Politics and Negative Results

March 15, 2011
Posted by Jay Livingston

Social scientists often end their reports and articles with suggestions for further research. Politicians, not so much. They want action. (Didn't Weber say something about this?)

There’s a government program for kids that is very popular among the low-income parents, but a recent large-scale government study shows that it produces no lasting educational benefits. What to do? If you’re a conservative, you get rid of it – all in the name of cutting costs of course. The program is Head Start, and funding for it is under heavy attack from the right.* (NYT story here).

Charter schools have a similar profile.. Low-income parents want them, but the broadest research shows that on average, compared with public schools, charters do no better and probably worse. Yet conservatives can’t fund enough of them.

Liberals may be almost as inconsistent, backing Head Start while opposing charters. I say “almost” because Head Start does produce some benefits. As the Times article reported,
Research on the program has shown that children who complete it do better socially and academically than children not enrolled in the program, and that they tend to have lower high school dropout rates. But the initial test score benefits tend to fade out by first grade.
So the “doesn’t work” mantra being repeated on the right is not quite accurate.

In both cases, what seems like a reasonable idea hasn’t worked out in practice. True, some charter schools and some Head Start programs do produce positive results. The trouble is that there’s no evidence of consistent, broad success. You’d think, especially if you are a social scientist, that the next step would be further research to figure out what the effective ones are doing that the ineffective ones are not (and vice versa). Maybe such research does exist, but if so, it’s not getting much press.

Instead, politicians to take a baby-with-the-bathwater approach. If a program fits with your ideology, fund it, no questions asked. If not, get rid of it.


* The first post on this blog (including an apt joke borrowed from Kieran Healy) was about the scant attention given to negative findings . I had not realized then that when the findings are in the interests of those with easy access to the media, the noise level can rise considerably.

3 comments:

mike3550 said...

The caveat that you have about the program not working is not only "not quite accurate," but I would argue actually wrong. Head Start is designed to level the playing field and, if Head Start kids have some advantages into first grade then it has done that.

Why we would expect a child's past participation in Head Start to outweigh variation in their current schooling? It seems unreasonable to expect it to do so; in fact, it is a large expectation of any program or policy to have benefits that last throughout one's lifetime.

That said, I don't know the study or whether liberals would be more likely to do the same (e.g., over abstinence-only education or charter schools); however, I believe that saying that Head Start falls short of expectations is partly due to the rather large expectations placed on the program.

Jay Livingston said...

Mike, I don't know the research, but I assume it's comparing Head Start kids with controls from a similar background. So, to me at least, it's disappointing that the gains don't last. Disappointing, but not surprising. I'm also guilty of those inflated expectations you mention, at least in my less sociologically aware moments. It would be nice if there were a silver bullet or a Superman.

PCM said...

To me it's both disappointing and surprising.

One of the the things I learned in graduate school and actually believed was that Head Start works. That it helps kids and was sort of a (partial) silver bullet.

So what happened to all that research from the 1990s? Was it wrong?