IQ Tales

In a previous post, I had this to say:

I have no data on this at my fingertips, but my intuition tells me that if Head Start’s “parenting practices” interventions are to have any impact at all, at age 3-4, they are coming WAY too late.

Today, I happened to catch a link to PBS Newshour’s interview with James Heckman. I was seriously irritated by Paul Solomon, but I will deal with that at the end. Here is what Heckman has to say about very early childhood education:

If you take disadvantaged, minority children starting at age six to eight weeks — I mean, they’re literally just born — and you follow these kids and give them intensive interventions for about eight years, you can boost their IQ at least up to age 21. You can see permanent differences between the treatment and control groups in both men and women, boys and girls.


I don’t want to make this into a very precise science where I can say, “At age three months and two days, you should do this.” I wish to God that we knew that, but we don’t and we’re working on it. But we do know roughly that the early years are very important and the later we wait, the harder it is. After age 10, IQ becomes very, very hard to change.

I think that other data vaguely indicate that the rigidity of IQ increases dramatically well before 10 years of age, so that was probably a conservative estimate by Heckman. Here is Heckman’s paper on the value of early childhood education. I don’t have a strong opinion either way on universal pre-school, and I have been known to make the case that even if it doesn’t have any effect on educational variables, the program still generates a lot of value as a daycare. I still think there would need to be a dramatic revolution in the educational delivery mechanism to reap any sort of outsized reward from educational intervention.

Now turning to Mr. Solomon, who really brings down the interview by interspersing ignorant bits like:

But work in “microeconometrics” — the statistical study of individual responses to public policy — has reached decidedly liberal conclusions.

Let me make this clear: Heckman’s research has made some significant discoveries, and has lead to a set of conclusions, period. These conclusions are not “liberal conclusions”, they are simply conclusions — and, like all studies, Heckman notes that much more research is needed. He even notes this in the interview itself. The “liberal conclusions” statement above is about as annoying as claiming models are useful for their “predictive power”. Models don’t make predictions, humans do. Models are explanatory tools. In much the same way, data doesn’t reach “liberal conclusions”, if a human happens to interpret data to support a liberal conclusion, it is a person that comes to the conclusion. Just for shits, lets come up with an ad hoc “conservative” conclusion based on the data: we need to embrace traditional family values and economic organization from mid-20th century so that women are able to spend more time interacting with their newborn children.*

That was easy.

Solomon also makes this childish remark in a question to Heckman:

You’re a University of Chicago economist, which suggests a certain conservatism with regard to economics, right?

Awful. Heckman answers well.

h/t Matthew Yglesias

*I do not endorse this conclusion.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s