Achievement Gaps Shrunk Faster in the 70s and 80s than Over the Past Decade

Yesterday I noticed Dylan Matthews of the Washington Post and Matt Yglesias of Slate tweeting that there is proof education reform is working. They cited this set of charts of NAEP score improvements since 1996, posted by Mac LeBuhn, a policy analyst at Democrats for Education Reform. 

I hate to be a downer, but attributing this good news to recent reform pushes, like No Child Left Behind, Race to the Top, and new teacher accountability schemes is extremely iffy, just as Stephen Sawchuck of Ed Week pointed out. Here's why: It just so happens that we have NAEP scores since 1971, and in the area of 8th grade math, which LeBuhn highlighted, the increase in raw scores and reduction of the achievement gap is actually a longterm trend. Take a look:

Screen Shot 2013-09-06 at 12.41.26 PM
In fact, achievement gaps shrunk much faster during the 1970s and 1980s than they have over the past decade, essentially because of skyrocketing performance from black children while white children remained relatively stagnant. Was this because of education reforms predominant back them, like school desegregation? Or because of demographic changes, since a more diverse group of students with more challenging backgrounds take the exam today? There are endless hypotheses, but no proof that any one kind of reform, or even reform itself, has led to these changes.

5 thoughts on “Achievement Gaps Shrunk Faster in the 70s and 80s than Over the Past Decade

  1. Longwalk Downlyndale

    Well yes, but shouldn’t we just file this in the proving-causation-in-social-phenomena-is-really-hard category? I mean try as he might Kevin Drum will never be able to definitively prove his theory that lead exposure to children is what caused crime rates to go up starting in the 60′s and down starting in the 90′s. But that doesn’t mean having lead in gasoline “doesn’t matter.”

    At the very least we can acknowledge that things have been getting better during the “reform era” and while we can’t definitively prove that say charter schools (or insert a reform policy of your choice) are directly responsible for this, we can say that modest education reform hasn’t caused our public school systems to get worse, as many critics of reform have predicted over the years. In fact it appears to be correlated with improving test scores. Which in the world of nation wide social phenomena is pretty good evidence comparatively.

  2. Gary Alan Chamberlain

    As I look at the graphs, they also suggest that the narrowest gaps were back in the ’80′s—1986, to be precise, early in the cycle which has seen the income gaps between the top and bottom deciles of our population expand. I read “The Smartest Kids In the World” (devoured it, actually, taking notes all the way) because of Ms. Goldstein’s excellent review. One of Amanda Ripley’s theses is that race and family income do matter, but they matter much less in other countries than we permit them to matter here. While i can’t, from the charts, adjust for the “adjustment” in format made a decade ago, it appears we’re going nowhere in the last thirty years. A good test (not for “proof” but as a basis for, say, Bayesian assessment) would be to identify comparable instruments in other nations. How is the UK doing with Pakistanis, France with Tunisians, Germany with Turks? What about Jewish refugee families in Israel? We do already know that our policies and practices in education place us near the bottom of the heap, while we rank second in the world in per-pupil expenditure. Top-dollar expense, third-rate results, in education as in health care. Whatever we’re doing badly, wouldn’t it behoove us to work much harder at identifying and changing the (almost-surely) multiple and interacting factors? I’ll keep checking this blog because I’m learning from it. Good stuff.

  3. mpledger

    Two things that I see that are different between the USA and other countries is
    a) other countries don’t always let illegal residents attend school (my country for instance doesn’t but we have very few illegal residents so it’s not a big problem)
    b) other countries don’t always mainstream physically and intellectually impaired children – the cost of their education (if any) gets hidden in the health system.

    Both of those differences make a pool of kids in America that are harder and costlier to teach.

  4. Steve Sailer

    “the narrowest gaps were back in the ’80′s—1986″

    The mid-1980s were a fairly good period for African Americans. For example, the black-white gap in homicide rates was relatively low then. In the late 1980s, however, black culture was hit hard by two related phenomena: crack and gangsta rap (which extolled the crack dealer ethos). Did this affect school test scores? Seems not implausible …

  5. Steve Sailer

    “How is the UK doing with Pakistanis, France with Tunisians, Germany with Turks?”

    Generally speaking, not well. The 2009 PISA international test scores include data broken out by recent immigrant ancestry. The children of immigrants score lower than the long-term native population in every OECD county except Canada and Australia. Those two countries have points systems intended to let in high human capital immigrants and keep out less promising would-be immigrants.

    In contrast, educational success story Finland has the lowest percentage of immigrants of any Western European country (and many of those tend to be Swedes, Estonians, Russians and the like).

Comments are closed.