Jay P. Greene's Blog

(Guest post by Greg Forster)

This week Jay highlighted the fact that a study showing a positive correlation between test prep activity in the classroom and improved learning is being portrayed as having shown a negative correlation between test prep activity (“drill and kill”) and improved learning. At this point it’s not known (at least to me) where the error arose, and I don’t have anything to say about the question of who said what. But I think it illustrates how the whole subject of drillandkillaphobia needs to be revisited.

Lately I’ve explored in some depth how testing has come to be the focal point of the fight between the two great factions in the education space, the “liberal artists” and the “pragmatists.” Liberal artists have gradually come to invest all their hopes in standardized testing. Pragmatists have gradually come to invest all their fears in precisely the same thing. This is a mutually reinforcing circle – over time, the liberal artists increasingly think testing must be good because the pragmatists hate it, and the pragmatists increasingly think it must be bad because the liberal artists love it. So naturally the battle line over testing has become more and more absolute.

The problem, as I’ve tried to show, is that this dynamic causes each side to reject something that’s essential to good education. The liberal artists seek curricula (or standards that can only lead to curricula) that emphasize sterile head knowledge of facts to the exclusion of practical problem-solving. We don’t know how to test what Daniel Willingham calls “deep knowledge” of subject area content, and even if we did the tests would be almost infinitely vulnerable to manipulation if they were ever used for accountability purposes. But we do know how to create rigorous tests for head knowledge of facts, so the liberal artists define “subject area content” to mean simply head knowledge of facts.

The pragmatists do the opposite. And since I was pushing pretty hard last time to emphasize what I thought were the dysfunctions of the liberal artists, I’d like to balance the scales with something about drillandkillaphobia.

Merely mention the subject of testing and it seems that pragmatists instantly jump to the conclusion that you want schools to look like this:

Now the interesting thing is that these days, they will engage in the most hysterical drillandkillaphobia while all the time affirming that we need to keep standards high, content knowledge matters, etc. To some extent I think that must be intentionally tactical – P21 knows that its flavor of loosey-goosey crunchy granola doesn’t sell these days, since they’ve lost a lot of battles in the war of ideas. But I don’t think that explains all of it. Take another look at that Ken Robinson “Changing Paradigms” video, where he begins by saying that of course we don’t want to lower standards. Clearly he, at least, really means it. It seems completely obvious to him that there’s no contradiction between his attacks on testing and his affirmation of high standards.

And that’s the problem. It seems so obvious to him that he fails to even take the question seriously. (That was one of the key points in Willingham’s stimulating critique of Robinson’s video.)

If the liberal artists need to get over their testophilia, the pragmatists need to get over their drillandkillaphobia. I’m not aware of any hard evidence linking test prep to worse outcomes. Sure, lots of people are really convinced that it must be the case, but that’s hardly a solid ground for making policy. (The floor is open in the comments section if you have some hard evidence you want to share.) And it’s not like this is a new question. I’ll admit that I haven’t done an exhaustive review of the research (again, if you have, the floor is open) but Jay and I conducted a study a while back showing that attaching rewards and penalties to a test doesn’t change the results; that would seem to speak right to the heart of drillandkillaphobia. This new Gates Foundation study, finding a positive correlation between test prep and learning outcomes, would seem to be another piece of evidence against it.

People can’t learn what Willingham calls the “deep structure” of practical problems until they’ve learned what he calls the “surface structure.” You can’t get from the pool deck to the bottom of the pool without passing through the surface of the water; similarly, you can’t get to deep (i.e. practical) content knowledge without first getting shallow (i.e. factual) content knowledge. If you like, it’s “merely” or “sterile” head knowledge. But head knowledge it is and head knowledge it will remain, even after you add the “deep” part.

People learn head knowledge by memorization. And any kind of memorization will appear, to those who wish to stigmatize and delegitimize it, to be “merely” “rote” memorization. You can call it “regurtitation” when people know facts, and in a sense you’re right – but people do need to know facts and be able to summon up that knowledge as necessary, whether you call that “regurgitating” or not. And for everyone but the real genius students, gaining head knowledge of facts will involve some kind of “drill.”

As the pragmatists themselves never tire of reminding us, real learning is hard. Well, yes it is. You can’t learn if you don’t memorize stuff. Memorizing stuff is hard and unpleasant, and it’s a lot more so for some kids than for others. That’s the world; deal with it.

I fully admit that if you really want to learn you are never just memorizing. You must be trying to understand the facts you absorb – understand their significance and the connections between them. But while it must be more than memorizing, it is never less than memorizing. Of course, if it must be more then by definition it can never be less.

Take that great, perennial boogyman of rote memorization – historical dates. People whine, why does it matter in what year a certain event occurred? Well – why does it matter? If you stopped and seriously asked that question and sought out the answer, you might . . . well, you might learn something.

The great irony, of course, is that at the same time the pragmatists are pushing this new bout of drillandkillaphobia, they’re working hard to impose a federal-government controlled system of national testing – excuse me, a totally “voluntary” system of “common” “assessment” that has nothing to do with the federal government, nothing to see here, move along, pay no attention to the man behind the curtain. That really is an attempt to handcuff real learning and turn us all over to the benevolent dictatorship of soulless men in white coats who must be trustworthy because, after all, they’re scientists. And that, of course, was always the original sin of the Progressive movement, in its educational form as in all other forms. Handing over all power to a tiny priesthood is the very epitome of “democracy” as long as we’re careful to call the priests “scientists.” But now I’m broaching a whole new and much deeper subject, one that will require another post to handle with any justice.