School smarts

Each year of schooling raises a child’s IQ, writes Catherine Johnson on Kitchen Table Math. She quotes a researcher on a review of over 50 studes:

These studies, conducted throughout the 20th century, comparing schooled and non-schooled populations, have estimated that the enhancement of IQ by schooling ranges from 0.3 to 0.6 of an IQ point for every year of school competed. Importantly, the association between IQ and exposure to formal education is not only due to children with higher measured IQ staying in school longer.”

I know there’s been a steady increase in IQ over the last century as education has expanded.

About Joanne


  1. wayne martin says:

    According to this WEB-site, Global IQ is constantly going down–

    Global IQ: 1950-2050:

    If this premise of the study is correct, US society is spending about $10,000+ per year per student to purchase .6 of an IQ point. This comes to about 7.2 IQ points for a HS Grad, and 9.6 for a College Grad.

    Of course, we’re going to have to ask: “Are we getting our money’s worth?”

  2. Indigo Warrior says:

    Actually the article states that good schools raise IQ, and that bad schools lower IQ.

  3. Miller Smith says:

    In ten years my child in a good school will go from 100 to 106? That’s it? Someone is seriously blowing a trumpet over this news?

    This is a remarkably small effect. This still places nature over nurture as the biggist IQ effect.

    6 points. Sorry, but I can’t get worked up over something that small.

  4. Hey!

    It’s better than having him go from 100 to 94!

  5. Isn’t no one worried about the fact that the “finding” is mostly the unproven thesis that the Flynn effect is primarily caused by schooling? The review of the literature points to a very mixed set of papers, not all of which would pass muster in the psychometric literature, few which really try to demonstrate that early year gains are stable into adulthood, none which dismiss the findings of the Minnesota twin studies (the closest thing we have to real controlled experiments), and few to none which make use of the sort of careful instrumental variables techniques that are being used to tease out causality in state of the art econometrics.

    It would be nice if this paper’s ideas were true, but this looks like the sort of “consensus” that psychometricians or economic statisticians like Levitt could easily destroy with a bit of careful statistical digging.