The Unz Review • An Alternative Media Selection
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 TeasersGene Expression Blog
Leaving Science on the Table
Email This Page to Someone

 Remember My Information



=>

Bookmark Toggle AllToCAdd to LibraryRemove from Library • BShow CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
AgreeDisagreeThanksLOLTroll
These buttons register your public Agreement, Disagreement, Thanks, LOL, or Troll with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used three times during any eight hour period.
Ignore Commenter Follow Commenter
Search Text Case Sensitive  Exact Words  Include Comments
List of Bookmarks
Citation: Vinkhuyzen, Anna AE, et al. "Estimation and Partitioning of Heritability in Human Populations Using Whole Genome Analysis Methods." Annual review of genetics 47.1 (2013).

Citation: Vinkhuyzen, Anna AE, et al. “Estimation and Partitioning of Heritability in Human Populations Using Whole Genome Analysis Methods.” Annual review of genetics 47.1 (2013).

The above are some commonly accepted values for the heritabilities of complex traits in the scientific literature. By heritability I mean to refer to the proportion of the variation in the trait within the population which can be explained by variation of genes within the population. The reason I am very precise is that heritability should not be taken to be some sort of obvious correlation statistic. Even though height is highly heritable in modern societies, the average sibling difference is 1.8 inches (4.6 cm). But, heritability is informative when it comes to populations and patterns of variation that we see around us. A highly heritable trait is amenable to selection for a breeder, while a non-heritable trait is not. But just because a trait is highly heritable does not mean that environment does not matter. One can imagine a scenario of “all boats rising” where environment shifts the trait value equally across the population, while all the variation within the population is still due to genes.

140122_XX_fig1.jpg.CROP.promovar-mediumlargeI’m bringing all this up because W. Bradford Wilcox has a new piece in Slate, What’s the most important factor blocking social mobility? Single parents, suggests a new study. The variable what Wilcox is alluding to in the title is illustrated in the figure to the left: the proportion of single parents in the community is very predictive. Importantly he’s arguing for very powerful community level effects. And certainly the correlation is pretty impressive. And knowing the Left’s love affair with the idea of science, he flogs this correlation rather for all that it’s worth:

Throughout his presidency, Barack Obama has stressed his commitment to data-driven decision-making, not ideology. Similarly, progressives like Krugman have stressed their scientific bona fides, as against the “anti-science” right. If progressives like the president and the Nobel laureate are serious about reviving the fortunes of the American Dream in the 21st century in light of the data, this new study suggests they will need to take pages from both left and right playbooks on matters ranging from zoning to education reform. More fundamentally, these new data indicate that any effort to revive opportunity in America must run through two arenas where government has only limited power—civil society and the American family.

The author is a visiting scholar at the Right-leaning American Enterprise Institute, but he does a good job of being evenhanded and not overtly ideological from what I have read. W. Bradford Wilcox seems to be sincerely driven by a commitment to the social issues he writes about so often. He’s also intellectually honest enough to admit that the lead author of the study which reports this stark correlation “has been careful to stress that this research cannot prove causation.”

At this point you know where I’m going with this. Wilcox admits the complexity and confounds at the heart of phenomenon he’s trying to describe. Naturally one aspect he leaves out are the innate dispositions of individuals due to their heritable makeup. More concretely, personalities differ, and those differences have consequences, and those differences partly have a genetic component. I haven’t thought about all the policy implications of this, but I do know that it makes the story that Wilcox and company tell more complicated, and likely alters the nature of the solutions that they might posit (and definitely the effect sizes they might see). Conservatives often accuse Left-liberals of being “social engineers” who neglect the complex interdependencies of organically evolved cultures. The argument is that they presume that they can model cultural outcomes as if they were as predictable as thermodynamics. But modern American conservatives have fallen into a similar trap, with the mantra “the family” “the family” “the family” as a catchall solution to all social ills. As if society is a simple pliable physical process, and the family one of its easy-to-modulate regulatory components.

It is somewhat tedious to get on this high horse over and over, but it needs to be done. Modern Left-liberals certainly won’t do it. While conservatives harp on the family, liberals focus on the economy. Ultimately both parties are missing a part of the picture, but neither is going to challenge the other on their shared lacunae.

 
• Category: Science • Tags: Behavior Genetics 
Hide 21 CommentsLeave a Comment
Commenters to Ignore...to FollowEndorsed Only
Trim Comments?
  1. The sociologist’s fallacy strikes again.

  2. Hear, hear!

    You’ve saved me from writing a critique of that Slate piece, since it’s not like time in abundance for me at the moment. 😉

    Readers should also check out this post by Staffan. The heritability of many traits appears to be even higher than those noted here (the true heritability of personality is in the .7 – .8 range, just like IQ, BMI, and height):

    The Nurture Enigma – How Does the Environment Influence Human Nature? | Staffan’s Personality Blog

    Indeed, the environment may matter a lot less then we think, perhaps not at all (the gross, large scale environment – the rising tide Razib alludes to notwithstanding). See my comment there:

    my comment on the true nature of the “environment”

  3. All of today’s socio-political beliefs involve unresolved complexities and imperfect data. People who wait for a perfect data and fully resolved complexities will never reach an actionable position on anything. Wilcoxx makes it clear he is aware of this. While, the idea of genetic predispositions is important and undersung, it’s not required in absolutely every discussion.

  4. Have any studies been done of single mothers who go the route of waiting until 40 to have kids, focus on their career, and then have IFV with donated sperm (or something along those lines?) Given the personality type of these single mothers is likely very different on average than the average (much higher in conscientiousness I’m sure), we should be able to dispel some of these concerns right off the bat if comparing them to a similar group of two-parent households shows no significant differences.

  5. mass, i don’t ask for new data. just integration of an old and robust model to explain some of the complexity. and when it comes to the things that wilcox talks about addressing confounds is essential. otherwise billions of dollars might be spent in vain. not just academic.

    karl, not to my knowledge. i should look into that.

  6. Upward mobility plot above (and in Wilcox’s study), Y25, is a very complicated number. Basically it is the median percentile, in the national income distribution, of a child whose family, at birth, stood at 25th percentile of income districution (from tax forms using specific sampling rules and interpolation). Basic defs: http://www.equality-of-opportunity.org/index.php/faq-s
    Details: http://www.equality-of-opportunity.org/files/mobility_trends.pdf

    So the data is narrowly specific to poor parents (not middle or upper class). And it tallies all children in a geographic area rather than specifically within the families consisting of single parents with children. The y-axis numbers, typically in the 35-50% range, actually show a pretty decent social mobility across the board – it looks like more than half of these born-poor kids achieve middle class income no matter the place they lived, and mores thereof.

    In the end, we have a supposedly nice correlation (but lacking confidence intervals, of course) but what’s even more impressive is the misinterpretation potential of the picture. The viewers are prompted to think that being born into single-parent families is a direct universal correlate of poor upward mobility, and to have appropriate knee-jerk reactions … even though the graph shows no such thing.

  7. Excellent piece, I’m just so tired of people addressing important social issues like single parents with their political ideological blinders on. Even though it was about thirty years ago I vividly remember the excitement of being at the hospital with my wife when she gave birth to our sons. I also remember all the women in the maternity ward who were absolutely on their own, no one came to visit them. It was very sad.

  8. Razib,

    You say,

    “It is somewhat tedious to get on this high horse over and over, but it needs to be done. Modern Left-liberals certainly won’t do it. While conservatives harp on the family, liberals focus on the economy. Ultimately both parties are missing a part of the picture, but neither is going to challenge the other on their shared lacunae.”

    But here’s the thing — while we have always had an economy that has gone up and down (indeed has failed to provide anything near the wealth today’s technology is able to provide) we have never had the breakdown in the family that we have today. So the reason social conservatives like me rail against single mothers all the time is because back when having bastards was shameful, there were fewer of them — a lot fewer. Bring back shame, bring back chastity and bring back a culture of marriage to the poor and you’ll have fewer poor people.

    Now, I know this is tough to do — I’m a true-blue HBD believer — but unlike JayMan I know we can’t keep doing what we’ve been doing (i.e. as Steve Sailer likes to say, “when you are in a hole, the first rule is stop digging”). So cut back on welfare, force private charity to pick up the slack (bring back the poor house!) and/or encourage marriage where you can:

    http://www.mdrc.org/publication/early-impacts-supporting-healthy-marriage-evaluation

  9. I’d say that by focusing on the family conservatives are not denying genetic deficits so much as they are assuming that even a family composed of persons with limited cognitive skills would fare better than those individuals would if left to their own devices. They think that a less than bright father and mother would do better than a single mother and child would do.

    I think this is true in a general way. That said, it is clearly true that families composed of limited, dysfunctional people can and do reenforce each other’s bad behavior. There are criminal families.

    A simple test of my proposition would be for any reader to ask a cop in his community if there is one family in which dysfunction and criminality runs intergenerationally. In any large community there is generally at least one. What to do in this case is another matter.

  10. Razib: Do you think that a stronger case can be made for the role of familial environment in differential social mobility and related outcomes by looking at *changes* in the latter (i.e., the difference) over time? For example, the crime wave of the 60s coincides with the sexual revolution and a decline in conservative sexual values; and if I’m not mistaken, not only overall crime but also differences in crime rates between different groups (notably, blacks and whites) increased. It seems less plausible that a change in genetics led to a change in differences between these groups, because a change in the genetics of the groups takes longer than a change in the social environment or values.

    Of course, this kind of argument doesn’t imply that genetics have no role in explaining current differences between, in this case, children raised by single mothers and children raised by two parents. But if having two parents provides or tends to lead to a familial environment that gives children even somewhat better life outcomes — granting that the better life outcomes we actually see are not *wholly* caused by these factors — then that still seems to me enough to ground the kinds of policy prescriptions people like Wilcox argue for.

    Karl: I agree that that would be an interesting study. I don’t think it would be decisive if you found that children raised by your group of single mothers had negligible differences from two-parent households, though. Conservatives don’t need to argue that traditional families cause better life outcomes among all groups. Many will argue that it’s among poorer people with lower impulse control, conscientiousness, etc. that traditional families are most important. In other words, the claim would be that two-parent families cause better outcomes primarily among the kinds of people that such a study would exclude.

    Such a result would still be some evidence against the conservative view, though, and I’d be interested to see one done.

  11. we have never had the breakdown in the family that we have today.

    who is this “we”? strikes me that this is an ahistorical model, though it captures some essence of specific changes. there have been multiple revolutions and recessions of ‘mores’ even in the west.

    bring back chastity and bring back a culture of marriage to the poor and you’ll have fewer poor people.

    i agree that a broad cultural framework matters. genes express themselves in an environmental context. but this is a big haul, and that’s why ppl tend to focus on atomic single unit aspects of the broader phenomenon. e.g., “get married.” that’s a marginal symptom of a broader problem. social conservatives do have a difficult time communicating the real issue in this sort of individualist culture, but you can’t tear down the master’s house with the master’s tools. saying that people getting married with reduce poverty is like saying making everyone go to college will make everyone smarter. it will have an effect. but it might be unpredictable and not as strong an effect in the direction you want.

    Do you think that a stronger case can be made for the role of familial environment in differential social mobility and related outcomes by looking at *changes* in the latter (i.e., the difference) over time?

    i generally thing this is a ‘non-shared environment’ factor. that, culture.

    in any case, my bigger point is that it isn’t too difficult to introduce a little behavior genetics/individual differences into the discussion. it will make it richer, and i think actually allow us to consider plausible ways to solve social problems. one-size-fits-all individualist monocausality isn’t going in the right direction, and is going to lead to hubris and disappointment.

  12. Razib,

    Thanks for the response — I thought this final comment of yours was interesting:

    “my bigger point is that it isn’t too difficult to introduce a little behavior genetics/individual differences into the discussion. it will make it richer, and i think actually allow us to consider plausible ways to solve social problems. one-size-fits-all individualist monocausality isn’t going in the right direction, and is going to lead to hubris and disappointment.”

    I do think it is important to introduce “a little behavior genetics/individual differences” into these discussions of social policy (in particular educational policy, as the blogger “Education Realist” likes to do). I’m just not sure welfare policy is a place where this additional information does help, unless the government wants to support some sort of soft eugenics policy via birth control promotion (i.e. “three generations of imbeciles are enough”!) As a Catholic, I can’t get behind such a policy, but it might appeal to a certain segment of Americans.

    I think Troy is much closer to the truth — while Mr. Meehan is certainly correct to suggest that a poor, stupid mother and father will probably produce poor, stupid children (don’t forget boys and girls — “Fat, drunk and stupid is no way to go through life, son.”), I suspect they will be much better off on all sorts of measures (school attendance, criminal records, earnings potential, etc.) with a mother and father at home — that’s what all those studies Wilcox and gang have been doing suggest. There is something very important to children about growing up with their biological mothers and fathers (or adopted mothers and fathers if necessary).

    You are still one of the best — keep up the good work.

  13. I suspect they will be much better off on all sorts of measures (school attendance, criminal records, earnings potential, etc.)

    where i would probably be skeptical is “much.” or would be curious how you define “much.” as someone who is intent on raising their children in a married two-parent household i think there are definite benefits for the children and the parents in this arrangement. but, i do think it is somewhat oversold in our culture as a utilitarian good (some will make the normative case, which i think is a different issue).

    . There is something very important to children about growing up with their biological mothers and fathers (or adopted mothers and fathers if necessary).

    well, i’d be OK with that. but what is that “very important” aspect? for example, shared family environment is often a much bigger effect for children and adolescents and tends to diminish. so stable two-parent households almost certainly on average allow for greater happiness and amity (though obvious there are many individual exceptions in homes with strife!) as an environment for growth and personal development. OTOH, it could be these are much smaller long term effects in terms of how someone behaves as an adult (i’m separating this from cultural expectations, which is a non-shared environmental effect whose magnitude i don’t deny).

    let me start a comment in the direction an alternative discussion which acknowledges genetics might go. the reality is america is now dividing between a bourgeois middle-to-upper class (college educated) and the rest. the norms and values of 1950s america probably did give some important foundations to the non-college fraction of the population, which is not as intelligent, and has a personality which is more amenable to short term gratification in a destructive fashion. these are people which need clear, simple, and hard and fast rules. the trade off here is that individual self-actualization is minimized. so for the elite the old order left a lot to be desired. liberals and conservatives may make different economic arguments here. for conservatives social liberalism is a lifestyle consumption good of the affluent, while liberals might respond that poverty is what is driving social dysfunction among lower orders. a society which acknowledged individual differences due to disposition might frankly admit that *universal norms are just not optimal in maximizing human flourishing* a crass/coarse way to go about it would be for conservatives to tell poor people that social libertinism is a luxury that they can’t afford, but that’s because american conservatives accept the universalist de facto homo economicus view whereby all individuals are interchangeable for all others. the premise is false, and different individuals need different guidelines in life. it’s like admitting the necessity of specialization and tracking vs. saying that everyone has to maintain a very high minimum of proficiency in academics.

  14. Both arms of the graph seem to be the flukes of social geography. On the left are localities in Utah and Idaho where people married and had children very early, when the parents were nominally poor but solidly rooted in the middle class and destined to belong to the middle class, and in the meantime strongly supported by extended families and the Mormon Church. On the right are the disadvantaged communities in the South and the Appalachia where national-level poverty was the local equivalent of middle-class status.

    Until they quantify the effects of single parenthood after controlling for other variables, one can’t be sure how important is the marital status. Until they quantify the effect of “shaming mothers into marriage”, one can’t be sure if the “marriages of shame” make any better children than (of recent news fame from India) “rapes of shame”

  15. Dmitry Pruss, my understanding is that Appalachia tends to rank high on social mobility in these studies. Sailer has noted that stylized fact should serve as a reality check against what the studies are supposed to show.

  16. Jeffery S, I must have expressed myself poorly. I did not mean to say that “Stupid” parents would tend to produce stupid children.

    What I meant to say was that even a family composed entirely of low IQ people would in most cases fare better for staying together for mutual support.

  17. Lumping data for pop headlines
    Looking just a little closer at the data

    In 2011, 62 percent of women between ages 20 and 24 who had recently given birth were unmarried.
    Among women ages 35 to 39, the percentage was considerably lower — 17 percent of women who recently gave birth were unmarried — but overall, 36 percent of the 4.1 million women who reported they had given birth in the past year were unmarried. That’s up from 31 percent in 2005.
    The birthrate among single mothers also varied along educational, socioeconomic and racial barriers. Sixty-eight percent of black women who had given birth in the past year were unmarried, compared to 11 percent of Asian women, 43 percent of hispanics and 26 percent of non-Hispanic whites. Fifty-seven percent of recent mothers without a high school diploma were unmarried compared to nine percent of recent mothers with a bachelors degree or higher. Sixty-nine percent of recent mothers who came from households with incomes with $10,000, in contrast to nine percent of recent mothers with households earning $200,000 or more.

    http://www.huffingtonpost.com/2013/05/01/single-motherhood-increases-census-report_n_3195455.html

    So it would seem like Genes Do Matter, not only in medical matters as is now begrudgingly being admitted, but including such matters as social and familial interactions, and personality structures.
    Culture is a strong environmental selector. “Albion’s Seed” for instance describes the persistence of a cultural style for generations. It would have had to have selective effect. We can clearly see its political influence today.

  18. Razib: You suggest one consequence of openly admitting the influence of genetics: a society which acknowledged individual differences due to disposition might frankly admit that *universal norms are just not optimal in maximizing human flourishing*

    I can only speak for myself, but personally, while I endorse the above line about conservative social structures and values (traditional families, religion, etc.) being most important for the lower class when it comes to the most *measurable* outcomes (poverty, crime, etc.), I think that they are important to those of us in the middle and upper class in less concrete ways — e.g., giving a sense of purpose in life and helping the individual see him or herself as something larger and more significant than him or herself. (The need for both of these, I think, is anecdotally attested to by the evangelical fervor with which many progressives approach “bettering society” through politics.)

    I think there is at least some empirical evidence in favor of this claim. For example, I think that the positive correlations between conservative social values and happiness, religiosity and happiness, and marriage and happiness hold for all income groups and social classes. If that’s right, it suggests that the cognitive elite may not be giving up that much when they miss out on “individual self-actualization.”

  19. For example, I think that the positive correlations between conservative social values and happiness, religiosity and happiness, and marriage and happiness hold for all income groups and social classes.

    be careful about relying on the happiness literature.

    1) it aggregates, so obviously it’s going to eliminate individual differences. do you think a gay man was happier in the 1950s? there’s only so much that values can do for people whose flourishing is not enabled by conventional ends because of their disposition

    2) happiness literature often also supports the proposition that having children makes you unhappier. yes, there are caveats and counter-arguments, but honestly that’s still the preponderance of the literature.

    also, as a point of fact, for most of their lives the middle-to-upper classes live traditional bourgeois lives. so i don’t see the overwhelming concern about moral decline in this sector. they espouse the values of the 21st century, while often living in the 1950s in their personal lives by middle age.

  20. Thomas Meehan:

    What I meant to say was that even a family composed entirely of low IQ people would in most cases fare better for staying together for mutual support.

    That’s probably true for the parents, but even among the not-so-bright, children generally leave by early adulthood. The important question is how much better off would a child of low or middling intelligence be at, say, age 40, if her parents had stayed together than if she grew up in a single-parent home. For kids who are college material, having two parents will make getting into and through college easier (due to the costs these days), but how much will it help for those not going to college?

    And how does that help manifest? Having a father in the home will likely make it easier for a young man to get better blue-collar jobs, just through connections. There seems to be a correlation between having a father in the house and not getting pregnant as a teenager, but teasing out the genetic vs environmental influence is hard. (It shouldn’t be impossible, as there’s a society a lot like ours, with lots of data, where low time horizons didn’t lead to nearly as many divorces as in ours. It’s called the 1950s. But what other confounds are there?)

  21. As for measurement error correction (aka correction for attenuation).

    http://en.wikipedia.org/wiki/Revised_NEO_Personality_Inventory#Reliability_of_the_NEO

    From Wiki data, the r_xx of Big Five with the usual test is on avg. .888.

    Given the usual number of .5 heritability for OCEAN traits, when this comes from MZT-DZT studies, this means the r_mzt = .5, and r_dzt = .25, because h^2 = 2(r_mzt-r_dzt) [Falconer’s Formula].

    So correcting for error yields:

    MZT: .5/(sq(.888*.888))=.5/.888=.563
    DZT: .5/(sq(.888*.888))=.25/.888=.282

    corrected heritability = 2(.563-.282)=.562, or 56%. Higher, but not 70%.

Comments are closed.

Subscribe to All Razib Khan Comments via RSS