Malcolm Gladwell - Outliers (The Story of Success)

Summary by Dan Dascalescu, 2011-May-23
This summary is ~3250 words long. Est. reading time: 10-15 minutes.

Official book website: Outliers
Book word count: ~75,000.
Sumary compression ratio: 23x.
Claimed summary coverage ratio: 95%.

The Roseto Mystery

In the 1950s, before the advent of cholesterol-lowering drugs and aggressive measures to prevent heart disease, the town of Roseto, PA, had a markedly lower incidence of heart disease than neighboring towns despite the similarly unhealthy lifestyle led by their inhabitants of Italian origin (obesity, smoking, diet with 41% calories coming from fat, and including lard and sweets). Exercise was not a typical pastime. Genetics was also ruled out when relatives of Rosetans who lived in other parts of the US were shown to be in no better health than the average American.

The cause of Rosetans' health and longevity could only be attributed to close family ties, dictated by their cultural background: "the Rosetans had created a powerful, protective social structure capable of insulating them from the pressure of the modern world".

The study 1950s studies were conducted by Stewart Wolf and John Bruhn. The book doesn't mention it, but by the '70s, Roseto lost its effect, along with increased Americanization.

1. The Matthew Effect

For whoever has will be given more, and they will have an abundance. Whoever does not have, even what they have will be taken from them.
-- Matthew 25:29

Most of the top Canadian hockey players are born in January and February. Extremely few are born in October-December. The explanation is that in Canada, the "eligibility cutoff for age-class hockey" is January 1, so kids who were born in January have an advantage of almost a year to kids who were born in October-December. At ages as young as 6, this difference is very visible in physical maturity and other sports skills. These kids get further special training, while the others don't, so by the time they're in high school, December kids have no chance against January kids at hockey. They also don't get a chance to accumulate 10,000 hours of practice (see next chapter).

Here is the breakdown of any elite group of hockey players, by Canadian psychologist Roger Barnsley:

  • 40% are born between January and March
  • 30% are born between April and June
  • 20% are born between July and September
  • 10% are born between October and December

If Canadians pick their top hockey kids based on what they think is skill, they are misled. They simply pick the oldest kids, and then train them a whole lot, and they do end up being the best hockey players. This is a beautiful example of a self-fulfilling prophecy.

The phenomenon has been observed in Belgium soccer, where the cutoff date was August 1st, and most of the top kids were born in August and September. A few years after the cutoff date was moved to January 1st, most of the top kids were, predictably, born in January or February.

Sports such as basketball don't suffer from a cutoff date effect, because a player of any age can always find other players and basketball courts. Hockey rings, on the other hand, are much more difficult to get into.

The Trends in International Mathematics and Science Study (TIMSS) shows that among fourth graders, the oldest children scored between 4 and 12 percentile points better than the youngest children. Economist Elizabeth Duhey calls this effect "huge": a score in the 80th percentile can qualify a kid for a gifted program, while a score in the 68th percentile would not.

Parents believe that disadvantages faced by younger children in kindergarten will eventually go away. But they don't. "At four-year colleges in the US, students belonging to the relatively youngest group in their class are underrepresented by about 11.6%."

[T]he outliers in a particular group [reach] their lofty status through a combination of ability, opportunity, and utterly arbitrary advantage.

2. The 10,000-Hour Rule

A study done in the early 1990s by psychologist K. Anders Ericsson at Berlin's elite Academy of Music on the school's violinists showed that:

  • elite performers had each totaled 10,000 hours of practice
  • there were no "naturals", i.e. musicians who effortlessly reached the top while practicing a fraction of the time their peers did
  • there were no "grinds" either, i.e. people who worked very hard but didn't get to the top

Once a musician has enough ability to get into a top music school, the [distinguishing thing] is how hard he or she works. That's it. And what's more, the people at the very top don't just work harder or even much harder than everyone else. They work much, much harder.

Bill Gates and Sun co-founder Bill Joy were hard working young adults, but they also were very lucky:

  • they were born in 1955, right on time to catch start of the explosion of personal computing
  • as an eight-grader, Bill Gates had access to one of the very few time-sharing computer terminals available at the time
  • Bill Joy had access to one of the few computers available for student use, and made use of that access to spend days and nights programming

By the time they reached their 20s, both had amassed 10,000 hours of computer programming experience.

Similarly, The Beatles had performed live 1,200 times by the time they had their first success in 1964. By comparison, most bands today don't perform live 1,200 times in their entire career.

What about young prodigies? Gladwell gives the example of Mozart. While he started writing music at six, his early works were not oustanding, and were all probably written by his father. Mozart composed his first completely original concerto regarded now as a materwork (No. 9, K. 271), at 21.

Ten thousand hours is an enormous amount of time. It's all but impossible to reach that amount all by yourself by the time you're a young adult. You have to have parents who encourage and support you. You can't be poor, because if you have to hold a part-time job [...], there won't be time in the day to practice enough.

On being demographically lucky

Data from the world's top 75 wealthiest figures as of 2008 shows a surprising characteristic: worldwide, and history-wide (from the Egypt pharaohs to today), almost 20% of the richest figures come from a single generation in a single country: Americans born between 1831 and 1840: Rockefeller, Carnegie, J. P. Morgan. They were worn right on time to catch the emergence of Wall Street and the continental sprawl of railroads, which led to immense commerce opportunities.

For computing, the key year is 1975. Steve Jobs, Eric Schmidt, Bill Joy and the three other founders of Sun Microsystems were born in 1954 or 1955.

3 & 4. The trouble with geniuses

IQ does play a role in success in life, but only to a point (an IQ of 120). Beyond that, what makes the difference is the family and community environment, and hard work. Lewis Terman, a professor of psychology at Stanford University has tracked about 1450 Californian kids with high IQs since 1921, and until 1955, before he died. The study, called Genetic_Studies_of_Genius is still going on. However, sociologist Pitirim Sorokin showed that their performance was indistinguishable from that of kids taken at random, from families of similar backgrounds. Terman's fieldworkers also rejected two students who were to become Nobel laureates because their IQs were too low. Terman himself concluded:

At any rate, we have seen that intellect and achievement are far from perfectly correlated.

In the end, only one thing mattered: family background. Almost none of the genius children from the lowest socio-economic class became famous.

The last 25 American to receive Nobel prizes in Medicine or Chemistry came from prestigious, but also from not-so-prestigious colleges (DePauw, Holy Cross, Gettysburg College). Malcolm concludes that to win a Nobel Prize, you have to be smart enough to get into a college as good as Notre Dame. "That's all."

At the University of Michigan law school, under an affirmative action program, each year 10% of the students who enroll must be members of racial minorities. To do that, the university lowers the admission criteria (otherwise, only 3% of its students would come from racial minority groups). White students indeed get better grades, which is not surprising, given that they had higher undergraduate grades than the minority group for which the admission standards had to be lowered. What is surprising, however, is that a followup study by Richard Lempert (PDF) showed that after graduation, the minority students "were doing every bit as well" as the white students:

[...] although an admissions index that combines LSAT scores and undergraduate grade-point average is a significant predictor of law school grades, it does not predict career success on any of our three outcome measures: log of current income, self-reported satisfaction, or an index of service contributions.

Divergence test

How many uses can you think of for the following two objects:

  • brick
  • blanket

A student with an IQ of 180 might say for brick: "building things", "throwing"; and would stop there. A creative student with a lower IQ might say "drawing on the sidewalk", and for blanket, "dampening it in water to help a thirsty person in the desert". Which of them is more suited for the kind of imaginative work that a Nobel Prize requires?

Geniuses such as Christopher Langan fail to realize their potentials (Langan, IQ 195, ended up working on a horse farm in rural Missouri) due to lack of "practical intelligence: knowing what to say to whom, knowing when to say it, and knowing how to say it for maximum effect".

Sociologist Annette Lareau concluded that there are only tow parenting philosophies, and they divided very well along class lines: wealthier parents adopted a style called "concerted cultivation" (the kids are encouraged to attend a variety of activities, and parents are permanently involved in their lives), while poorer parents essentially let kids develop on their own. The result is that "cultivated" kids develop a sense of entitlement and shift interactions to suit their preferences. Kids left to their own devices appear better behaved, less whiny, but develop a sense of distance, distrust and constraint.

5. The three lessons of Joe Flom

Most successful lawyers in New York share these characteristics:

  • of Jewish descent, with parents who worked very hard in the garments industry, ultimately establishing their own shops. Their work was meaningful - it had autonomy, complexity, and a direct connection between effort and reward. These parents fostered an entrepreneurial spirit in their children.
  • born around 1935, in a "demographic trough", when in response to the Great Depression, many families stopped having children. The generation born during that decade was much smaller and benefited from large schools and qualified professors prepared for the large generations before. The job market was also in their favor, with high demand and low supply.
  • initially rejected by established law firms because of their non-Christian origin, the lawyers had to work independently and take cases refused by the big law firms. These cases were most often takeover. The lawyers became very good in this particular area, and when takeover cases exploded in the 70s, they were the best available, while the previously established law firms had gotten left behind.

6. Harlan, Kentucky

Several mountainous towns in Kentucky exhibited an unusual pattern of family feuds that lasted for several generations. This is explained by the origin of the townspeople: Scotch-Irish herders, who brought with them a culture of honor. As a farmer, one's existence depends on collaboration with others. As a herder, it's much more independent. The herder needs to defend his animals and land, to the point of killing another herder's animals who venture on his land. A study on University of Michigan students revealed that those who were most affected by an insult uttered towards them ("asshole") differed from those less affected by their origin: Southern US. This was despite a number of factors: all students were several generations removed from their farming or herding ancestors (if they had any British ancestors at all), they were all fairly cosmopolitan and had traveled significantly, they were of varied size and physical strength etc.

7. The ethnic theory of plane crashes

Industrial and airplane accidents are almost always caused by a number of small defects occurring at the same time. For example, the famous Three-Mile Island incident happened because "five completely unrelated events occurred in sequence". If any of those malfunctions occurred independently, or even two at a time, the plant would have kept functioning without a glitch. For aircraft, it typically takes 7 malfunctions to cause a crash. Since planes have numerous warning and redundancy systems, the key element in causing crashes is human error, and more precisely, the communication between pilots.

Commercial airliners have a captain, and a pilot, and sometimes a flight engineer in the cockpit. If the captain becomes too tired, sick, or otherwise incapacitated, it becomes the job of the other pilot to take over the aircraft. The problem is that in some cultures, the "power distance index" is so high, that the captain is perceived as a demigod who is never wrong. This prevents co-pilots from clearly warning the captain of dangers he missed, and had been the cause of most of the Korean Air crashes until 1999, when Korean Air[lines] (the name was changed after 1997) decided to take measures specifically aimed at addressing the power distance issue. In Korean, there are six forms of address: formal deference, informal deference, blunt, familiar, intimate and plain. It's also the job of the listener to figure out what the speaker is trying to say, in complete contrast to the Western "transmitter orientation" - the speaker being responsible for communicating their ideas clearly and unambiguously.

English as a universal language

To allow pilots to escape their cultural legacy, Korean Air mandated the use of English in the cockpit. The foreign language let pilots have different personas while flying, allowed them to communicate clearly with their superiors, and be assertive when the latter were in the wrong. This reduced the number of Korean Air incidents dramatically, and the airline's record has been spotless since 1999.

Similarly, Cuban pilots considered American Air Traffic Controllers their superiors, and were simply afraid to let them know of critical details about the plane, such as being out of fuel. A Cuban pilot responding to a New York ATC who directs their plane to circle JFK and report on status, can be: "We're circling around the airport, we're good, and uh, we're running out of fuel". (Plane are "running out of fuel" naturally as they approach their destination, and this is not the terminology a pilot would use). This type of speech, in which the speaker attempts to downplay what's being said, is called "mitigated speech". An American pilot would say "Listen buddy, my fuel is almost zero. I have to land".

8. Rice paddies and math tests

In English, the numbering system is quite irregular:

  • we don't say "one-teen" and "two-teen"
  • numerals from "thirteen" to "nineteen" are irregular as well, vs. those over twenty ("twenty-one", "thirty-two", "sixty-four")
  • "thirty", "forty" and "fifty" are like "three-ty", "four-ty" and "five-ty", but not quite (vs. "sixty", "seventy", ..., "ninety")

In Chinese, on the other hand, the system is perfectly regular. After 10, the numbers are expressed as "one-ten-one", "one-ten-two", ..., "ten-nine", "two-ten-one", "two-ten-two", ... "three-ten-seven" etc. Try this exercise: compute thirty-seven plus twenty-two. If you are an English speaker, your brain has to decode that into "three tens and seven" plus "two tens and two". For a Chinese speaker, the information is right there in the question: "three-ten seven plus two-ten two". By the age of five, this keeps American kids a year behind Chinese kids in counting. It also makes math easier to learn, and conceivably more attractive.

Another factor that explains Asians' prowess with math is their culture of work. As rice paddy owners, a Chinese family has to work year-round, and on a very small surface (2 or 3 paddies, each the size of a hotel room). The work is also very complex, involving irrigation at a precise water level, and simultaneous harvesting and planting of the new crop. Wheat or corn, on the other hand, require much simpler work, with almost no work to do between harvest and sowing, and relatively little work to do between sowing and harvest. As a side note, peasant life, even well into the 19th century, consisted of brief episodes of work, followed by long periods of idleness. Moreover, mechanized agriculture makes it possible for an average Midwestern American farm today (450 acres) to be used by just one family, while in China the same area of land feeds a village of 1,500 people.

This culture of work can also be gleaned from folk wisdom: Chinese proverbs sound like "If a man works hard, the land will not be lazy", or "He who wakes up before dawn 360 days a year will never fail to make his family rich". By contrast, Russian proverbs from areas where agriculture consists of growing wheat, sound like "If God does not bring it, the earth will not give it".

Applied to schoolwork, the problem of mastering mathematics, for example, is not one of ability, but one of aptitude:

You master mathematics if you are willing to try. Success is a function of persistence and doggedness.

The Trends in International Mathematics and Science Study (TIMSS) exam shows a striking correlation between the math rankings and the rankings of how many items are answered in a tedious 120-item questionnaire about students' background (parents' level of education, their opinion of math etc.) "They are exactly the same." In other words, countries students whose students take the time to answer the questionnaire, are the same countries whose students do the best job at solving math problems: Singapore, South Korea, China (Taiwan), Hong Kong and Japan. Interestingly, the lowest-achieving Chinese Americans come from the edges of the Pearl River Delta, where agriculture is less intense.

9. Marita's Bargain

The culture of work influences Asians' education in simple but very effective way: their school year is longer than the American school year by as much as 35%: the school year in the US is, on average, 180 days long. In South Korea, it's 220 days long. In Japan, its 243 days long. Asian countries don't have a long summer vacation. No doubt Asian children do better in university once they immigrate to the US - they simply have had much more time to gain an education.

An analysis of the reading skills of American kids shows that from the beginning of the school year until the end, these skills predictably increase. Also predictably, kids from higher-income family do better than kids from lower-income family. What is surprising is that during the summer vacation, the reading skill of higher-income kids also increases, while that of lower and middle-income kids remains constant, or decreases slightly. Explanation: lower-income kids spend their vacation playing or watching TV, while higher-income kids go to summer camp, are taken to museums, or their parents otherwise provide them with educational experiences and opportunities.

Gladwell concludes that "School works, it's just that there's not enough of it for kids who aren't achieving".

In the South Bronx, randomly chosen children from very poor families enroll in KIPP (Knowledge is Power Program) charter schools. By the end of the 8th grade, 84% of them perform as well at math as privileged children from high-income suburban American families. By comparison, in the South Bronx, only about 16% of middle school students are performing at or above their grade level in math.

These kids don't need a very high IQ, shiny laptops, smaller class size (KIPP class size is 35), or teachers with a PhD. They need a chance, and to do hard work: most wake up at 6am, because school starts at 7:25. They study until 5pm, then do homework for a few hours in the evening as well. Saturdays, school goes from 9am to 1pm, and from 8am to 2pm in the summer.

The problem with math education is the sing-or-swim approach." More time spent on math means understanding is more solid, teachers have time to explain things, and retention is better. KIPP's solution is to provide 50% to 60% more learning time.

Outliers are those who have been given opportunities - and who have had the strength and presence of mind to seize them.


Buy me a coffee to sponsor more cool posts like this!

My tags:
Popular tags: