In the WSJ, Robert Plomin writes,
DNA is the major systematic influence making us who we are as individuals. Environmental influences are important too, but what look like systematic effects of the environment are often genetic effects in disguise: Parents respond to their children’s genetically driven traits, and children seek, modify and even create experiences correlated with their genetic propensities.
His book is Blueprint, which I just finished. His thesis:
DNA is the only thing that makes a substantial systematic difference, accounting for 50 percent of the variance in psychological traits. The rest comes down to chance environmental experiences that do not have long-term effects.
What he calls “chance environmental experiences” could be measurement error. Measurement error always holds down correlation. This raises the possibility that some traits that are measured with error are more heritable than they appear. For example, Gregory Clark found that social status is much more heritable across many generations than would be expected based on parent-child heritability estimates. I explained that this is likely due to error in measurement in social status, which lowers immediate-generation correlation more than multi-generation correlation.
Educational interventions are apparent environmental influences that wear off over time. You raise a test score but do not fundamentally alter ability. That is an element of what I call the Null Hypothesis, which Plomin strongly endorses, although of course he does not use that term. Related: Scott Alexander on pre-school.
This is one of the most important books of the year. Coincidentally, the NYT has an article on economists’ use of polygenic scores. Tyler and Alex both linked to it.
But you should know that I came away from Plomin’s book less than impressed with polygenic scoring. So much data mining. So little predictive value. Also, there is serious criticism of his view that environmental factors exhibit no systematic influence, but he does not confront it. I did a search inside the Kindle edition for “Flynn” and found no results.
Arnold, you say:
“Also, there is serious criticism of his view that environmental factors exhibit no systematic influence, but he does not confront it.”
What is the “serious” criticism? Serious = high quality criticism? Lots of people don’t like it, but what is their evidence? 50 years of twin studies, adoption studies, studies of half-siblings.. all show that the reason siblings are alike to each other and their parent is genetics, and the reason they are different is the “non-shared environment”, aka, stuff that cannot be explained. The shared environment can explain essentially nothing. What does “confront it” mean. I’ve read this book, but I’ve also read his textbook, “Behavioral Genetics” (7th edition, lead author now Knopik).
I recommend the textbook for people who have questions.
“I did a search inside the Kindle edition for “Flynn” and found no results.”
IQ is but one trait. BMI has been increasing in western countries (and some developing countries). But the variance within the population shows a high heritability for BMI. Likewise for IQ.
You say “So much data mining. So little predictive value.”
Educational attainment is predicted at 11-13% of variance.
https://www.nature.com/articles/s41588-018-0147-3
This is on par with what SAT tests predict. The cap is probably around 15% of variance explained, since that’s close to the best you can do with grades/SATs/transcripts. So with more data and better scoring that amount should be reached.
Social science prediction is fundamentally difficult, highly multicausal. I think doing as well as the SAT test is in fact good. It’s far better than much of social science.
So I’m not sure where your comment on low predictive value is coming from. If you’re saying things are overhyped, that’s fair enough. If you are saying the predictions are not very valuable, then I suppose that’s also fair, but then your comment also applies to other types of predictions based on grades, SAT test, since they are comparable.
Maybe more discussion on what ‘explains’ means.
Please allow me to make a plug for Judith Rich Harris’ two books, _The nurture myth_ and _No two alike_. _The nurture myth_ is largely devoted to debunking the belief that parenting and environment explains a vast majority of things (controlling for genes).
_No two alike_ argues for the role of peers, peer groups, peer networks, and competing for status in a peer group. Intuitively this makes sense. Which peer group you end up in matters. Some of us remember drifting or jumping from one peer group to another. It really does matter, I will argue.
Her basic argument is that 50% of variance in a measured variable is genetics, 40% comes from “peer effects” defined broadly, and 10% is random / measurement error.
Her argument makes sense to me. I can’t do justice to her books here. She’s a funny and engaging writer, also, and heterodox in part because she was pushed out of academia early on and had to think for herself. Steven Pinker wrote a Foreword to her first book, presumably to help give it a boost and because he thought it merited attention.
I think peer effects can matter, but it’s something of a zero sum game. Good peers are those with good traits. Bad peers have bad traits. If you give a person with bad traits a peer group with good traits you’ve degraded the good peer group some, harming those individuals.
This is then further compounded by the fact that mixed peer groups can be net negative for the entire pie. Bad apples spoil the bunch. Or simply one slow learner slows down the group.
It’s probably best to focus on training good habits within peer groups, often tailored to the needs of that peer group. Rather then to assume we can solve our problems by distributing enough “good peers” around like a prize to be awarded.
Your point is a good one.
I think an intuitive way of thinking about the “peer effect” phenomenon is to imagine a family with two children and both are in high school.
One sibling in high school has peers and friends who do sports or band or drama club and generally do their homework and turn in their assignments. The kids in this group think it’s normal to go to college and get a decent job and “launch” into functional adulthood with a decent career and earnings stream.
The child has a sibling in the same family–same mother and father, also in high school. That child chooses friends who treat school as something one suffers under duress. In this peer group bad behavior is normalized. Recreational drug use is accepted as something to do which is normal, highly pleasurable, and not very risky or imprudent. One gains status within the group by engaging in “juvenile delinquent” type behavior, breaking the law (but not getting caught).
(Pay attention: According to Judith Rich Harris, neither group of teens wants to imitate their parents. THey want to imitate their high status peers. But they don’t have the same high status peers. What is normal in one group is not normal in the other group.)
The two siblings may very well be different in IQ, reading proficiency, frustration tolerance, sitzflesch, impulse control, vulnerabilty to addiction, novelty seeking behavior, ability to predict consquences of poor choices. But the peer group tends to push the two siblings in different directions, also.
To expand the argument a bit, here is this. Recreational drug use among teenagers spreads in part by “contagion,” some have argued. James Q. Wilson actually made this argument, so it’s not a crackpot notion.
Another way of thinking about peer groups is they facilitate a “snowball” sort of development in which you start at the top of the hill and roll snowballs down the hill to grow in mass and move farther from each other as they get farther from the top. I believe Martin Seligman made this argument for teenagers growing more different as they age. It’s in _What you can change and what you can’t_, ca. 1995.
My wife and I married each other at age 33, both divorced, both brought kids to the marriage: 9 in all (and all under the age of 14).
While it’s hard to tell which is the chicken and which is the egg, it’s clear to us 32 years later that the peer groups each chose — really starting in elementary school, but especially in junior high and high school — had a major influence on the course of their lives through their 20s, for good or for ill. Note that this cuts across their actual genetic heritage, that is, this appeared to be more of major influence than whether they were her kids or mine.
The good news is that all 9 are now independent, self-supporting adults with good productive lives; we love and adore them all, and vice-versa. But their choices of teenage peer groups were a solid predictor of the struggles some of them faced in getting to this place.
Maybe I am wrong, but if most of the important things about human beings are highly heritable (IQ, big 5 personality traits, beauty, etc.), doesn’t that lead you down the path that people should be fighting about the “environment”? That is, if you know that you are always going to have lots of people with below average IQ, and lots of people with below average conscientiousness, etc., shouldn’t you maybe think about how culture and institutions are going to interact with and impact those people? And likewise how they will impact people with different traits? That is to say, especially when thinking about institutional design and policy analysis, you cannot take a one size fits all approach, but instead should build a model in which you have different groups of people with different psychometric profiles who will respond to any short of change differently?
The title of Judith Rich Harris’ first book (justly recommended by charles w abbott above) is The Nurture Assumption. The Revised and Updated 2009 edition really is a significant revision of the 1997 original.
my bad. Thank you.
Oops, that shouldn’t have gone there.
P Burgos is exactly right. For example, if lots of people are just never going to be high in school smarts, it is ridiculous to predicate national policy on the idea that if we only make the schools good enough, everyone can be a high school graduate–or more ridiculously, a college graduate.
You are exactly right.
And that reality runs smack into the gains an individual or group can get from telling low-IQ and/or low-conscientiousness that the problem is really that the other group is keeping them down. Once that group gets large enough, the Heinlein Effect kicks in:
“Throughout history, poverty is the normal condition of man. Advances which permit this norm to be exceeded — here and there, now and then — are the work of an extremely small minority, frequently despised, often condemned, and almost always opposed by all right-thinking people. Whenever this tiny minority is kept from creating, or (as sometimes happens) is driven out of a society, the people then slip back into abject poverty.”
Welcome to Western Civilization 2035
Good gravy.
Exactly- policy should be formulated so that things work pretty well not only for the kids that graduate from college, but for everyone else as well.
Instead of dismissing environmental factors altogether, it seems like it would make more sense to explore which genotype groups respond to which environmental factors more of less than other groups. For example, Tyler Cowen pointed to the Gershenson, et al study the other day that, using demographic subgroups found that “Leveraging random student-teacher pairings in the Tennessee STAR class-size experiment, we find that black students randomly assigned to a black teacher in grades K-3 are 5 percentage points (7%) more likely to graduate from high school and 4 percentage points (13%) more likely to enroll in college than their peers in the same school who are not assigned a black teacher.” Suggests to me that you can find meaningful information when you disaggregate.
As far as I can tell, no one is dismissing environmental factors at all. From the point of view of behavioral genetics, half of human traits are unexplained. Not from want of looking for these environmental factors.
The key point is that if you ignore genetics when doing social science research you miss the largest confounding factor. Social science research just about always ignores genetics.
Judith Rich Harris’ work is good at teasing out how the term “environment” needs to be carefully specified. For example, siblings growing up in the same family with the same biological parents can have different peer groups.
Another example, Thomas Sowell once corrected a niece who said “Uncle Tommy, I went to the same school as you [but didn’t turn out as smart was the subtext].” Sowell replied “No, you didn’t. By the time you got there it was the same building but a different school.”
Even the same classroom can be experienced differently. Some sit in the front and take notes. Others sit in the back and pass notes to their friends (or text, nowadays).
What tends to be a more uniform environment? Perhaps military Basic Training. But that’s speculation and I’ve not experienced it first hand.
Nonetheless, let’s ponder that. To make the environment the same for all students, you have to be very authoritarian and strict. There is one way to make a bed or field strip a rifle. The more loosey-goosey or anarchic, the more the setting is different for the various students.