kitchen table math, the sequel: Michelle Hernandez on admissions & SATs

Wednesday, January 30, 2008

Michelle Hernandez on admissions & SATs

The most obvious result of recentering is the fact that what used to be considered an unusually high score on the SAT I is now just an average score. To be exact, for years a combined score of 1,400 was considered the dividing line between normal scores and extremely high scores for highly selective colleges. This si no longer the case. The average combined score for the admitted class of 2000 at Dartmouth was over 1,410. In fact, for the class of 2000, the average scores of all 11, 400 applicants who applied to Dartmouth (this includes all the weakest applicants in the pol) was 662V, 677M, almost 1,340 combined. For the class of 2001, the average was even higher: 664V, 683M.

While it used to be extremely rare to see students with verbal scores over 700 (remember, only 1 percent of all students taking the unrecentered test scored over a 700, and those students could not have missed more than three or four questions on the whole test to achieve that score), now it is much more common. In fact, half of the members of the class of 2000 scored over 710 on the verbal section, about the same number who scored over 710 on the math section.

Here is an SAT I breakdown of the acceptance rates at Dartmouth for the class of 20000:

Verbal SAT Math SAT
399 or less 0%
400-499 1.6%
450-499 4.1%
500-549 5.4%
550-599 7.5%
600-649 10.1%
650-699 14.8%
700-749 28.2%
750-800 52.2%

Math SAT
399 or less 0%
400-499 2.4%
450-499 6.3%
500-549 5.4%
550-599 7.1%
600-649 9.5%
650-699 15.1%
700-749 24.4%
750-800 44.0%

All these statistics taken together mean that parents, students, and counselors need to readjust their standard for excellence dramatically in terms of scores. When a college counselor or a parent calls an admissions officer to ask whether or not a student has a chance of being admitted, he usually starts out by announcing that the student has very high SAT I scores. The problem is, when the officer asks what he means by "very high," he usually says, "Over fourteen hundred," which, as we have just seen, is not only not "very high"; it is actually below the class average.

[snip]

It is still, as it has always been, more impressive to see high verbal scores than high math scores, since most students at the highly selective colleges will be doing much more writing and reading than math. Verbal ability is still a good indicator of how strong a reader the student is. The ability to read well will ultimately have a bigger impact on most college students than the ability to do SA I math very well, especially since the level of SAT math is not particularly high. There are many students who do terribly on the SAT I math and yet who manage to get the highest score of 5 on the AP calculus exam. If any math is useful at the college level, it is calculus, not the basic math covered on the SAT I. Therefore, SAT I scores of 750 V, 630M would be much more impressive for most highly selective colleges than a 640V, 780M, even though the latter score has a higher combined total by forty points.

The final point I want to make about interpreting SAT I recentered scores is that because most admissions officers are not gifted in math, there is still the tendency to use the 700 cut-off as the magic number between good and excellent scores, even though, as we have seen, the averag score is over 1,420 combined....Remember probably 99 percent of all current admissions officers took the old SAT I, the nonrecentered test. In their own personal histories, 1,400 has always been a high score. It is extremely difficult to consider applicants under a totally different system from the one you had in high school. [Hernandez' book was published in 1997, so this is 10 years ago.]

[snip]

Oddly enough, and I saw this during my last two years in admissions, officers still respond to any scores over 700 by saying that they are strong, whereas if the scores are in the mid to high 600s, they tend to refer to them as average. This is simply not accurate. In Ivy applicant pools, a 700V, 700M is now an average, even below-average, score, although some officers will not be able to draw the correct conclusion when analyzing student scores.

Despite this inability on the officers' part to be 100 percent accurate in interpreting your scores 100 percent of the time, they will undoubtedly improve as the years go on and they see years' worth of recentered scores. In the meantime, all applicants should be aware of the reality of SAT I, which is that if you want to stand out in the Ivy applicant pool, you should aim for a combined score of about 1,490 or higher, or at least a verbal score over 730 or so, which would allow you to have a modest 650 math score and still look strong in the applicant pool

source:
A Is for Admissions: The Insider's Guide to Getting into the Ivy League and Other Top Colleges
by Michele A. Hernandez
pp. 46-49


Is she right about SAT-level math?

That strikes me as wrong -- at least based in the fact that I constantly see algebra called the "gatekeeper to advanced courses in math," etc.

Hernandez was an assistant director of admissions at Dartmouth. She's a high end college counselor now. Mathew says she's fantastic, and her books are terrific.

26 comments:

VickyS said...

You can go to the SAT website and have their Question of the Day delivered to your inbox everyday. We've been doing this for a year and the math questions are really very basic. Whether this reflects the actual test accurately or not, I don't know.

Anonymous said...

The SAT I math test seems to be mostly algebra and geometry, so yes, it's basic stuff (in the grand scheme of things). Remember that students generally take it starting in junior year, so it covers through what would be covered in 10th grade in the "college track". The College Board site lists the concepts covered on the SAT I.

Another difference between when we old fogeys took the SAT (ca. 1980) and now is that back then, you were not allowed to use a calculator, so your answer could be incorrect because of pencil transcription errors or stupid arithmetic mistakes.

Yet another difference: A minority of people took the SAT more than once back then, and hardly anyone used a prep course. I think back then, prep courses were for people who were in danger of really flubbing up. (I think the twin sisters in my class could have benefited from a prep course or something to raise their scores from the abysmal 950 total that each got.)

SteveH said...

"There are many students who do terribly on the SAT I math and yet who manage to get the highest score of 5 on the AP calculus exam."

Bad at algebra and a 5 on the AP test? I don't believe it one bit.


"If any math is useful at the college level, it is calculus, not the basic math covered on the SAT I."

No. For all of those non-science and non-math people who go to "highly selective colleges", algebra is a whole lot more important. For science and math people, both are important.


It's interesting to see how (probably math-phobic) admissions people view math.



"... if you want to stand out in the Ivy applicant pool, you should aim for a combined score of about 1,490 or higher, or at least a verbal score over 730 or so, which would allow you to have a modest 650 math score and still look strong in the applicant pool"

Huh? 730 + 650 = ?

Where's my calculator?

Anonymous said...

That was the first phrase that jumped out at me and it's one of those burning questions that I have, "What exactly is the level of SAT math?" Because anyone can slap the label of geometry or algebra on anything, it's not the subject area that makes the level, so much as the depth of the questions in the subject area.

My father in law (now a retired research mathematician) says that because of the context of the SAT they simply can't ask high level math questions. The context being that the questions must be answerable within a minute or two and according to him any real math question worth its salt requires more than a minute or two of pondering and thinking and inferencing in order to answer.

At any rate, I haven't seen a good discussion of this by math experts and look forward to it only because I've seen so many people act as if the SAT is the gold standard of math knowledge. Perhaps it is not, perhaps it's more like a DMV driving test which only serves to weed out the incompetent.

Catherine Johnson said...

I'll have to check to make sure I typed the passage correctly - the math error could be from me, not Hernandez.

I mentioned on another thread that years ago, when I took the SAT with no prepping, etc., I got a 620 on math.

When I took a sample test after finishing Saxon Algebra 1 I got a 620 again -- and I got every algebra 1 & geometry question correct.

The questions I missed were algebra 2 questions, primarily functions. (Or functions entirely.)

The test is much more advanced than the test I took.

I wouldn't say it's "hard," especially, but it covers much more material.

If I had gotten every question on algebra 1 & geometry right back in the day, I would have had a math score of 800.

Catherine Johnson said...

Perhaps it is not, perhaps it's more like a DMV driving test which only serves to weed out the incompetent.

The way I think of it -- and this may be completely wrong, so please tell me if so -- is that SAT-M is a test of essential procedures and concepts.

Sort of like a test of grammar.

As I understand it, this is one of the reasons people place so much emphasis on SAT-V, which does test thinking, making inferences, etc.

The SAT-V is a very substantial test. When admissions officers are honest they'll tell you that SAT-Vs in the 700s perfectly predict the ability to do college-level work. Period.

Same goes for SAT-V in the 500s.

Ed did one of the test passages, which was on history. He is a historian and it took him the full allotment of time to read the passage and answer the questions.

A couple of the people on the Frontline interviews say that SAT-V is in fact an IQ test.

SAT-M is a content or subject matter test.

Catherine Johnson said...

nope, I typed it right

sigh

you'd think an editor would have caught that (I should have caught it, but otoh this isn't my day job)

Catherine Johnson said...

they simply can't ask high level math questions

Absolutely.

I think it's a pretty solid test of "basic skills" in algebra 1 & 2 & h.s. geometry.

Catherine Johnson said...

Yet another difference: A minority of people took the SAT more than once back then, and hardly anyone used a prep course.

Actually, I took it twice.

I cruised in to the testing place after spending the night before at a party and ended up with a verbal score in the 600s -- which somehow I figured was bad. (How would I have known this, living there in the middle of IL?? Maybe it was a lot lower than what my PSAT would have predicted?)

I took it again and my verbal score jumped 100 points. Math stayed the same.

After that I read zillions of articles on how it's impossible to change your score by taking the test more than once or practicing.

Of course, they weren't talking about taking the test once while paying attention and once while skimming distractedly....

Catherine Johnson said...

When I took the sample SAT test a year or so ago it was nighttime, I had been drinking red wine, I was tired, and the lights were on so low I was having trouble seeing the page.

Plus I forgot I was supposed to use test-taking strategy, so I was sitting there doing the SAT the way I did Saxon problem sets: focusing on each one & sticking with it until I could solve it.

About halfway through I remembered: oh, wait. This is an SAT test.

So then I started skipping unfamiliar problems, doing the ones I could do, then returning to the others.

It wasn't until afterwards that I thought: That was a waste.

Catherine Johnson said...

I did get a 790-V.

Missed two questions, one of which I dispute, the other of which I chalk up to the red wine/tired/no light situation.

Catherine Johnson said...

Bad at algebra and a 5 on the AP test? I don't believe it one bit.

That struck me as wrong, but could you forget a lot of your algebra & still do calculus??

Anonymous said...

So then I started skipping unfamiliar problems, doing the ones I could do, then returning to the others.

Is it computerized or on paper these days? I didn't think that you could skip and come back to more difficult problems on a computerized test. You are forced to enter an answer before you move on.

Anonymous said...

So then I started skipping unfamiliar problems, doing the ones I could do, then returning to the others.

Is it computerized or on paper these days? I didn't think that you could skip and come back to more difficult problems on a computerized test. You are forced to enter an answer before you move on.

Catherine Johnson said...

I was just using a test prep book, but offhand I don't see why computerized paper would make a difference.

Chris took the ISEE in November, and the test-taking strategy was to circle the questions you couldn't do immediately and come back -- something I never managed to think of.

He used two symbols:

* a star by questions he knew how to do but needed more time to think about or do computations for

* a circle for questions he had no idea how to do offhand and was going to look at one more time before he simply marked in a random answer

(ISEE scoring doesn't correct for guessing so the practice book advises you to fill in an answer whether you can do the problem or not.)

I just asked Chris about it; he said he could definitely go back to earlier questions.

What you can't do is go back to earlier sections of the test.

He did really quite well, considering all we've been through. He was in the 82nd percentile for "mathematical reasoning" (essentially, quantitative comparisons) and the 74th percentile for "mathematmics achievement" for 8th graders taking the test.

The ISEE population is, apparently, an elite group; the Fordham admissions director said that they informally tack on 10 points to each score to arrive at an equivalent score for the TACHS. (Test for Admission to Catholic High Schools). He said that, in turn, the TACHS population is more selective than the PSAT population.

This is still not where we want him to be, given his reading comprehension....his ISEE scores show a massive split between reading comprehension and everything else. That is a sure sign of inadequate education.

It really is astonishing, looking at his scores.

I went to a rural school and had the same split between reading and math that my son has attending his chi-chi $22,000/student suburban Westchester school.

plus ca change

SteveH said...

"That struck me as wrong, but could you forget a lot of your algebra & still do calculus??"

No.

Try differentiating and simplifying:

[3x^(1/2)*(4x-1)]/[x^2 -4x -5]

You might get the differentiation pattern correct (top times the derivative of the bottom ...), but you'll never survive the simplification.

My reaction to that statement (which was used to make a huge, sweeping generalization) is "prove it".

This raises an interesting question. Who, exactly, are the people who decide who gets in and who doesn't? How are the borderline cases judged? It sounds like they are not math people and have a bias against math SAT scores.

Take two applicants with the same combined score. It sounds like the one who has the higher verbal score will get in, even if that person wants to go into engineering.

Catherine Johnson said...

It sounds like the one who has the higher verbal score will get in, even if that person wants to go into engineering.

I'm sure she's correct in saying that elite schools favor SAT-V over SAT-M.

I also think they're right to do so. SAT-V is an IQ test; SAT-M isn't (as far as I know.)

It would be interesting to know how they select students who are interested in fields like engineering.

Dartmouth is a liberal arts school; you can major in math, but not in engineering.

I suspect that the Ivies favor SAT-V over SAT-M no matter what the applicant's stated area of interest.

On the other hand, this book is 10 years old and I gather from the recent WSJ article that schools are looking for kids who are strong in math and have a specific interest in math/science.

I'll check the NYU scores.

Catherine Johnson said...

I don't know calculus, so I can't judge statements like this.

Nevertheless, it sounded wrong.

She says, in the chapter on the "Academic Index" score admissions officers calculate, that admissions officers aren't comfortable with math and don't know it well.

Anonymous said...

I'd like to hear more about who gets in, especially where a school has separate undergraduate schools for its engineering and arts and sciences programs. Does the engineering school have its own set of admissions officers, who are more likely to be wowed by a higher SAT math? Or do all applicants to a selective science program already have 800M scores, making the verbal score sort of a tie-breaker.

BTW, I disagree with the statement that SAT-V is closely linked to IQ. I think that SAT-V is closely linked to how well read your parents are. SAT-M probably has a greater correlation to IQ.

Lastly, you are not going to botch the SAT-M and get a 5 on either the AB or the BC. The test-taking skill sets for all three tests are not that much different.

SteveH said...

"SAT-M isn't (as far as I know.)"

Wouldn't that imply that practice will greatly improve your score?


My wife has an extremely high IQ, but couldn't survive a college calculus course if her life depended on it. Her math SAT score was good. My guess is that a 5 on the AP test is a better indicator for technical degrees, but I can't imagine anyone doing poorly on the SAT-M and then get a 5 on the AP test.

Anonymous said...

I think my earlier link to the list of concepts covered by the SAT-M got buried in the general discussion, so here it is again:

Concepts covered by SAT-M

Each section contains a link to some review material. For example, under the "Review Number and Operations" section, the "Percent Increase and Decrease" page contains this question:

"If the price of a computer was decreased from $1,000 to $750, by what percent was the price decreased?"

Anonymous said...

FWIW, I also took the SAT twice, because I couldn't stand having an almost-perfect math score and wanted to shoot for the 800. So on the second go-round, my math score went up 10, but my verbal went down 20. Then again, if you think of the score as an estimate, with some built-in error, such variance is to be expected.

Linda Seebach said...

The SAT-M is much more purely an IQ test than the SAT-V (g-loaded, as they say). It does require a certain specific content, roughly the amount of math a college-bound student should have studied by the end of 10th grade, but by definition almost all the students who take it have done so. Since hardly anybody learns math outside of school, the score is far less susceptible to family SES factors, except insofar as they reflect shared familial IQ.

Well, ok, outside the school curriculum. Catherine is teaching math to C., but only because the school is incompetent.

Hernandez can't add, which tells you something about the credibility of her views on the greater significance of verbal over math scores. She is probably also right about a similar blindness on the part of her former colleagues, with a similar cause.

Her comments would be less misleading if she were careful to insert "at Dartmouth" where it is logically necessary, e.g.: "When a college counselor or a parent calls an admissions officer to ask whether or not a student has a chance of being admitted, he usually starts out by announcing that the student has very high SAT I scores. The problem is, when the officer asks what he means by "very high," he usually says, "Over fourteen hundred," which, as we have just seen, is not only not "very high"; it is actually below the class average."

Over 1400 is very high; but not at Dartmouth.

That the mathematical content of the SAT-M is fairly low-level is not a bug, it's a feature. Properly designed questions about elementary material are excellent at separating genuine talent from routine competence, which is why some graduate programs pay scarcely any attention to GRE-V scores. They're all high, but math is an effective filter.

Anonymous said...

Catherine --

Regarding the question about computerized tests: Some standardized tests, such as the GRE, have been converted to adaptive computer-based tests. "Adaptive" means that the test program learns from your performance. As you answer questions incorrectly or correctly, the program "learns" what areas give you trouble, and it gives you questions to target more specific areas.

For example, say you are taking an algebra test. If you ace four point-slope linear equation questions in a row, it may figure that you know linear equations cold, so you won't see any more of those, and instead you will get more difficult quadratic equations. If you ace those, you'll get higher polynomials or quadratics with complex roots or something. But if you start making mistakes on the quadratics, it may give you more of them, but concentrating on whether you are tripped up by an x^2 coefficient that is greater than 1, or by negative coefficients, or some other detailed concept.

This is a simplified example, of course, but should illustrate how the adaptive CBT works.

Because the test is adaptive, you are not allowed to skip questions that you don't know and come back to them later.

More info about the adaptive CBT GRE is here.

Anonymous said...

---raises an interesting question. Who, exactly, are the people who decide who gets in and who doesn't? How are the borderline cases judged? It sounds like they are not math people and have a bias against math SAT scores.

Remember that it's no longer a borderline "should we take this student or not." It's "should we take this student OVER this other student."

Admissions departments are often biased away from the geeks, even at schools like MIT or Cal's Engineering school, toward the more interesting, well rounded student who is also excelling in math and science. And the folks in those departments decide by some basic criteria who is even considered at all, and then they whittle further down by their own interest in the person; they then make the case to the rest of the admission staff, who gives the score for go/no-go.

Anonymous said...

Have u try the SCIENCE online bookstore Cocomartini

http://www.cocomartini.com/

I get all my textbooks for this semester from this bookstore. All are brand new and half price discount.

Good luck and wish some help.

hehe ^_^