for a loop when it's used as an example in a question.

Laura Vanderkam said...

Only partly perfect! I never did figure out what I missed in the English section...

Anonymous said...

"if you make one silly mistake, your score drops precipitously. A perfect score isn't much different from an imperfect score."

yes, this is my issue, too. I do not understand, though, how an intelligent person with sufficient preparation was unable to break the 500's on the math score. I am guessing insufficient or inappropriate preparation (for example, if she just "reviewed" math problems, instead of doing them herself). Haven't read the book, though. And, maybe I'm just playing into my own bias of thinking that it can't be all that hard to get into the 700's on the SAT (though I think a perfect score is a matter of some randomness, because of the small number of questions that differentiate those scores).

Otherwise, it seems to me that we're saying we can't teach the fairly low level math that's on the SAT (though, I guess another possibility is that we can't teach it to be done as fast as it needs to be done on the SAT).


Calee said...

As someone who scored a 570 on the SAT math in high school and then a 750 on the math GRE 9 years later, I am not a huge fan of the SAT. The thing that raised my scores after 9 years - I taught SAT prep during and after college. It was, of course, the verbal and later the essay side I was hired to teach, but I still heard the math teachers' tricks from time to time. It was pure arrogance to turn down test prep when I was 16. Yes, the GRE is a different test (and it's computer based so there were fewer questions and more incentive to get the first few right) but I don't think I actually learned math in that time. It's all about test prep and the accompanying strategies. It can make a huge difference.

Gail Post, Ph.D. said...

Very timely post, given the recent announcement that the College Board plans to revise the SATs. I think we all know that the SATs are not a measure of IQ. Although improvements can be made through test prep, there is a limit to this, and some kids just perform better on these kinds of tests, regardless of how much test prep they have. It will be interesting to see how changes in the test play out over time.
Gail Post/

Nother Barb said...

My 9th grader is taking the SAT on Saturday through an academic talent search, and I'll be interested to see his score alongside the ACT he took last month (and in 6th grade). Because they won't be used for college admissions, the only prep was to take the sample test one time at home. If he takes SAT "for realsies" later, it will be interesting again to see the difference.

But the book makes me wonder, what does the test test, and what does the prep do? Perhaps the author has a learning style for math that does not lend itself to such tests or prep?

Laura, if it isn't too personal, did you take a prep course before you took the exam in middle school? And did you take the exam as a prerequisite for the academy?

Laura Vanderkam said...

@Nother Barb - I did wonder about this, too, why the math section proved so difficult given that she was able to improve on the verbal. No, I never did any test prep courses any of the times I took it. I did MathCounts in middle school, which involved practicing doing algebra and geometry questions very quickly. So that probably helped. I think I checked out a book of practice problems from the library once before I took it "for real" in high school, but I didn't do a formal practice test.

Anonymous said...

I went to college with a girl who claimed to have had a perfect SAT score and I never doubted her, she was incredibly smart and I won't go into detail about her for the sake of her privacy.

My goal for myself to have had a successful try at the SAT was to score in the "A" letter range and do equally as well in English and Math.

I accomplished that goal and it felt like nice closure for the end of high school.

The SAT score probably means something different to different people and tells something about each person. I love English and Math equally. I have one parent who was gifted in math and the other who was gifted in English, so it can show family history and genetic trends.

No matter what standardized test they choose I think a well-prepared student should feel confident and not worry.