Outliers, The Story of Success
3.
One of the most widely used intelligence tests is something called Raven's Progressive Matrices. It requires no language skills or specifie body of acquired knowledge. It's a measure of abstract reasoning skills. A typical Raven's test consists of forty-eight items, each one harder than the one before it, and IQ is calculated based on how many items are answered correctly.
Here's a question, typical of the sort that is asked on the Raven's.
Did you get thatI'm guessing most of you did. The correct answer is C. But now try this one. It's the kind of really hard question that comes at the end of the Raven's.
The correct answer is A. I have to confess I couldn't figure this one out, and I'm guessing most of you couldn't either. Chris Langan almost certainly could, however. When we say that people like Langan are really brilliant, what we mean is that they have the kind of mind that can figure out puzzles like that last question.
Over the years, an enormous amount of research has been done in an attempt to determine how a person's performance on an IQ test like the Raven's translates to reallife success. People at the bottom of the scalewith an IQ below 70are considered mentally disabled. A score of ioo is average; you probably need to be just above that mark to be able to handle college. To get into and succeed in a reasonably competitive graduate program, meanwhile, you probably need an IQ of at least 115. In general, the higher your score, the more education you'll get, the more money you're likely to make, andbelieve it or notthe longer you'll live.
But there's a catch. The relationship between success and IQ works only up to a point. Once someone has reached an IQ of somewhere around 120, having additional IQ points doesn't seem to translate into any measurable real-world advantage.*
“It is amply proved that someone with an IQ of 170 is more likely to think well than someone whose IQ is 70,” the * The “IQ fundamentalist” Arthur Jensen put it thusly in his 1980 book Bias in Mental Testing (p. 113): “The four socially and personally most important threshold regions on the I Q scale are those that differentiate with high probability between persons who, because of their level of general mental ability, can or cannot attend a regular school (about IQ 50), can or cannot master the traditional subject matter of elementary school (about IQ 75),can or cannot succeed in the academic or college preparatory curriculum through high school (about IQ 105), can or cannot graduate from an accredited four-year college with grades that would qualify for admission to a professional or graduate school (about IQ 115). Beyond this, the IQ level becomes relatively unimportant in terms of ordinary occupational aspirations and criteria of success. That is not to say that there are not real differences between the intellectual capabilities represented by IQs of 115 and 150 or even between IQs of 150 and 180. But IQ differences in this upper part of the scale have far less personal implications than the thresholds just described and are generally of lesser importance for success in the popular sense than are certain traits of personality and character.”
British psychologist Liam Hudson has written, “and this holds true where the comparison is much closerbetween IQs of, say, 100 and 130.But the relation seems to break down when one is making comparisons between two people both of whom have IQs which are relatively high....A mature scientist with an adult IQ of 130 is as likely to win a Nobel Prize as is one whose IQ is 180.”
What Hudson is saying is that IQ is a lot like height in basketball. Does someone who is five foot six have a realistic chance of playing professional basketballNot really. You need to be at least six foot or six one to play at that level, and, all things being equal, it's probably better to be six two than six one, and better to be six three than six two. But past a certain point, height stops mattering so much. A player who is six foot eight is not automatically better than someone two inches shorter. (Michael Jordan, the greatest player ever, was six six after all.) A basketball player only has to be tall enoughand the same is true of intelligence. Intelligence has a threshold.
The introduction to the / vs. 100 episode pointed out that Einstein had an IQ of 150 and Langan has an IQ of 195.Langan's IQ is 30percent higher than Einstein's. But that doesn't mean Langan is 30percent smarter than Einstein. That's ridiculous. All we can say is that when it comes to thinking about really hard things like physics, they are both clearly smart enough.
The idea that I Q has a threshold, I realize, goes against our intuition. We think that, say, Nobel Prize winners in science must have the highest IQ scores imaginable; that they must be the kinds of people who got perfect scores on their entrance examinations to college, won every scholar-
ship available, and had such stellar academic records in high school that they were scooped up by the top universities in the country.
But take a look at the following list of where the last twenty-five Americans to win the Nobel Prize in Medicine got their undergraduate degrees, starting in 2007.
Antioch College Brown University UC Berkeley University of Washington Columbia University Case Institute of Technology MIT Caltech Harvard University Hamilton College Columbia University University of North Carolina DePauw University University of Pennsylvania University of Minnesota University of Notre Dame Johns Hopkins University Yale University Union College, Kentucky University of Illinois University of Texas Holy Cross Amherst College Gettysburg College Hunter College N o one would say that this list represents the college choices of the absolute best high school students in America. Yale and Columbia and MIT are on the list, but so are DePauw, Holy Cross, and Gettysburg College. It's a list of good schools.
A long the same lines, here are the colleges of the last twenty-five American Nobel laureates in Chemistry:
City College ofNew York City College ofNew York Stanford University University of Dayton, Ohio Rollins College, Florida MIT Grinnell College MIT McGill University Georgia Institute of Technology Ohio Wesleyan University Rice University Hope College Brigham Young University University of Toronto University ofNebraska Dartmouth College Harvard University Berea College Augsburg College University of Massachusetts Washington State University University of Florida University of California, Riverside Harvard University To be a Nobel Prize winner, apparently, you have to be smart enough to get into a college at least as good as Notre Dame or the University of Illinois. That's all.*
This is a radical idea, isn't itSuppose that your teenage daughter found out that she had been accepted at two universities Harvard University and Georgetown University, in Washington, DC. Where would you want her to goI'm guessing Harvard, because Harvard is a “better” school. Its students score a good 10 to 15 percent higher on their entrance exams.
But given what we are learning about intelligence, the idea that schools can be ranked, like runners in a race, makes no sense. Georgetown's students may not be as smart on an absolute scale as the students of Harvard. But they are all, clearly, smart enough, and future Nobel Prize winners come from schools like Georgetown as well as from schools like Harvard.
The psychologist Barry Schwartz recently proposed that elite schools give up their complex admissions process and simply hold a lottery for everyone above the * Just to be clear: it is still the case that Harvard produces more Nobel Prize winners than any other school. Just look at those lists. Harvard appears on both of them, a total of three times. A school like Holy Cross appears just once. But wouldn't you expect schools like Harvard to win more Nobels than they doHarvard is, after all, the richest, most prestigious school in history and has its pick of the most brilliant undergraduates the world over.
threshold. “Put people into two categories,” Schwartz says. “Good enough and not good enough. The ones who are good enough get put into a hat. And those who are not good enough get rejected.” Schwartz concedes that his idea has virtually n o chance o f being accepted. But he's absolutely right. A s Hudson writes (and keep in mind that he did his research at elite all-male English boarding schools in the 1950s and 1960s), “Knowledge of a boy's IQ is of little help if you are faced with aformful of clever boys.”''
Let me give you an example of the threshold effect in action. The University of Michigan law school, like many elite US educational institutions, uses a policy of affirmative action when it comes to applicants from disadvantaged backgrounds. Around 10 percent of the students Michigan enrolls each fall are members of racial minorities, and if the law school did not significantly relax its entry requirements for those studentsadmitting them with lower undergraduate grades and lower standardized-test scores than everyone elseit estimates that percentage would be less than 3 percent. Furthermore, if we compare the grades that the minority and nonminority students get in * To get a sense of how absurd the selection process at elite Ivy League schools has become, consider the following statistics. In 2 0 0 8 , 2 7 , 4 6 2 of the most highly qualified high school seniors in the world applied to Harvard University. Of these students, 2,500 of them scored a perfect 800 on the SAT critical reading test and 3,300 had aperfect score on the SAT math exam. More than 3,300 were rankedfirstin their high school class. How many did Harvard acceptAbout 1,600, which is to say they rejected 93 out of every 100 applicants. Is it really possible to say that one student is Harvard material and another isn't, when both have identicaland perfectacademic recordsOf course not. Harvard is being dishonest. Schwartz is right. They should just have a lottery.
law school, we see that the white students do better. That's not surprising: if one group has higher undergraduate grades and test scores than the other, it's almost certainly going to have higher grades in law school as well. This is one reason that affirmative action programs are so controversial. In fact, an attack on the University of Michigan's affirmative action program recently went all the way to the US Supreme Court. For many people it is troubling that an elite educational institution lets in students who are less qualified than their peers.
A few years ago, however, the University of Michigan decided to look closely at how the law school's minority students had fared after they graduated. How much money did they makeHow far up in the profession did they goHow satisfied were they with their careersWhat kind of social and community contributions did they makeWhat kind of honors had they wonThey looked at everything that could conceivably be an indication of real-world success. And what they found surprised them.
“We knew that our minority students, a lot of them, were doing well,” says Richard Lempert, one of the authors of the Michigan study. “I think our expectation was that we would find a halfor two-thirds-full glass, that they had not done as well as the white students but nonetheless a lot were quite successful. But we were completely surprised. We found that they were doing every bit as well. There was no place we saw any serious discrepancy.”
What Lempert is saying is that by the only measure that a law school really ought to care abouthow well its graduates do in the real worldminority students aren't less qualified. They're just as successful as white students.
And whyBecause even though the academic credentials of minority students at Michigan aren't as good as those of white students, the quality of students at the law school is high enough that they're still above the threshold. They are smart enough. Knowledge of a law student's test scores is of little help if you are faced with a classroom of clever law students.