secrets of the sat
absolutely nothing here

Robert Sternberg: He is a Yale University professor of psychology and education and a researcher of human intelligence. He is the author of Successful Intelligence: How Practical and Creative Intelligence Determine Success in Life.  In it, he argues for broadening the definition of intelligence and creating new tools to measure it.
navigation, see text below
We went and filmed at Fair Haven Middle School. What were we seeing there that we wouldn't see in a traditional classroom?

What we are adding to the curriculum is more higher level thinking, so in addition to whatever may already be there, we'll have some analytical activities, such as analyzing what happened on this national road or why there was a use for it or why there was a need for an underground railroad in this country during the times of slavery.

There will be amplified creative activities and these would be things like if you're planning a family reunion, how would you actually do it? What kinds of gifts might you create? What kinds of menus might you create? So, it's something where you have to plan something, you have to design something.

And then some of the activities would be things that are practical, like how would you implement this reunion? How would you make it work? What are the difficulties if you have an underground slave railroad in making that work so that it's not detected? And the idea is to encourage this analytical, creative and practical thinking.

Are there kids who may actually be benefiting from this? Why?

My concern, given the relatively low predictability of the tests, is that there may be people who have tremendous talents--creative and practical talents--who, because they don't do well on tests, never get the chance to show what they really could do in important jobs. Well, this is a good example of the idea that kids learn in different ways. The example here is some of the stuff we'll do is visual, with pictures and graphs and image things. Some of it is more verbal. If it's verbal, it can be auditory or written, but that applies not just in reading, it applies to anything. I've taught statistics, math courses and what I've found is that often if you teach them algebraically the formulas, you'll have one group of kids doing well. And a teacher might conclude that those are the smart kids and the rest of them are dumb and I was that teacher, so I've been through it.

Then they said, hmm, I don't happen to like what you're talking about. I don't like visualizing things. I'm not good at it. So that year, I started not only teaching with the formulas, algebraically, but geometrically, with more diagrams and with more figures and an interesting thing happened. A whole bunch of kids who had not been doing well, now started really grasping the material well and some of the kids who had before been doing really well now found it difficult to understand. The principle is general.

In other words, if a teacher only teaches in one way, then they conclude that the kids who can't learn well that way don't have the ability, when, in fact, it may be that the way the teacher's teaching is not a particularly good match to the way those kids learn. And if the teacher tried another way of teaching, she or he might find that a lot of kids who now seem dumb really are able to learn the material.

What do you say to the people who say, "does practical or creative thinking or assessments of those abilities predict anything that's going to get my kid into Princeton?"

None of the criteria we use are really soft. All of the studies we do in my group are quantified. In other words, unlike some people with new theories, we will go out, we'll go into a school and we get products and the products are evaluated, whether it's by teachers or others. The scores are quantified and then we compare performances.

There's sort of two questions. One is, do these kinds of creative and practical skills really matter in life? Well, we've been studying that for now almost 20 years and in probably two dozen studies, we've shown that, for example, measures of practical abilities predict job performance as well as or better than do IQ tests and IQ like tests. But they do not correlate with, relate to these IQ tests.

So, we have the statistical data. It's been published in scientific journals that show these kinds of tests, to the extent you care about what people really do on the job. And I'm talking about a variety of jobs, sales people, managers, military officers, elementary school teachers, college professors, janitors, clerks, secretaries, so we're talking really the range and these tests do predict success on the job. That's question #1.

Question #2 is, will it help them get into Princeton? I hope not. I hope it helps them get into Yale, but that little detail aside, the answer again is yes. And the reason is that people have the misunderstanding and I think anyone in the Yale admission office would say the same thing. In fact, I worked in the Yale admissions office at one time, so I think I have something of the mind set.

People might think that the only thing they care about is SATs and that's totally not true. The reason it's not true is #A, they could fill a class with people who have high SATs and still have plenty more with high SATs. What they look for is not just high SATs, but people who are distinguished among those, who have creative and practical talents. That's #1.

#2 is you'll find that many people are accepted who don't have the high SATs, but they have some degree of distinction in something else because the competitive colleges don't want just a bunch of people who are good at getting As, they want diversity. They not only say it, they believe it. And I'm not just talking about any one school.

So they'll look for musicians, they'll look for writers, they'll look for potential scientists, potential engineers, potential athletes. They look for a wide range of abilities.

And so, you're actually doing your kids a favor if you help them develop these other skills, not only for their success in life but for giving them something that will distinguish them from other kids whose academic records and test score may be fine, but who don't have anything else that sets them off.

You've got a three part theory of intelligence but what you're doing is you are forgetting about a single G factor. Is that's what's important?

Well, we're not forgetting about it at all. The three parts of the theory are analytical ability, the ability to analyze things to judge, to criticize. Creative, the ability to create, to invent and discover and practical, the ability to apply and use what you know.

And analytical ability is very similar to conventional notions of so-called G. Where we go beyond those conventional notions is with creative and practical abilities.

Now, you may wonder then, well, why do people get the so-called general factors? Well, the reason is because they limit their tests to analytical tests. If we just give analytical tests, and we have, then we'll get a G factor, too. So, if you restrict the range of tests you give, so that you only look at a narrow spectrum of abilities, and you restrict the kinds of situations in which you give them. You give people paper and pencil tests in a classroom to measure traditional skills, then we'll get a G factor. But we go beyond that.

Let's take the issue of the so-called general factor or G, which is an ability that some people believe pervades, is common to, all tests that require any brains at all. We did a study to show that this is, we believe a myth, that it's common to some tests, academic type tests, but it's certainly not common to all.

Well, first of all, we did lots of studies where we show practical intelligence doesn't correlate with G. We have probably two dozen studies that practical intelligence better predicts job success than IQ.

But let me tell you a study we did in Kenya. We did the study among little children in a rural village and we said, all right, what do you need to adapt in a Kenyan rural village? Well, one thing you need, you need to know the names of medicines to medicate yourself and the reason is, that almost 100% of the kids have intestinal parasites, worms, and that's really--it can be a fairly serious illness. It can actually kill you, if you have a bad enough case.

So, knowing how to medicate yourself is as important to those kids as knowing how to eat food is here. It's a part of daily life and they use this knowledge continually in their lives. So, we created a test of something that's important to these kids and we correlated scores on that test with their scores on IQ tests, tests of G, general ability.

What did we find? We found that the correlation was significantly negative. In other words, the better they did on the IQ test, the worse they did on the practical test and the better they did on the practical tests, the worse they did on the IQ test.

Now, why? I'll tell you. In our society, we assumed that it's really important to do well in school, to get good grades, to be able to get into the meritocracy, as we call it. But suppose that the roots are different for some kids, not just in Kenya but here, too. Suppose a kid wants to be a plumber. Why not? It pays really well, probably better than I get paid, or a carpenter, or any of a number of occupations where college and graduate school and law school are not the route to success.

So, why is that kid going to invest his time learning these academic kinds of skills? That's not going to get him a good salary, it's not going to get a nice house, it's not going to get resources for the kids. Rather learning the trade will be what's important.

And that's what we found in Kenya. To the Kenyan families, school doesn't really matter because none of them are going on to college. Almost all of drop out of school and so, they're spending their time learning things that are important to them.

The only point I'm making is that once you get out of this very small box we've created for ourselves, I believe the general factor evaporates. But what many psychologists have done, probably because they did well on a test themselves and everyone wants high self esteem, is to create this little box and then do their research inside it. And so, you can do hundreds and hundreds of studies showing a general factor and just so long as you restrict your populations, your testing materials and the kinds of situations you look at, you can keep finding the same wrong thing again and again.

So, what about people who say, well, schools are for academic subjects only and that creative and practical skills should be ruled out of the academic assessment?

Well, there are two reasons why I don't think that's true. Let's make it three. The first is, what our studies show is that if kids learn creatively and practically, they learn better, even if the tests are for memory. So, if your goal is the same as mine, which is that the kids learn as much material as possible, learn it well and they're able to use it, then this is the way to do it. The way to do it is not to cram them with facts and most parents know that they weren't good at learning that way. So, that's point #1.

Point #2 is, to the extent that you want the kids to leave school with a good feeling about it, rather than a bad feeling and to want to spend their time on it, to feel like this is worthwhile, then why bore them to death? If you bore them to death and say, this hurts me more than it hurts you, #A, they're not going to believe it, and #B, they're going to invest their time in other things anyway. They're going to get interested in other things, maybe anti-social things.

The third point is what do you really care about? Don't you care about ultimately whether your kid is successful later in life? That seems important to me. And in order to succeed in later life, you need creative skills because look at how fast the world is changing. When I grew up, there were no computers that anyone used, except for the big scientists. There was no internet, there were no VCRs. If you're not adapting to the very rapidly changing environment, if you can't think creatively, you lose big in this society because there are very few jobs for you left. That's #1.

If you look at practical abilities, how many people do you know who, the last really important thing they did, they'll mention at a cocktail party, they'll slip in that they were Phi Beta Kappa or that their SAT scores were high 700s. And the reason they slip it in is that's the last important thigh they did. They have nothing else to talk about.

I'm not saying everyone's like that. There are people who do well on these tests who are successful. But the people who do well on the tests who are successful, it's because they have something else.

You've been interested in IQ for a long time...

IQ is a mess for me in that when I was very young, I did really poorly on IQ tests. I was very test anxious or maybe just stupid, I don't know. But in any case, I did poorly on the tests and so, in the first three years of school, I had teachers who thought I was stupid and when people think you're stupid, they have low expectations for you. And, oddly enough, you tend to want to meet their expectations. That's human nature.

And it's also human nature that if you meet someone's expectations, they're happy, because they were right. So, they expected me to be dumb. I acted dumb. They saw that they were right, they're happy. I'm happy that they're happy. And pretty much everyone's pretty darn happy. And what can happen and this happens to a lot of kids is that each year, they go down a little bit more. It's a downhill fall.

In my case, I had a fourth grade teacher who thought I was smart, for whatever reason. She had high expectations. She wasn't satisfied to say, you have a low IQ, so you should be a crummy student. And so, I started being an A student to meet her expectations and that turned around my career.

And my concern is, what if you don't have that teacher? And most kids probably don't. Then, you never get the chance in this society and the reason you don't get the chance is because you get a little older and your achievement's low and then the SATs don't save you because you're already on this downward swing. You don't do well on the SATs and then if you don't do well on those or the ACTs, the competitive colleges all use them or almost all use them, so it's hard to get into any of those.

And then, if you want to go to graduate school, law school, graduate school requires the GRE, law school requires the LSAT, business school requires the GMAT and everywhere you look, there's another test. So, if you don't do well on that kind of thing, everywhere you turn, the access routes to success in our society are blocked.

So, should we get rid of the SAT?

I don't frame it that way. It's not a question of getting rid of the SAT, it's that the SAT, I believe, as it's now constituted, is too narrow. And if we don't have a test, what we may end up doing is going back to what this country has done before. We could use social class and we still do, but in the 50s, it was, do you have the right last name and are your parents in privileged positions?

We could use religion, as we have. Places like Yale used to have quotas for certain religions. I think those times are well past. And we can use race. In other words, we could substitute things that are much worse.

The problem isn't quite with the SAT, it's with the use of it. What we need to do is include other kinds of abilities, such as creative and practical abilities and at the same time, recognize that they're modifiable, that all these tests can do is give us an approximate reading at a given time.

Do you think it's good, right, proper, okay for kids to begin training for their SAT's five years ahead of time?

Well, I think it's a waste of time to train for the SATs five years ahead of time. The best thing to do five years ahead of time is to invest in getting a really good education and that in itself will probably help scores. If there's going to be an SAT, it's probably practical to invest in a book or perhaps in a course, but I'm sorry to say, I went to some classes that my kids took and it was clear in school that what they were doing was just SAT training.

In terms of educating the kids, this is really a waste of time. So, I would hope that the society would have smarter goals than that.

You said labels don't just describe something. Labels shape reality. What did you mean by that?

What I mean by that is when a person is labeled, it's not just the description. It can become a reality. So, for example, if a child is labeled as having a learning disability, it has very concrete consequences for the kinds of services and potentially accommodations that child will get. So, regardless of what that child can do, how well they can ready, for example, their world is changing as a result of having the label.

If you're labeled as stupid, then people have certain expectations for you.

It sounds like the typical stereotype doesn't apply only on an individual isolated basis but throughout society because of a perceived racial image.

The stereotype thereat can operate with any group of people at all. It can operate on the basis of race, it can operate on the basis of religion, in some countries, on the basis of caste, whether you're nobility or a commoner. So, many factors affect it, but if you keep hearing messages that you're this kind of person, why wouldn't you believe it? If you hear it enough, you'll start to think almost anything is true.

Are you saying that we're wasting talent by clinging to one size fits all testing?

Well, the idea that traditional tests shortchange society is really a simple one, and that is, ask yourself, what do you need to succeed on the job? If you were to make a list of traits, you might come up with things like a sense of responsibility, you need some good ideas or you need to show initiative, you need to work, you need to know how to get along with other people, whatever list you come up with.

And I've asked people in groups when I speak to them to just make a list of traits you need to succeed on the job or to succeed in your personal relationship. And then I'll ask them, now, make a list of traits that you need to succeed on an IQ test or an SAT or even in an introductory college course that prepares you for your job.

And the overlap is very small. Research has shown that IQ type tests account for about 10% of the variation in how successful people are in various aspects of their adult lives. 10% isn't much and, maybe it's a coincidence, but when I ask people what it takes to be successful on the job or in a personal relationship and what it takes to be successful on one of these tests, or in an introductory classroom, the overlap is probably about 10%.

So, if we're determining peoples' futures in large part by traits that are only slightly relevant to success on the job, that's frightening, because it means, #A, we may be enabling people who don't have much competence for those jobs to go into them and we all know people like that in our own occupations who just don't really have it. And there may be other people who could be much better at the jobs who never get the chance because of their low test scores. And I think that's really a tragedy.

John Katzman says the SAT is a scam. I take it you don't agree with that.

Well, I don't think that the SAT is a scam. I see the problem as a little different. The way I see it as a technology, it was a really good technology, it was state of the art about 60, 70 years ago. The problem is that there are very few technologies that essentially haven't changed for 60, 70 years. 60, 70 years ago, computers, what? VCRs, telecommunications. If you look at any technology, they evolve very quickly. Why? Because there's a lot of competition and if you don't evolve and come up with better and better products, you're out of business.

Testing hasn't evolved that way. Different test publishers have their own niches and there hasn't been whole lot of competition in the big testing. ACT and SAT each have their own parts of the country. The GRE has its lock on graduate admissions. And so, one could blame the companies, but really, economically, they have no incentive to change things very much because they're getting the business.

So what I'm hoping is that the incentive can come from somewhere that they will more want o innovate in the content of the test. The problem again is the system. It's very easy to point to one party or another, but I see the system really needs to be changed. And it can be changed and that's what I hope I see in my lifetime.

But how? ETS or whoever, ACT, will continue, as long as they're making money off of this.

You said that you talked to Henry Chauncey. Henry Chauncey was a great leader. What he crated may or may not have been everything we'd want, but at its time, what he was doing was really innovative. What he was trying to do is to give kids a chance who didn't have the family status or the family clout or the money, because they'd have the chance of getting by high testing scores.

At the time, that was a big innovation. Today, we realize that the system has a lot of flaws. What's needed is leadership and we haven't had it. It's that simple. Changes could be made but we haven't had that kind of leadership and I wish that the boards of trustees would get off their chairs and establish that kind of leadership. It can be done. It can be done right now.

Let's say ETS calls you up and suddenly, it's "Robert Sternberg--head of the testing change plan for the next five years." What do you do?

First of all, I want to make it clear that I'm not looking for a job there, but the second thing is that we now have the wherewithal to devise tests that measure creative skills. They're not going to be perfect tests. The analytical tests when they started weren't so great, either. We have the wherewithal to start measuring practical skills. They're putting almost zero budget into really innovative research. They have some research budgets. And I'm not talking about ETS in particular, these testing companies, but most of the research is not directed at the distant future.

In our own research, for example, when we tested creative and practical, as well as analytical abilities, what we found is that on the analytical tests, the kids who did well were mostly white. They were mostly upper middle class. They mostly went to so-called good schools. They looked like a typical high SAT group and they should, because they're high analyticals.

When we looked at the high creative and the high practicals, we found that they were much more diverse and this was not an affirmative action story. It's simply by expanding the range of skills you test, inevitably you find that some of the kids who didn't look so talented before now really look quite smart.

Before Proposition 209 passed in California, UC and especially UC Berkeley relied very much on SATs. It was probably the most exclusive SAT admit in the country.

Universities, private schools, colleges and the like use numbers, so that's why I say, given that that's probably what they're going to do, having been an admissions officer, I know that when I'm reading applications and as the night goes on and I'm reading more and more, it gets more and more tempting to count the SATs. It's easier than reading these long essays and teacher recommendations. It's human nature.

So, that's why I say what we need to do is make better numbers. Given that humans will always want things in easy form, then let's at least have the numbers represent a wider range of skills.

Have we just substituted one elite for another?

We have substituted one elite for another. One of the points I make in Successful Intelligence is that all societies do that and it's not going to go away. What differs is the criteria they use. So, as I was talking about before, we've decided on test scores to be our image of who should get ahead, but socio-economic class can count a lot, here and elsewhere. See, whoever gets power in their given system thinks it's a good system, because heck, they benefited from it, so it must be good.

So, once you get people of high test scores in power, which we now have, why shouldn't they think test scores are good? They're succeeding and they did well on the tests. If you have people who are wealthy getting into positions of power, they would say, who cares about test scores? Test scores, those are pointy headed intellectuals. It's the ability to make money that shows who's good. Or it could be done by religion.

I'll give as an example, why don't you use height? So, what we could do, since we already do use height. We know that CEOs are taller than the people who work for them. Military generals are taller than people below them. Taller presidential candidates almost always win. And the advantage to height is that it can be measured objective. It can be measured reliably. You can't buy a book, you can't take a course. It's hard to cheat. If you wear elevated shoes, it's easy to detect.

So, now let's say, to get into Harvard, you need to be 7 feet tall and maybe to get into Yale, you have to be 6' 11", but good looking, and then you go all the way down to Squeedunk, where you only have to be 3' 2". But now, it's going by height. And to get into law school, you have to be 7' 3" or something. Well, in 25 years, we're going to find that the people running things are tall and tall people are going to think that, yeah, look at how tall people are succeeding.

So, the point I'm trying to make is that every society creates its own meritocracies. I don't think that's going to go away. The question is, if that's going to happen to a greater or lesser degree, then how do you want to do it? How do you want to distribute resources fairly? Some countries do it randomly.

Let's look at the people who are going to have the most to contribute to the world. And then, I think, we'll look for things like analytical, creative and practical abilities and perhaps wisdom, too.

Is the SAT an IQ test or is it not?

Well, ETS certainly would say it's not, but all of these tests correlate very highly with each other and no one at ETS would say anything differently. So you can call it whatever you want, for reasons of political correctness or for reasons of being careful, whatever, but these so-called G-based, general ability based tests for ability that isn't really generally, tend to highly correlate with each other.

You talk about analytical abilities being the best predictor of academic performance. Is it?

This is what's so upsetting. You can make any test the best predictor, simply a matter of what you care about. If you care about the kids being very original and having their own ideas, then creative abilities will be the best predictor. If you care about whether they can use what they learn in their everyday life, then practical abilities will be the best predictor.

But we tend to create these multiple choice or fill in the blank tests that are very similar to the kinds of things that are measured by tests like the SAT and that's why they work. There's nothing sacred or mysterious or God given about the fact that tests like IQ tests or the SATs predict well. The only reason they do is because that's the way we teach and assess our kids. If we change the way we teach and assess our kids, the other kids start to do well.

You said it was a tragedy for those who get lost because we rely on certain tests.

I think the main thrust of my remarks today is that as a society, we want to create the best world we can. We want to be a leading society, we want the individuals to be happy and successful and the question is, how can we achieve productivity and satisfaction. I think it is very sad that by relying on a very narrow band of tests, there are some people who do well on those tests and who will do perfectly fine. But my concern, given the relatively low predictability of the tests is that there may be people who have tremendous talents, creative and practical talents, who, because they don't do well on these tests, never get the chance to show what they really could do in important jobs.

And on the other hand, there were people who do well on these tests and that's kind of what they do well and so they go into positions of power but they turn out to be good in memory and analytical skills, but sometimes rather poor in having good ideas and knowing how to get along with people, and being effective in their lives.

And I think that's that other side of the coin and that's tragic, too. So, it's not to say that you cant have high test scores and be effective or low test scores and be ineffective, but in the tests that are really so modest in their validity, we put too much reliance on them...

To the extent that we want diversity, one way of getting it is affirmative action, but I personally think a better way of getting it is to create broader tests and to assess students for abilities and talents where right now, they're not getting the chance to show what they can do. And what we found in our research is that when we created a broader test that measured creative and practical abilities, the kids who did well on those were broader spectrum. They were more racially and technically diverse, they were more economically diverse, they were more educationally diverse.

So, I think this represents another route. Rather than making a political decision, let's test for a broad range of abilities that truly are relevant to success in life and then greater diversity in college campuses and elsewhere will naturally come as a result of that.

home | discussion | who got in? | interviews | the race issue | sat & test prep | history of the sat
the screening process | test score gap | getting in to berkeley | bibliography | links | tapes & transcripts | press | synopsis

FRONTLINE | pbs online | wgbh

web site copyright 1995-2014 WGBH educational foundation



../test/ ../race/ ../interviews/ ../who/ ../talk/ ../