Saturday, October 14, 2006

beautiful lies

Omar: I also heard Malcolm Gladwell speak at the New Yorker festival.



Me: I hate that guy!

Omar: [laughing] He was such an amazing storyteller. He had no slides, just him standing there talking, and the audience was spellbound.

Me: I have no doubt of it. His books are beautiful stories. They really work their way into your heart. But they're depressing and they're lies. For example, he says that 4% of the population is over six feet tall, but some large percentage of CEOs are, like 50%. He concludes that height leads to people thinking you're powerful, and hence success.

Omar: I remember this.

Me: It's really annoying for me, since I'm not as tall as you are, Omar. But there could be some third factor that leads to both! Maybe it's actually that rich families feed their kids well (so they grow up to be taller) and give them more education and connections (so they grow up to be CEOs). Maybe height isn't the cause at all.

Omar: Yeah, there's another part where he says that crime in New York City fell, and it was due to punishing small crimes like littering. But there are studies that have shown that wasn't the reason.

Me: In Freakonomics, they actually did regression tests on all the factors. They ran a scientifically controlled algorithm, and found that it wasn't the small crimes at all. It was by legalizing abortion.

Omar: Right.

Me: At least if Gladwell's books are uplifting, there would be some redeeming value. But they're not. He has another section on sexism, and how it's subconscious and pervasive. He doesn't motivate by saying, 'Hey, let's band together to fight and make a difference!' No, he ends on a note about how it's everywhere and there's nothing we can do because it's so deeply rooted.

Omar: The story of the female trombonist.

Me: His stories are so compelling that it's hard not to be influenced, and a lot of it is not even factual. I fall into all the unfortunate categories that he describes. I'm not six feet tall, I'm a woman, I'm not Caucasian. If I were to believe his books ...

Omar: You're fucked!

Me: His message is: "Well, you're boned. But thanks for buying my book!"

13 comments:

Anonymous said...

but he says such nice things about _you_, niniane

Anonymous said...

People believe all sorts of things based on convient "facts" - just see what the democrats and republicans put out there. Then there is the convient conclusion that increased green house gases definitely cause global warming. I bet you believe that one!

Anonymous said...

Me: In Freakonomics, they actually did regression tests on all the factors. They ran a scientifically controlled algorithm, and found that it wasn't the small crimes at all. It was by legalizing abortion.

I was inclined to trust that result, since it was researched by a top economist and also made some intuitive sense, but it turns out there's a surprisingly strong argument against it. Take a look at http://www.isteve.com/abortion:

As I tried to explain to Dr. Levitt when we debated in Slate in 1999, what happened, simplifying greatly, was that the vast youth crack crime wave first emerged in the later 1980s in the socially liberal states where legal abortion also had taken off first about 17 years earlier, most notably New York and California, which legalized abortion in 1970, three years before Roe v. Wade.

In other words, contrary to Levitt's theory, there was at the state level, a positive correlation (when appropriately weighted by population of state), between the legal abortion rate in the early 1970s and the teen homicide offending rate in the late 1980s and early 1990s among those youths born after legalization. Unfortunately, Dr. Levitt initially only looked at crime rates for the years 1985 and 1997 (and only looked at the uselessly crude age groups of over and under 25), so he completely missed how his theory had catastrophically failed its most obvious historical test.

Second, and also contrary to Levitt's theory, this vast youth murder wave took off first specifically in the demographic group that had the highest legal abortion rate: urban blacks. The non-white abortion rate peaked in 1977, well before the peak of the white abortion rate. The peak years for homicide among 14-17 year old black males were 1993 and 1994 -- i.e., the cohort born at the peak of the black usage of legal abortion in 1977. As Donohue and Levitt wrote in 2001, under their theory, the opposite was supposed to happen:

"Fertility declines [following the legalization of abortion] for black women are three times greater than for whites (12 percent compared to 4 percent). Given that homicide rates of black youths are roughly nine times higher than those of white youths, racial differences in the fertility effects of abortion are likely to translate into greater homicide reductions."

I now think that the beneficial effect of abortion on crime is at best very minor.

Given a data trend, statistical algorithms can tell you the most likely causal factors among a list of candidates you provide. However, if you overlook a key class of candidates, a statistical algorithm will not discover it for you. If you don't choose the most relevant data to analyze, a statistical algorithm will not tell you. The bottom line is, there are still no substitutes for a broad knowledge base and high-quality critical thinking.

Anonymous said...

There was a recent paper that came out that showed that taller people also tend to have higher IQs, at age 3 and throughout childhood. (The early age may point to something about early childhood, like nutrition, that may in turn affect IQ.) The researchers then show that the IQ difference is, in their analysis, enough to explain why tall people earn more. (http://papers.nber.org/papers/w12466)

Anonymous said...

Wow, what a load of crap! Taller people are smarter. What? Are these ladies smoking crack? Their lame academic lives got a bit too boring so they came up with this bullshit. I can't believe they have the audacity to advocate this shit. I cite Napoleon as the most famous example.

Anonymous said...

Wow, what a load of crap! Taller people are smarter. What? Are these ladies smoking crack? Their lame academic lives got a bit too boring so they came up with this bullshit. I can't believe they have the audacity to advocate this shit. I cite Napoleon as the most famous example.

One example does not disprove a statistical tendency. The claim is not that the taller person is always smarter than the shorter person, instead it is that there is a small positive correlation between height and intelligence.

(Of course, as Niniane mentioned, there may be no direct connection between them; a third factor like nutrition may be responsible for raising both.)

Anonymous said...

Well actually they do make this claim "In this paper, we offer a simpler explanation: On average, taller people earn more because they are smarter." Not my words but theirs. They do a disservice to people by saying this. To an impressionable young child there's the problem of self-fulfilling prophecy. A correlation does not a causation make so why blow this up into an inflamatory paper other then to get famous, even if it is for all the wrong reasons.

Dog of Justice said...

Well actually they do make this claim "In this paper, we offer a simpler explanation: On average, taller people earn more because they are smarter." Not my words but theirs.

You are ignoring the words "on average". Napoleon doesn't tell me much about the average intelligence of short men vs. tall men. Large random samples tell me a lot more.

They do a disservice to people by saying this. To an impressionable young child there's the problem of self-fulfilling prophecy. A correlation does not a causation make so why blow this up into an inflamatory paper other then to get famous, even if it is for all the wrong reasons.

*uncloaking sound*

To quote Ilkka Kokkarinen, "even though correlation is not causation, as everyone quickly learns on the Internet, correlation requires an explanation. And more importantly, it invalidates all worldviews that predict that no such correlation would exist." ("Two girls for every boy" entry, January 5th)

Now, I agree with you that it may be good educational strategy to avoid telling this to young kids; ideally they hear it after they have enough self-confidence to not care or even "prove it wrong", and the mathematical sophistication to understand the difference between past correlation and future destiny. This is a very different thing from not ever telling them, though; that approach has enormous negative externalities.

Dog of Justice said...

Oops, I just realized that one of my sentences in the last post can be parsed to mean something totally antithetical to what I believe.

"ideally they hear it after they have enough self-confidence to [not care] or even ['prove it wrong']" is what I intended.

Anonymous said...

I had a friend disagree with the book Freakonomics because it said people who are adopted tend to test lower. His proof? His girlfriend was adopted and was really smart.

Another example: Prisons are disproportionately full of minorities. That can't be true, my girlfriend is black and she's not in prison.

Just because you are an exception to a generalization does not disprove the entire generalization.

mtbouchard said...

A few notes:

I wonder what the correlation is of height to being a GOOD CEO and not just the fact of being one.

The abortion theory about the reduction of crime in NY has been discredited from several sources (as the anonymous commentor noted). It is still widely believed that having a strong public perception that the rule of law is in order via zero tolerance of petty crimes like grafitti or even jaywalking are useful.

And if you want to watch a true master story teller who can mesmorize without props or slides, watch "Swimming to Cambodia" by the late Spalding Gray.

Josekin said...

Awesome post. I agree with you completely on the assessment on MG. Nonetheless, I'd like to point out that his book is merely a collection of researches. He does not have an opinion other than "here are the numbers and these are the facts." Anyway, all these are just statistically significant (height and CEO); they aren't predictive at all. So you have nothing to worry about really.

Anonymous said...

Niniane disdains all persuasive rhetoric that does not place her on a pedestal.

She disagrees that tall people are smarter and have advantages in business and life as Malcolm Gladwell espouses.

Yet she is so into physical appearances. If Niniane were to ever date / marry a short guy with a keg (not a six pack)
I may start to believe that physical appearance (specifically height and proportion) have no bearing on how hard you
have to work to get where you are.

:-) Being a bit shorter than 90 % of my close friends I see the subtle slights from people. (As do they)

But I work harder, make more money then most and tell them to lick my little balls..

Napoleonesque? Maybe.. But the warm tongues feel sooooooooo good. :-P

In the words of Randy Newman...

"Short people have no reason to live"

He makes me laugh as does Malcolm.

You have to laugh.. Or be angry...

I prefer to smile.

:)