Tuesday, April 19, 2011

Race, Class, Intelligence and Genes - bit more

Following on from previous post I just had an amusing thought.

Assuming anonymous's premises

(1) intelligent people are more likely to end up in high paid jobs, and
(ii) intelligence is partly inherited (genetic)

He (I assume a "he" - don't know why) concludes that the children of the higher paid are more likely to be intelligent, on average. That's true, other things being equal.

However, anonymous then goes on to claim this shows kids of working class people are less intelligent on average. That does not follow. The conclusion certainly isn't "unavoidable", as anon claimed.

If 2 groups A and B (e.g. black and white, working class and middle class, gays and straights, whatever) make a up the population, but group A members are largely or wholly prevented from competing effectively for intelligent, higher paid jobs because of non-genetic factors, then yes, those in such jobs will be more intelligent on average than the members of either group A or B on average.

However there's clearly no reason as yet to suppose that, as the children of group A don't show up much in those jobs, the children of group B are, on average, more intelligent than those of group A.

Moreover, and this is the amusing bit, if there's no independent, prior reason for supposing the members of group B are genetically more intelligent than group A on average, we can also conclude those in group B who are not in intelligent jobs will then be LESS intelligent on average than those in group A.

So, assuming no prior reason to suppose working class people are innately less intelligent, anonymous's premises should lead us to suppose that, other things being equal, those middle class people not in intelligent jobs will be less intelligent on average than working class people. This is because they have actually been weeded out on the basis of lower intelligence, whereas the members of group A - the working classes - largely have not.

So, knowing what we do about the way non-genetic factors prevent both black and working class people from competing effectively for the intelligent jobs, we should conclude that the children of middle class people/white people not in intelligent higher paid jobs will be LESS intelligent than those of working class people/black people, on average.

This would account for Tim, nice but dim, syndrome.

Yet Woodhead claims it's primarily lack of innate intelligence which explains why many kids are uninterested in his lessons, disruptive, etc. Actually, if it's lack of innate intelligence which supposedly produces such problem behaviour, it should actually be the kids of the middle-classes-not-in-intelligent-jobs whom Woodhead should be singling out.

Yet he just ignores the obvious non-genetic factors preventing working class kids from competing effectively for intelligent jobs, and jumps to the conclusion that the children of the working classes are less intelligent, on average.

That does say a lot about him, I'd suggest.

12 comments:

Anonymous said...

"However, anonymous then goes on to claim this shows kids of working class people are less intelligent on average. That does not follow. The conclusion certainly isn't "unavoidable", as anon claimed."

It is if the two premises are correct. But, as Wombat said in comment to your previous post, you seem to disagree with premise (1), which would require intelligent people from Group A to stand a chance of getting the well-paid jobs. So your problem seems to be a factual one with one of the premises, not with the article itself.

"Yet Woodhead claims it's primarily lack of innate intelligence which explains why many kids are uninterested in his lessons, disruptive, etc. Actually, if it's lack of innate intelligence which supposedly produces such problem behaviour, it should actually be the kids of the middle-classes-not-in-intelligent-jobs whom Woodhead should be singling out."

His proposals to improve the quality of education (more vocational training, more academic streaming) would, as far as I can see, apply to everyone, including dim middle-class people.

Hugo said...

Let's skip these predictions and go straight to the evidence:

http://super-economy.blogspot.com/2011/04/iq-income-and-wealth.html

Nick said...

Not intended to be entirely rigorous, but how about the following for a brief sketch of an argument:

P1: Middle class adults are, to a significant degree, defined as those who have completed tertiary education and have a professional, technical, managerial, or academic occupation (as opposed to a skilled or semi-skilled manual occupation, for example) [demographic ABC1]

P2: Adults who have completed tertiary education and have a professional, technical, managerial, or academic occupation probably have an average IQ that is greater than the average for the population of adults as a whole [according to the available empirical data]

C1: The average IQ of middle class adults is probably greater than the average for the population of adults as a whole [from P1 and P2]

P3: The variability in IQ in adults is primarily due to genetic factors (around 80%) [as per scientific studies]

P4: Variability in children’s IQ is partly heritable (probably between 30% and 70%) [as per scientific studies]

C2: To the extent that variability in children’s IQ is hereditable, the children of middle class parents will probably tend to inherit genes that predispose them to having a higher IQ than the average for the population of children as a whole [from C1, P3, and P4]

P5: The variability in children’s IQ that is not hereditable is due to environmental factors (family environment, peer-group etc. - probably between 30% and 70%) [as per scientific studies]

P6: Adults who have completed tertiary education and have a professional, technical, managerial, or academic occupation probably tend to provide an environment for their children that is on average more conducive to boosting their IQ (or to realising their genetic IQ potential) than is the average for the population of adults as a whole [a plausible, albeit speculative inference based upon limited data]

C3: To the extent that variability in children’s IQ is due to environmental factors, the children of middle class parents will probably on average tend to have their IQs boosted (or their genetic IQ potential realised) more than the children of the population of adults as a whole [from P5 and P6]

P7: If the children of middle class parents have, on average, genes that predispose them to having a higher IQ than the population of children as a whole, and an environment that is more conducive to boosting their IQ (or to realising their genetic IQ potential), then they will probably tend to have a higher IQ than the average for the population of children as a whole

C4: The children of middle class parents will probably tend to have a higher IQ than the average for the population of children as a whole [from P7, C2, and C3]

Corollary:

P8: The overwhelming majority of adults who are not middle class are working class, and they each comprise very approximately 50% of the population (the upper class being a very small minority – around 2% of the population)

C5: The children of working class parents will probably tend to have a lower IQ than the average for the population of children as a whole [basic probability theory]

More to follow..

wombat said...

FWIW Woodhead for all his other failings does mention in the original Guardian interview that "the nurture is better" as well, so he is at least open to suggestions that other factors are at play.

I think this new example (maybe) shows the difference between your model and anon's.

Anon seems to be pointing to the significance of the fact that unlike skin colour it is easier to change class. Indeed in the model with no interbreeding a colour change can be taken as impossible (after all thats presumably one of the reasons why it stipulates no interbreeding) but a class change is on the other hand quite possible even if there is powerful discrimination against intelligent working class people taking up well paid jobs there appears to be nothing stopping a dim child of a middle class family taking up a poorly paid job. How many generations of this can the middle class aura survive. Well, Jocasta's Daddy left her some money in the previous post but what happens when that runs out? Won't later generations eventually become working class? On the other hand n-th generation white folk of a line selected for lack of intelligence will never be black will they.

Richard T said...

To me, odious as he surely is, Woodhead is in the same company as Paul Johnson and Charles Lord Snooty Moore. He combines a monumental arrogance with irretrievable silliness.

John said...

Stephen,

Obviously true. But have we actually established that, to the degree that working class kids fail to gain high paying jobs, non-intelligence factors predominate?

This whole argument could be resolved with some empirical evidence, but I still think both you and anonymous need to define 'middle class' and 'working class', since your arguments seem to diverge on this unstated premise.

Stephen Law said...

Richard T.

Anon agrees that class is not defined wholly by income. In the same way that the category black-but-not-in-a-high-paying-job is not defined entirely by income.

Hugo - your stats are interesting, but irrelevant here as I made no predictions. Also, in any case, they say nothing about genetic components.

As to premise (i), it is ambiguous I think. "intelligent people are more likely to end up in high paid jobs". Suppose black people or working class people are entirely prevented, by other means, from competing for high paid jobs. Is (i) true? If the answer is "no", then anon's argument just begs that question. But I was reading (i) in such a way that it is true even if such an obstacle exists. Being intelligent if you like multiplies the probability of your getting such a job. But if what it is multiplying is zero, then you are still no more likely to get one.

Stephen Law said...

Here's an argument for you to consider.

(i) ruthless, self-serving and shallow (obsessed only with in money) people are more likely to end up in high paying jobs.
(ii) such character traits are partly inherited.

Therefore children of people in high paying jobs are more likely to be ruthless, self-serving and shallow.

Therefore (because middle classness is defined partly by income) the children of middle class people are more likely to be ruthless, self-serving and shallow.

Is this conclusion "unavoidable", to quote anonymous?

I am very skeptical about it. The conclusion might be true. But it has hardly been established beyond reasonable doubt. There are just too many other factors in play here, including most obviously the fact that lack of ruthlessness, self-servingness and shallowness may not be what primarily prevents working class people from getting such jobs.

Someone who thought that conclusion did follow "unavoidably" would be either a fool or someone with a prior "agenda" re middle class people that was skewing their judgement.

Anonymous said...

"Anon agrees that class is not defined wholly by income."

It's not defined wholly by income in that there are some cases in which a high-paying job wouldn't be considered middle-class (a professional footballer, for example). However, I think it fair to say that the great majority of "middle-class" jobs pay better than the great majority of "working-class" jobs, to a degree that, for the purposes of this argument, "middle-class" and "on a higher-paying job" can probably be treated as synonymous.

As for your A vs. B example, you seem to imply that it's impossible for someone to move between the two groups -- if you're born an A, you stay an A, no matter what job you get. In some cases (e.g., colour), that's certainly true. However, it's not the case in the class issue: it's entirely possible (if difficult) for a working-class kid to get a job as a doctor or lawyer or whatever, and hence become middle-class.

If it's possible to move between groups A and B, then we would expect to find group B to be the more "intelligent" group, even if the field is levelled against group A. Say for argument's sake that prejudice and lack of proper education mean that, whereas 75% of people in group B get high-paying jobs, just 25% in group A do. We would still expect the 25% of group A to be drawn disproportionately from the most intelligent members of that group, and the 25% of group B who fail to get high-paying jobs to be disproportionately drawn from the least intelligent members of that group. IOW, the most intelligent As would become Bs, whereas the least intelligent Bs would become As. Even if the two groups had exactly the same average intelligence to begin with, we'd expect a difference to arise over time.

Adam said...

At the risk of going off point slightly, I was wondering if this argument would be valid.

If we were saying that the set of middle-class children, on average, are intelligent. So if you are a member of the set Middle class you are then, on average, intelligent.

so formally:

'p ⇒ q'

p

therefore: q

Where p = middle class children, and q = intelligence.

However, to then say that if you are intelligent you are middle class would commit the fallacy of confirming the consequence. The argument is uni-conditional and therefore it does not follow the other way, that intelligent people are middle-class.

So the fallacy would be:

'p ⇒ q'

q

therefore: p

This, not in an attempt to show that there are some middle-class people that are less than 'intelligent' - which it doesn't - shows that you don't have to come from the set of middle-class to have the property of 'intelligence'. I know this does not tackle the original argument invoking high-paid jobs, but it is worth noting...I think.

Stephen Law said...

Anon

"it's entirely possible (if difficult) for a working-class kid to get a job as a doctor or lawyer or whatever, and hence become middle-class."

This is the only bit of what you said that has any relevance so I will focus on it.

Yes, of course, if there is some social mobility, even if not much, then there's some reason to expect the kids of those in group B to be genetically smarter than those in group A.

But then similarly there's some reason to suppose the kids of those in group B will be genetically more ruthless shallow and self-serving.

In each case, "some reason to suppose that" doesn't mean "unavoidably the case that". Because the situation is complex, I certainly would not bet on those in group B being genetically more ruthless, shallow and self-serving. I'd be skeptical about drawing that conclusion on the basis of such evidence. Ditto genetically more intelligent. There may be some difference in intelligence, as I agreed at the outset. But it has not been established. And it certainly has not been shown to be at all significant - e.g. significant enough for Woodhead to justified in pointing a finger at working class kids and say they're the problem because they are genetically thick."

So why do I get it in the neck when I express skepticism about you having established that working class kids are innately less intelligent - that this conclusion is "unavoidable"?

Your insistence on jumping to that conclusion says much about you, I think!

PS what do you think of my suggestion that middle class kids not in such high paying professions may actually be less intelligent on average than working class kids? After all, the working class kids are largely weeded out of that job market by other factors. Not so the parents of those middle class kids.

Of course, I wouldn't jump to the conclusion that such middle class kids will be genetically thicker on average than working class kids. That's because I am not a twat with a chip on my shoulder.

Anonymous said...

http://www.wspublishers.com/uhh.pdf
Michael H Hart - Understanding Human History

Read the introduction. Hopefully it'll pique you interest enough for you to read the rest of the book. The first hundred pages or so are a good introduction to the state of the literature about race and IQ.