« If Not Data, What? | Main | Dan Kahneman Puzzles With Us »

October 26, 2007


In this sense, God has screwed over each and every one of us- in three billion bases of DNA, there's bound to be alleles which we really don't like.

What I find amazing is that no article I read actually quotes Watson as saying Africans have lower IQs. What he said was that he is “inherently gloomy about the prospect of Africa” because “all our social policies are based on the fact that their intelligence is the same as ours – whereas all the testing says not really.”

His claim was ONLY that Africans' intelligence is different than "ours."

Is there much doubt as to his meaning? Perhaps not, but I should think on this blog we would not commit the sin of assuming too much.


1. Intelligence describes some quality that individuals have.
2. Individuals have different aamounts of or ability in that quality.
3. Larger groups of can be created by grouping these same varying individuals and this group can be said to have a group average in terms of intelligence.

Some conclusions that follow from these assumptions:
1. Any group of people, divided by almost any arbitrary measurement into a large enough sample will differ in the average amount of intelligence each particular subgroup has.
2. Arbitrarily dividing the worlds population into "races" based on their skin pigmentation will create subgroups that also have a differing average intelligence.

Therefore, different racial groups have a different average intelligence. This logic doesn't depend on what exactly you define as intelligence, as long as something is defined as intelligence. This logic doesn't imply that any particular racial group must have more intelligence than any other particular racial group. It just implies that by definition, individuals are different from each other and thus groups of individuals will also tend to differ in their different characteristics.

Now for the fun part.

Does it matter that different races have varying "average" intelligences? Only if you are racist and judge people by what arbitrary "race" you've mentally classified them as. To me, if you insist on classifying people based on their physical appearance, it would make more sense to classify people by height, or eye color, or their weight. Those attributes tend to make more of a difference, genetically.

However, perhaps you find race an easy way to filter individuals into categories. In that case, based on the extensive evidence, in some geographic regions, people with darker skin have a lower average intelligence as judged by the best tests and definitions of intelligence available. What's interesting is that in some other geographic regions, people with darker skin have a higher average intelligence.

So used as a bayes filter, skin color isn't going to typically give you the ability to improve your judgement of intelligence unless you have other demographic information that has a much better correlation.

In actual fact, if you want to judge a person's intelligence based on some quick rules of thumb, there are much better demographic methods available to you than skin color. National origins, wealth levels, education obtained (for older individuals), etc...

On the other hand, if you really care that much about it, just ask them to take whichever intelligence test you prefer for you.

But to say that races differ in intelligence? That statement should be as non-controversial as a statement of fact as it is useless as a guide to actions and policy.

Sharper, if the divisions are truly arbitrary and on a basis uncorrelated with individual intelligence then the difference in population average will be (crudely speaking) on the order of 1/sqrt(N) where N is the size of each population. N is so large here that the sorts of interracial "differences" that are inevitable on the grounds you give are also clearly negligible, and no one who claims that races do, or don't, differ in intelligence is talking about such tiny differences.

Eliezer, leaving aside the bit about God, I think there's at least one way in which differences (in actual intelligence, or in others' estimates of your intelligence, or in other things that matter) can be worse when they're distributed according to race: network effects. Nations, societies, cultural groups, families, etc., tend to be somewhat homogeneous by race. So if you belong to a race that's systematically poor (to take an example that uncontroversially does occur) then you're doubly screwed: you have no money, and all the people around you who might have been able to help you also have no money, and all the social structures that might have been built up to help you aren't there because no one else around you has any money either.

As I tried to make clear a while ago, humans are very particular about the kinds of inequality that bother them. Race tied inequalities are among the ones they care most about.

IME most people only think individual IQ differences are ok because they believe other qualities compensate the difference. If they say that some person has a higher IQ, they usually (at least implicitly) question their social skills, financial success, physical prowess, etc.. Also they always talk about much smarter people, not about the 50% under the average, conveying the idea that difference is due to the genius' unusually high IQ not because most people are stupid in comparison. OTOH group comparisons usually imply that one group is smarter and the other is dumber, by comparing the average values for each group. While race is a sensitive issue, if we exchange race by gender, economical status, birthplace, weight, etc., the controversy is pretty much equivalent.

About the IQ vs. GDP "controversy" both Lynn and Vanhanen should be ashamed. They're not even decent scientists, their methodology is flawed and they manipulated the data to fit their results! You can't say "I don't have the real data so I'll just put a number here and argue that it's true because I say so." and expect it to be taken at face value. It's not an experiment if it isn't reproducible (which rules out almost everything except biology, physics and chemistry ;) and you can't reproduce it if you force the data to fit your pattern.

Now, speaking about IQ itself, does make sense talking about it? Is there (at least) a significant correlation between IQ and any useful metric? Can we say that IQ improves our utility, for example? Are we (as a scientific community) sure that IQ measurement isn't just self fulfilling (i.e. it measures what high IQ people have, but not much more)? I know of the (methodologically valid) studies that show people with higher IQ earning more but those studies don't show if these cases are a direct result of IQ (i.e. they're more effective) or a indirect result due to employers favoring people with high IQs (or SATs). Also other (methodologically valid) studies show that IQ doesn't correlate to financial growth (i.e. becoming richer) because people's investment and saving habits don't correlate with IQ.

IMO IQ is a poor metric, it can't give reliable predictions about things that really matter (e.g. GDP, personal finance, scientific achievements, etc.). I fail to see how it's better than trying to measure how fast can people divide long numbers, surely it may be impressive and have a couple of use cases, but mostly it doesn't matter. IMNSHO it's telling that those people trying to correlate IQ with other values always use bad methodology and end up trying to convince the reader that correlation (i.e. their results) equals causation (i.e. their hypothesis).

About the commenting program:

1. Why require both javascript and a captcha to prevent spamming? Both are very bad for accessibility.
2. Moving to another URL is quite bizarre. Additional negative points if after submitting doesn't show any results at all.
3. Combining 1 & 2 in a javascript requirement for two (seemingly) unrelated URLs makes the whole process much more complicated than it needs to be.

I browse with javascript disabled (security reasons) and usually can post in most blogs. I also write software for a living, so I know that any of those aren't required. Please consider improving your blog software to something simpler and less restrictive.

There are no racial inequalities there are only human inequalities.

Sharper, dividing humanity into the subpopulations "male" and "female" does not result in average IQ differences, though brain-volume does differ, as does the visuo-spatial vs verbal component of intelligence, and the standard deviations for them differ as well. It is usually an assumption of statistics that large enough groups with arbitrary divisions (say, by a "natural experiment") will not differ on average.

In which geographic region were you thinking of where darker skin indicates higher intelligence?

Daniel Yokozimo, nobody had been discussing Lynn yet, so you should have mentioned that that the book you are referring to is called "IQ and the Wealth of Nations" so other readers aren't left in the dark. I believe if you remove all the data they made up/estimated the result is pretty much the same, although I haven't actually read the book myself. If you know of a study finding no correlation between IQ and GDP or GDP growth, provide a link. The reason people are interested in IQ it because it has more predictive power than any other variable in psychology and possibly any social science. It isn't just employer preference, it also predicts likelihood to be in an accident on the job (within professions).

Group differences are distressing because the more of a difference there is, the more grounds there are for discriminating on the basis of race or sex, and it is incredibly frustrating to find oneself continually judged not for one's own manifested abilities, but for the average abilities manifested by other people who who share one's race or sex.

Yes, if there is an average group difference, then race or sex alone would count as Bayesian evidence of a particular level of ability. A perfect Bayesian, looking at all the evidence, would be justified in statistically discriminating. But there aren't any perfect Bayesians; that's why this blog exists. One needn't contend that all groups have identical average capacities in order to be worried about discrimination. The problem is not discrimination itself, considered abstractly in some toy domain; the problem with discrimination is that in the real world, people are going to do it wrong, and do it wrong in all sorts of harmful and oppressive ways.

In "An Intuitive Explanation of Bayesian Reasoning," Eliezer writes that people don't pay enough attention to priors, but I wonder if the opposite isn't the case when the task is evaluating people whose race and sex are known.

Perhaps I misunderstand the purpose of this web site, but why not just stop categorizing people and look at each individual as they are in the present with love and respect?
Eliezer-excuse me if this is inappropriate, but have you read the book of Job? I don't mean as the word of God that must be believed at all costs, but as an insightful look at the human condition. (I'm not religious, but that doesn't mean I have to discount the wisedom that does show up in the Bible or other religious texts, does it?)

Who has suggested not looking at each individual with love and respect?

Could you be more specific about the insights you find in the book of Job that are relevant here? (It doesn't seem to me to say anything about individuals versus groups, or about the implications of varying intelligence or other unequally distributed benefits; and what it says on the question of whether God is just amounts to "how dare you question his justice? he's bigger than you". But no doubt I'm missing something.)

U can't possibly say what I want about this and keep it short, but I"ll try anyway. I'll sketch the bare outlines and if you want you can fill in the details.

The US race problem comes entirely from insufficient miscegenation. If we mixed things up sufficiently we'd have no separate races in 2 generations. But we haven't, yet.

Consider Gause's Law. Populations that don't mix are like separate species, they won't stably fill the same niches. The alternative to one population going away is a caste system -- guarantee that each race has niches that are theirs alone, that they have a guaranteed place in the society.

The developing US caste strategy was abandoned for one that officially lets anyone compete for anything. We *have to* claim that the competition is fair, or equality is a sham.

In reality, africans have far more genetic diversity than anyone else. Without detailed information it would seem to follow that they'd have more variability in intelligence, and in anything else that hasn't been intensely selected. And also in traits that have been selected; a diverse population will tend to find multiple solutions to challenges.

Within a single breeding population, inequalities tend to average out. If you cheat me in a land deal, maybe my grandson marries your granddaughter and it's all in the family. In 5 generations there's more chance for that, in 10 generations it's likely. If one of my ancestors cheated another of them, he shouldn't have done that but it doesn't mean much to me. Between separate populations the outrage tends to build up. If you cheat me and get the advantages at compound interest, my great-great-grandchildren just have it worse. It's predictable we'll have problems when separate populations interact too much.

About gods, if you suppose that you will have an infinite life for which this short life is only some sort of audition, then it isn't at all important about injustices you endure here.

Also, who died and left you deciding what's good and what's evil? A god might have some higher purpose that makes your petty concerns irrelevant. All of nature is arranged in feedback loops -- when one population gets too large it degrades its environment and then the population size gets adjusted downward. "Too large" can be defined as "large enough to degrade its environment to support fewer". This goes on everywhere. When a human baby dies in a degreaded environment, isn't it the same thing? Should we be exempt from the feedback loops that run the rest of the world? Of course we'd think so, but why would anybody else agree?

TGGP: Eliezer referenced the book (the wikipedia url on the "real" link, lookup for the phrase "Is Idang Alibi about to take a position on the real heart of the uproar?"). I thought everybody followed the links before commenting ;). Anyway I assume that if something is referenced its discussion is on topic.

Regarding their data, we can't just remove the data they fudged, we need to redo the analysis with the original data. We can't just discard data because it doesn't fit our conclusions. Using their raw data without fudging we are left with low correlation, many data points outside the curve.

Ditto for any other studies. I highly skeptical of sociologists or psychologist papers because they always (again IME) have use very bad statistics. Most assume a gaussian or poisson distribution without even proving that the process generating the data has the right properties. The measurement process is highly subjective and there's no analysis to assess the deviance of individual measures, so they don't properly find the actual stddev of their data. If one wants to aggregate studies, first one must prove that the measurement process for each study is the same (in the studies mentioned in your "predictive power" link this is false: at least two Lynn studies use population samples with different properties, also another couple use different IQ tests) otherwise we are mixing unrelated hypothesis.

I'm highly skeptical of IQ measurement, because it's too subjective. Measuring the same individual over and over on a long interval we get different results, but we shouldn't. A physicist wouldn't use a mass measurement process that depended on subjective factors (e.g. if the measured object is pretty or the time of measurement isn't jinxed), in a similar way we shouldn't use a measure of mental capacity that is highly dependent of stress (which has no objective measurement process) or emotional state. In this situation one of the best approaches would be using many different data measurements for each individual and aggregate the data with Monte Carlo analysis to find the probability of each results. We can't just fudge the data, discard sample we don't like and use a subjective methodology, otherwise it isn't science. When a physicist does a experiment he has a theory in mind, so he either already has an equation or ends up discovering one. The equation must account for all variables and the theory must prove why the other variables (e.g. speed of wind in Peking) doesn't matter. "IQ and the Wealth of Nations" fails to prove that any other factors influencing GDP are irrelevant to the IQ correlation, that alone discredits the results.

Correlation is the most overused statistical tool. It is useful to show patterns but unless you have a theory to explain the results and make actual predictions it's irrelevant as much as the scientific method is concerned. If we ignore this anything can be "proven".

Skin color is a function of the latitude at which ones ancestors lived.

This is obvious through observation:
ancestors lived on the equator - brown skin.
ancestors lived at extreme north/south latitude - white skin.

I don't believe I specified a quantitive amount by which they would differ, just that they would differ. You're right, normally (pun intended), the groups wouldn't differ by much. That's part of the point, isn't it. Why care that they differ at all? There isn't a useful reason to care that group X has a different average IQ than group Y. Does a particular group of dark skin people have a lower intelligence due to their skin color? Not likely. Other factors are much more significant. There's no causual relationship between the two factors. The only reason it's mentioned at all is because some people have a hang-up about defining themselves or others by the color of their skin.

If you took two groups. One group with high intelligence and a second group with low intelligence and genetically modified them to change only their skin color, would their intelligence change drastically as well? The point is that skin color is not a causal factor in determing intelligence, so it's meaningless to use as a filter when presented with a sample size of one random individual with skin color X.

For a bunch of reasons, recent black immigrants from the East Indies in NY have a much higher average intelligence than blacks in NY with a longer US family pedigree (and a higher average intelligence than your average white in NY, for that matter) Does it matter to their intelligence what color their skin is? Apparently other factors matter a lot more. My point is that racial grouping is more arbitrary than most other groupings as a basis for making judgements and is only really used by people out of historical inertia. The groupings of people into "race", whether done by individuals or by governments tends to be pretty damn unscientific compared to say, geneticly-based groupings. I don't know first-hand about the UK's classification schemes, but have you seen the racial categories governments in the US use?


Dividing the human population into two groups, male and female, does in fact result in one group having a higher average intelligence than another. Depending on your actually used definition of intelligence, you may decide that one group has a higher average than the other and which group that is may change over time. I suppose if you modify your definition of intelligence based on the actual level of that quality in males and females you might be able to come up with a definition that for an instant in time made the two groups equal, but the actual population changes quickly enough that your definition would also quickly become outdated and the two groups would no longer be equal in your proposed definition of intelligence.

On the question of if there is a usefully measurably large difference between the groups (as opposed to an actual difference), then I agree that it's not useful to use male/female group membership as a filter or test for an individuals estimated intelligence, regardless of which of the popular definitions of intelligence that you subscribe to. The variance is way too high within the population to make it a useful indicator for practical purposes.

Again, that's why the fact of there being differences should be as unremarkable as it is useless as a guide to decisions and policy.

Take a step back. Theoretically if someone divided the world population into two groups by randomly assigning each individual the letter "A" or the letter "Z", you would have two groups that on average have a miniscule difference in intelligence, or athletic ability, or whatever you want to measure.

But practically speaking, knowing the results of the "A" and "Z" groups and also knowing which group your individual belongs to is useless as compared to a factor that actually has a casual relationship to what you are trying to measure that individual for.

No offense to the sociologists among us, but individual characteristics matter far more than group characteristics when making any sort of judgement or decision.


Sorry, forgot to answer your additional question in my reply. Whenever Thomas Sowell and Walter Williams get together in a small group (as they do occasionally), within that small geographical region darker skin color is a great indicator of higher intelligence. Makes all these divisions seem just a little more arbitrary, doesn't it?

On a more serious note, Queens, NYC, NY has a higher than normal proportion of black West Indies immigrants. In that geographic region, blacks have an average IQ higher than those with lighter skin. Other US locations include Pembroke Pines, Florida, Rialto, California and Brockton, Massachusetts. They also have a higher median income, are more likely to live in a two parent household, own their own home, etc...

g- Nobody has suggested treating people other than with respect and love. It seems to be a fairly common thread in the things I'm reading here.
Instead of asking "what group has a lower or higher IQ?", why not ask, "How do we raise an indiviual's IQ?"
I may be misreading Job, I see more like- "don't forget the beauty that surrounds you"

Sharper -- Pembroke Pines? Holy schmoley, that's a pretty small neighborhood. I'ts more known in my mind for Canadian snowbirds than high IQ dark skinned people. Do you have a reference for your assertion? I'm personally intererested because it's about 5 miles down the road from where my sailboat is right now. Plus it's where a friend of mine lived when she was in the escort business.

Douglas, I don't see why we can't ask both questions, but in any case the question this post was about wasn't "what group has a higher or lower IQ?" but "why do people think that group IQ differences matter more than individual ones?" -- decrying group-over-individual emphasis just as much as you are.

All the stuff about natural beauty in Job is there to make the point "God is bigger and cleverer than you are, so who are you to question him?". (Hence the constant refrain of "Do you know ...?, Have you seen ...?, Were you there when ...?".) It's admittedly rather grand, at least once you get over what now reads like bizarre Bad Science (storehouses for the hail, etc.), and people whose judgement I respect have claimed it's great poetry, but I still don't quite see what insight into the human condition it offers beyond "Sometimes bad things happen for no readily apparent reason", which most people over the age of three have noticed even before they read Job.

"in his humble opinion": Jim Watson? Humble? There's a first for everything.

>Race adds extra controversy to anything; in that sense, it's obvious what difference skin colour makes politically. However, just because this attitude is common, should not cause us to overlook its insanity.

You can thank the Nazis for making race so political that it won't be touchable for generations. You can also thank the segregationists for irritating that political gland and making race untouchable for more time.

Francis Galton brought a lot of ideas to the world, but the one that was amplified to the point where it will take centuries for the controversy to die down was "eugenics." And when people hear the word "eugenics" I bet you that they also hear "master race." I read a lot of science fiction, and I believe that there have been very few novels that discuss eugenics as anything other than a project to make "supermen." Tepper's The Gate to Women's Country and Sawyer's Hominid trilogy are the only ones that come to mind at this time.

Weasel words: I use "thank" purely out of sarcasm.

Why isn't the concept of fairness a bias?
It seems to meet the criteria.

People care about race because race is about who your blood relatives are, and who, to some extent, your descendants will be.

A racial group can best be defined as an extended family that has more coherence and cohesiveness than a typical extended family because it is partly-inbred.

So, that's why people care so much.

Also, please, can we stop equating "race" and "skin color"? Haven't you ever seen an African albino? Being white in skin color doesn't make him white racially.

Or consider the famous golfer Vijay Singh, who of South Asian origin and was born on Fiji. He is darker than the average African-American (but has Caucasian features). He is never, ever considered to be racially black or African-American in America. Never. You can make up a list of other dark-skinned people who aren't considered black in America, such as pundits Dinesh D'Souza and Ramesh Ponnuru.

Race is about ancestry.

Different societies have different ways to deal with the inevitable complexities of genealogy in assigning people to races, but they are genealogy-based, not skin color based.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30