SAT was designed to measure aptitude

[mathwonk]

When I was a high schooler, there were two advantages that have apparently been removed from todays students:

1) The old SAT was designed to measure aptitude, not achievement or learned skills.

2) There was a real "merit scholarship" program, generous enough to cover more than the tuition to Harvard for a relatively poor boy like me.


Thus the SAT allowed pupils who had received poor educations, such as those available essentially everywhere in the American south, to demonstrate college potential. Then there was money to provide the opportunity to get a good education.

The Economist this week argues that those things essentially changed America from a country for the privileged, into a meritocracy. (It seems George Bush got into Yale before the SAT revolution took place there.)

The result was that Americans began to succeed and obtain jobs and money and status, based more on brains than connections, a boon for the entire country, in increased wealth and opportunity.

This powerful engine of change was the brainchild of James Conant of Harvard, an educational visionary of the early 20th century.

That has all been reversed. The SAT has been revised twice in the last decade, once to inflate the scoring, and now to remove the one skill I have found most crucial in my entire career as a research mathematician, namely analogy.

[Here is an example of reasoning by analogy: if a line segment has 2 endpoints, a square has 4 sides, and a cube has 6 faces, how many "faces" does a "4 dimensional cube" have? Didn't know that before did you? Neither did I.]

The merit sholarship stipend was long ago reduced to a token amount that would not come close to paying Harvard tuition for even one semester.

I suggest you young people start asking why this is? Although many people argued that the old SAT discriminated against poor and ethnic minorities, actuaoly the opposite seems true. A test designed to measure intelligence rather than essay writing skills, helps the less well educated but bright poor child.

Anyone could afford an SAT practice book for under 20 dollars retail, and free in libraries. Now an education in good essay writing is not so cheap.

The real travesty, according to the Economist, and I agree, was not that old SAT scores failed to identify the best candidates for college, but that colleges deviated from using them in admitting students for political or monetary reasons.

Since this revolution backwards in time affects you, I recommend you take an interest in it.

 

[juvenal]

Haven't read the essay - will take a look at it.

I don't think the old SAT's were that useful for colleges to find diamonds-in-the-rough. It seems to me that the older SAT was a better measure of raw intelligence (if that term is even definable) and the newer SAT is a better measure of useful skills for college.

College, at least from my own experience, is not exactly the most supportive environment out there. You're expected to come in with fairly decent skills, including essay writing. I don't think raw itelligence will cut it by itself. So I think that a scenario where a student would have been identified ONLY because of his high, older SAT score is rare, and such a student would have a fairly tough time in college.

There are a lot of side issues of course, such as whether essay-writing skills can really be measured objectively in these standardized tests.

 

[mathwonk]

Well you are right, essay writing skills are more useful in college, but the point of the old SAT's of finding diamonds in the rough was valid, and borne out by the huge enrollment of poor boys at schools like Harvard in the 1960's as compared to very few or none in the 1930's. (There were several hundred merit scholars at Harvard at the time I was there, maybe just in my class. There were two of us from my high school class of 60 in Tennessee, both of us poor.)

We did have a rough time of it for lack of good preparation [my first essay was marked "unoriginal and dull!, C-"], but I qualified for honors math and chemistry ahead of most others, on placement exams, and even was placed in a graduate logic class as a freshman. Stats showed at Harvard in those days that the public school boys lagged behind for the first year or so and then pulled ahead in the later years, ahead of the prep school boys with their superior background.

Some like me even flunked out temporarily partly from the culture shock, but were readmitted after working hard for one year, and eventually excelled. The ultimate goals we reached would have never been achieved without the opportunity to have a good education based primarily on raw intelligence and drive.

Many of the answers I give on this forum in the maths sectiopn are based on the education i had at harvard. The books used there (Courant, Apostol) were not used everywhere, and the professors were far more qualified, as I discovered by comparing notes with friends who went to other schools back then.

Mike Spivak, author of perhaps the most popular and influential honors calculus book today was one of those bright undergraduates at Harvard, being taught by professors whose standard of teaching was hard to find in books in those days.

The point is the argument for changing the test is that it was biased against poor students and minorities, exactly the opposite of the truth. It enabled those groups to have the same shot as a rich kid, with sufficient hard work. While there one enjoyed side effects like hearing Malcolm X speak in person, or Adlai Stevenson, or William Sloan Coffin.

I suspect the combination of the old SAT and the merit scholarship program caused a temporary revolution, access to the best higher education for those students best able to use it, that is largely lost today.

It may be too that there is a political motive to deny this opportunity, as those students from the 1960's did not support the political goals of the established business community as did the scions of wealthy families, and knowledge is power.

There is a reason that most math positions in universities, both graduate students and faculty, are filled by foreign students today, who greatly excel their American counterparts.

 

[juvenal]

I understand what you are saying, and I also attended Harvard, as an undergrad in the 90's. I think that the playing field is different now than it was in the 60's. There are fewer prep schoolers and more public schoolers. Probably less legacies as well. So I don't think the caliber of the student has gone down - in fact - it's gone up because it's much harder to get in. The SAT, as it rightfully should be, is only one of many different things that admissions officers look at when making a decision. (And I got a 1540, 3 wrong, back in the days when almost nobody was given a 1600).

Because of this, there are probably fewer opportunities for diamonds in the rough who might get a second look because all they have is a high SAT score, but that is not to say that the students who do get in are less qualified. Far from it.

 

[misskitty]

I'm not sure what to think of the new SAT. I'm retaking it this year to improve my scores. Some people say it's an advantage to taking the new one. I'm not sure though. Everyone I've spoken with makes it sound like you'll never get anywhere if you score below a 1500. That's intimidating and VERY scary for a person such as myself.

 

[juvenal]

Another point I'd like to add is that I don't think that the college admissions process should be really responsible for finding those diamonds in the rough. If there are truly gifted students out there with raw ability and who are not putting those talents to use, they needed to be spotted early on in high school and put into some gifted and talented program and nurtured, so that by the time they are ready to attend college, they'll be as fully prepared as any other entering student. If these gifted and talented programs don't exist, then that's where lobbying efforts should go. The old-style SAT isn't necessarily the most reliable indicator of raw ability and talent anyhow.

 

[motai]

I've never really been that good at standardized tests. I had a fairly low score on the SAT and only slightly better scores on the ACT (two 26s). I also had the tendency to... psychologically bash myself from these low scores because I was generally too slow even after months of test preparations. But then again, that was how I was raised. My grandfather (who was a physicist himself) taught me to take problems slowly and methodically, and I agree with him. I see little reason for me to rush my calculations in the workforce, it will only lead me to errors.

Do my low scores mean that I will be a failure in college? Probably not. I already almost have a year of college just from Dual Enrollment/AP courses alone, so the score itself is becoming increasingly meaningless to me.

As for the essay portion, that is a big maybe for me. I've done writing essays for other standardized tests (the Florida FCAT) and got screwed over (and didn't find out why). I'm very strong at formal papers, and even made A's in a college level Dual Enrollment Composition course the preceding year, yet somehow I got one of the lowest grades (a 2.5 out of 6) on that portion of the test. Quite odd. And I have read sample 6 papers from the FCAT study guides... laughable. I still don't know how I got such a low score.

Personally, I think that a person's willingness to learn should be the basis of collegiate success, but then again, I am probably a very small minority and that criteria (that being the willingness) can't necessarily be quantized.

 

[mathwonk]

motai,

I agree that willingness to learn, i.e. determination, is absolutely the most important predictive factor in success, in almost all situations. You sound like someone who will be very successful.


I suggest however you consider taking honors courses in college even in subjects you might exempt by your AP scores. AP courses in high school, especially in math, are often nowhere near the level of good honors courses in college.

For one thing they are seldom taught, at least not here in the south, by professors with PhD's and research qualifications, as they are in college. Of course there are exceptions.

Some college professors consciously try not to teach above the level of AP courses. So you have to ask.

 

[juvenal]

Mathwonk - I do agree that it's more expensive to nurture gifted and talented students. However - I just don't think the SAT is a very great substitute.

I don't remember whether I took the SAT once or twice. (I think I took it once in my sophomore year, and then once again in my junior year). I took the PSAT, though, before the SAT. My point of giving the score was not to brag, but rather to make the point that even though I scored very well, I still don't think much of the exam. Also - I wasn't exactly a superstar or anything at Harvard, though I did decently. And I had a very rough time my first semester, even though I went to a decent public school in the northeast.

You mentioned that SAT scores have gone down over 30 years. I haven't seen this data. But - since women and underprivileged minorities tend to score lower on the SAT on average, and the percentages of these groups have gone up, I would assume that this would account for some of the effect. And once again, a high SAT score is only a partial indicator of talent.

Mathwonk - do you not believe that college admissions have gotten more competitive? If the competition level is higher, then shouldn't the caliber of students be higher? This is anecdotal, but I do know people who were admitted into Harvard in the late 80's - family friends - and they were generous enough to let me see their entire applications back when I was applying. I can't imagine that they would still get into Harvard from the mid-90's on, knowing all their qualifications.

I think there are enough other things in a college application (SAT II scores, AP scores, state and national awards, recommendations, essays, extracurriculars, etc) that the SAT score isn't all that important these days. Also - with all of the information available on the internet these days, applicants are just more sophisticated.

I'm not sure what your point about southern students is, btw. Harvard, and other schools, do try to spread their admissions out geographically. If anything, Harvard has gotten more aggressive about trying to recruit students who aren't from prep schools. And this is totally independent of any changes to the SAT over the last decade. I think the Harvard admissions officers know that it's a lazy way out to use SAT scores as a primary way to find good candidates. The best way to find those diamonds-in-the-rough is to do some leg work and visit some high schools in more far-flung parts of the US then the northeast. And that's what they are doing more.

 

[misskitty]

Some colleges starting to look more closely at a student as a whole person instead of just a test score. So even though the test has changed (again), colleges aren't requiring these tests. It helps to have them, but some colleges don't require them. Granted that definitely doesn't apply to everyone. However they are beginning to look more at the person and what they can bring to the college versus what the student's test score is.

 

[mathwonk]

juvenal, somehow i think you are missing my point. I am not arguing that students at harvard were smarter in the 1960's than in the 1990's. I am arguing that the SAT is a valuable tool that is being politically compromised.

one of the reasons you got into harvard was your high sat score. harvard was one of the first schools to exploit the talent pool out there by recruiting the smartest students, and the sat score was one method of identifying them.

a big advantage harvard has is its wealth, as it can afford to support poor smart students who otherwise could not go there. even at stanford or other top schools, they need more of the money the students' parents provide.

of course identification of talent is one of the hardest jobs in the world. I know, I am a member of a math department which is always trying to hire the "best" people, but who are they? how do you spot them? For some reason the analysts always think the analyst candidate is the best and the topologists always think the topology candidate is best. There is always a fight over "strength" versus "fit". ["Well yes your candidate may be stronger but ours meshes well in interest with our current personnel."] Strength should win this argument most of the time, or something will soon go wrong with your program.

harvard and every other school uses many different measures, one of which is the sat. but the sat has always been a test that was believed, i think rightly, to measure raw intelligence, and it seems ETS (not Harvard) is trying to change that. as a professor who teaches these students I regret that. i think it is a mistake.

nothing you mention as an alternative to sats is new. my friend was a student recruiter for harvard in 1962 to find suitable talent in coastal north carolina. (he couldn't find any.) essays and personal interviews and evidence of character have always been important, as have family history, sports, musical or scientific ability, even ones surname. [You may know there are scholarships to harvard available based on your ancestry. My friends who were ex olympic soccer players, or world class tennis players, or sons of founders of famous multinational companies, or descendants of 18th century new england literary familes also seemed to get in easily.]

nothing new is being added that increases the information about a student. but the evidence of intelligence available is apparently being diminished. this makes no sense to me, unless one no longer wants to know who the smart ones are.

I have the greatest respect for harvard. i will say that in 40 years of exposure to various academic institutions, I believe harvard is the one most likely to select people based on excellence. at other schools, other factors weigh more than they do there. that i believe is why they continue to be the best. they choose to be the best. other places choose to be less and they succeed.

Jim Harrick could never have had Tony Cole admitted to Harvard (i hope).

This is not about which harvard classes were best qualified. The SAT is a tool that helps everyone choose more wisely, even schools that are not harvard. I hope it is not being compromised consciously for poltical reasons, as it seems to be now.

 

[juvenal]

Mathwonk - this has been an interesting discussion, but in the interests of productivity, this will have to be my last post on the topic.

Is the SAT a measure of raw intelligence? Assuming equal preparation among test-takers, there is certainly a good correlation. But not perfect. Would a chess aptitude test also measure raw intelligence fairly well, maybe even better than the old-style SAT? Yes. Does it mean that students should start taking exams consisting of 3 hours of chess against a computer? No (at least I don't think so).

Is it always better to have more information via a standardized test? Not always. First - students may overemphasize test preparation at the expense of more important things - such as learning for the sake of learning. Second - admissions officers may subconsciously overemphasize the results of the tests. It's easy to say that "yes - we only consider the SAT as a flawed proxy for future success" , but when you have some number that you can assign to a student, it's too easy to subconsciously place a great weight on that number. Also, it's human nature for people to ignore error bars (the standard deviation of the test result). So there is a such thing as too much information in some cases.

Last point: while the old-style SAT was indeed a flawed, but useful tool in making college demographics more egalitarian (giving non-prep-schoolers and non-legacies a chance) - I think that the combination of a more diverse and sophisticated applicant pool and more modern-thinking, diversity-minded admissions committees has made the SAT somewhat obsolete. If the SAT were to completely disappear today, I don't think things wouldn't change very much, and might even change for the better. (I'm also pretty sure I've seen studies that showed that old-style SAT scores had little correlation with college success, at least when you're comparing scores that vary by 200 points or so).

 

[motai]

Another thing that I noticed about standardized tests is that they are in a way discouraging to many prospective students. I, for one, know that I will probably never make it into an Ivy League school just because my scores alone are too low, even though I am more of the type of person who places a larger emphasis on learning for the sake of learning rather than spending that time on test preparation alone. So in a way those schools are skipping over interested and qualified applicants who have a willingness to learn (and who would probably do well at their institution as well), though their scores may not necessarily reflect their intelligence.

For me, as long as I put the required effort in, no concept is particularly out of grasp or too difficult to understand. Unfortunately standardized tests place a psychological stigma on people like me, causing me to mentally tell myself that I am inferior to someone with a higher score (which is real damaging btw). Then again, I didn't put that great of a time investment in test preparation because I didn't really see a point in the first place. I'd personally rather study something that I don't know, something that is interesting and will benefit me in the future. I think I fell into a hole though, and my logic back then apparently isn't one that most colleges will look on favorably if all they place a greater emphasis on scores. After all, who wants someone who has a willingness to learn if the other candidate has near-perfect SAT scores but might not be that interested in learning new material? I am probably a minority though, most of my classmates tend to do better than me, and I haven't really met anyone else who shares my sentiments. Guess that makes me stupid then

 

[devious_]

I haven't read all these posts yet, but I will when I've got free time. I just logged on to say:

The SATs are crap. I got a 660 on SAT Math, but an 800 on SAT-II Math 2c. I also got As at O-level Math and Additional Math and at A-level Math and Further Math. Plus I absolutely love math and spend large quantities of my spare time studying higher-level material and doing challenge problems. The 660 is probably going to kill my applications. (I'll know in a few weeks!)

That's it.

 

[misskitty]

I had the same problem with my scores. I don't think those test are accurate measures of anything.

I'll give you an example. I reviewed and prepped for my SATs, I wasn't thrilled with my results...(however I was only a sophmore). My friend took them as a senior, guessed on a ton of the questions, and got a 1600. How is that an accurate measure of anything if he guessed? Not comparing him with me and such.

 

[omagdon7]

The one thing I must say is that if you don't know how to do the math on the SAT I but you are in more advanced math it doesn't mean the test is flawed. It means you learn things and then forget them because the SAT I is extremely basic. Anyone who *****es about standardized tests needs to be quiet and realize that you need something that is an objective standard for everyone. The rigor of a school varies so the only things they have to go on to judge that are accurate measures of your aptitude are essays and standardized tests. And though you may feel more qualified than some people who get into MIT, they don't simply let you into MIT based on yourscores, and all things being equal the people with higher scores are obviously the people who merit placement.

 

[misskitty]

Of course. Otherwise there would be little point to get an objective view of a person. I just don't think that your scores are All the college considers. Its more about who you are as a person and not what kind of a test score you are. Some people who score highly, have nothing or vey little to offer because they don't care. But they get in on a technicality like that when someone else is more rounded and cares about what they do? Thats what I was trying to get at: person is a bit more important than the score...its the person attending your school, not the score.

 

[GCT]

The new SAT seems better in my opinion. A better math section, essential reading comprehension/verbal ability section, and writing section...all are essential to succeed academically, college is not the place to struggle around these basic skills. I suppose a higher standard will result add a bit more decency to the college classroom.

Particularly the analogies, in my opinion, were more biased favorably upon the wealthy upper classes.

I don't believe that the ACT has received so much criticism as the SAT, is it being revised also?

 

[devious_]

The question, omagdon7, is whether the SATs are an accurate measure of aptitude or not.

Personally, I didn't grow up in America, so I wasn't really used to taking tests similar to the SATs. For people like me, who are in abundance inside and outside the USA, the SATs are in no way a measure of aptitude but instead are a measure of test-taking capabilities. There are strategies and tactics you can do to minimize damage and walk out with a better score, a score that doesn't accurately assess your grasp of the material. Is that really fair to students who don't have the privilege to learn such techniques?

Some schools, like Yale, have adopted a smart philosophy: they waive the SAT requirement for certain students who, for example, have taken other types of standardized tests like A-levels or the French Baccalaureate. This is a better way to measure aptitude, especially since those tests are, by far, superior to the SATs (for two main reasons: guessing won't get you anywhere and the fact that you get marks for your answers and NOT your final answer -- after all, is the ability to avoid making arithmetic slips equivalent to having high aptitude?).

 

[misskitty]

The new SAT seems better in my opinion. A better math section, essential reading comprehension/verbal ability section, and writing section...all are essential to succeed academically, college is not the place to struggle around these basic skills. I suppose a higher standard will result add a bit more decency to the college classroom.

Particularly the analogies, in my opinion, were more biased favorably upon the wealthy upper classes.

I don't believe that the ACT has received so much criticism as the SAT, is it being revised also?

I'm curious to see how much different the new SAT is. I'm a wicked nervous about taking it to be honest.

As far as I know the ACT is not being revised. They must be happy with the format its in.

I completely agree college is no place to be struggling with any kind of basic skill. However if you don't do as well as you would like to do, there are other ways of displaying your competancy with basic skills (i.e. your essay).

 

[Eratosthenes]

Just remember that doing bad on the SAT is not the end of the world!:)

If you do bad you can still succeed and do well in college.
I never even finished high school and I am now at a decent university studying and doing really well. I learned everything I know about math, english, science, history, politics, computers, etc in college.

During what high school year do people typically take the SAT? When I was in high school I don't remember having to take it. I dropped out in the 10th grade though.