Thursday, May 9, 2013

Migrant selection, IQ, and Richwine

I am starting to dip my toes into the Richwine thing. I haven't read his dissertation or the Heritage paper yet. I really despise the big picture: closing the door to the United States because we find immigrants undesirable in some way, etc. Talking about the IQ differentials of different groups just makes me queasy too.

But I'm also concerned about how some people are going into this and making their arguments against Richwine.

Intelligence is a real thing, OK? It does vary across the population. I defer to the psychologists on what to think of any particular metric (paging my own favorite psychologist on this one - commenter Dr. J). But it's a real thing. It's not the only thing we ought to care about, of course, which is why I don't like making policy on the basis of a few qualities we find to be desirable. But it's a real thing.

One thing the IQ of an immigrant population relative to the native population is going to depend on is selection effects: is there negative or positive migrant selection? This is dictated by a lot of things including geopolitics and the specific immigration statutes - but it's also dictated by the comparative advantages of potential migrants.

Let's say Latinos in Central and South America have precisely the same IQ distribution as native Americans. Exactly the same. If there is negative migrant selection, Latino immigrants to the United States will have a lower IQ than natives. There's nothing racist or controversial about it - that's math. The U.S. average IQ will go down and the Central and South American average IQ will go up (because relatively lower IQ residents will migrate to the United States).

If IQ is highly determined by genetics (I defer to the geneticists on this) and marriage is highly determined by ethnic and immigration status (I defer to the sociologists on this), the differences will persist.

If IQ is highly determined by child nutrition (I defer to the child development people on this) and Latino immigrants remain relatively low income, the differences will persist.

This is just a fact of life people.

Now what bothers me about Richwine is I can't imagine what motivates a person to get into these questions. Maybe Latino immigrants are lower IQ. That is entirely plausible to me.

But I really don't care about that and that doesn't really register on my list of things that ought to guide immigration policy.

I guess some people do, but I am suspicious of people who have a deep interest in comparing the IQs of different populations.

7 comments:

  1. I know that intelligence is a real thing, but can I measure it or even describe it in a positive sense? I've run into this problem a lot when dealing with intelligence, I cannot describe it but in terms of what society values or what I value. It's somewhat paradoxical for me in that way.

    ReplyDelete
    Replies
    1. Ya I defer on that one. My understanding is that we do have pretty good metrics actually, but that the problem is it is multifaceted.

      And as you point out (and I tried to express in the post), even if we can pinpoint it it still leaves open the question of how to value that quality.

      Delete
    2. I wouldn't be so sure of this. Metrics aside — just speaking conceptually — "intelligence" is pretty clearly a mess: it's supposed to be a kind of generalized aptitude stripped of any particular application, as if intelligence tests somehow stand outside of human culture. Without resorting to a kind of Romantic idealism, it's hard to see how or why the intellectual aptitudes or faculties should be valued more highly than the ability to hit a baseball, empathize with others, knit a scarf or any of the other infinity of human abilities that we value. All of these activities are equally embodied.

      Allow me to direct you to the wikipedia page on Gilbert Ryle's The Concept of Mind, which is the foundational text of this line of inquiry in the Anglo-American tradition. It might be of interest.

      Delete
    3. That's basically always going to happen when you try to study something which lay people talk about. At the end of the day, if you're going to do science, you need to operationalize your variables. Inevitably, your operationalized variable will not perfectly match the vague ideas that the lay have in mind.

      In this case, g-factor tests appear to be pretty-well correlated with each other and with many of the characteristics that are frequently cited as markers of intelligence such as mathematical and musical ability.

      Delete
  2. Not so fast. I mean, possible, but easy to test El Salvador population is about 6.6. There are easily about 1.5 immigrants from El Salvador living in the US. So this is a decline in the population of about 18.5. Do you seriously believe that El Salvador IQ's today is noticeably higher than in other countries? I mean, seriously? Besides, Richwine dissertation strongly suggests it's about the origin--"European" IQ is (and presumably according to Richwine) has always been higher.

    ReplyDelete
  3. Regarding IQ, infectious desease is a large contributor.

    "We tested all these ideas. In our 2010 study, we not only found a very strong relationship between levels of infectious disease and IQ, but controlling for the effects of education, national wealth, temperature, and distance from sub-Saharan Africa, infectious disease emerged as the best predictor of the bunch. A recent study by Christopher Hassall and Thomas Sherratt repeated our analysis using more sophisticated statistical methods, and concluded that infectious disease may be the only really important predictor of average national IQ."

    ReplyDelete
  4. Mr. Kuehn, you are totally out of your depth here.

    To add to the comments already made, I would cite the "Flynn effect," described in a 2007 article in the LA Times: 'James R. Flynn, an emeritus professor of political science at the University of Otaga in New Zealand, discovered two decades ago that IQ test scores were steadily rising in the developed world despite failing schools and stagnant standardized test scores -- a phenomenon called the "Flynn effect."'
    http://articles.latimes.com/2007/oct/27/science/sci-iq27

    The "Flynn effect" and ten other findings about "IQ scores" are summarized in an article in Psychology Today. These findings suggest it is highly unlikely your a priori logical analysis and hypotheses about test scores include any productive lines of inquiry.
    http://www.psychologytoday.com/articles/200107/iq-the-test

    ReplyDelete

All anonymous comments will be deleted. Consistent pseudonyms are fine.