It’s back! The undead duck (canard in French) that class size doesn’t matter is on the loose again, quacking now in The New York Times. The question must therefore be asked: why do teachers say they would reject a $10,000/year raise if it meant an increase of 3 in class size? Why do parents want smaller classes for their children? The obvious answer is that class size does matter.
Garlic, stakes, & silver bullets don’t seem to work, so let’s examine the “science” behind the claim. Very little of it is reliable, even by the wildly generous standards of the Ed Biz, but one study most people seem to accept is the Tennessee STAR Study. In it there appears to be no difference in the “educational effect” of classes of fifteen and of twenty-three.
This is a difference in class size of 8, nearly three times the number a plurality of teachers would not want $10,000/year to accept. Something appears not to add up, but the reality is not that the educational effect is equal. Rather, it is that a class of fifteen is no better at preparing its students for the Stanford Achievement Tests in English and Mathematics than is a class of twenty-three. Raise your hand if you think that everything learned in a classroom—even an English or math classroom—is assessed by the SATs. No hands at all? Very good! This class is obviously not gulled by studies that use inadequate proxy values.
As yesterday’s posting points out, much that students learn in a classroom goes beyond what can be transmitted by didactic instruction or assessed by multiple-choice tests. What is more, the SATs do not penalize students for wrong answers, thus giving them a probabilistic “gift” of 20% of their guesses as right answers. I have dealt with “test-taking skills” that have nothing to do with knowledge, but the answers gained by the use of these “skills” or by guessing are counted no differently from genuine knowledge. Let a student try to bamboozle the teacher of a small class in a carefully graded essay or a personal interview and see how far he gets.
This is what the parents and teachers who want small classes recognize. When a writer therefore says that mandating large classes is “good policy but bad politics,” she has got it quite wrong. It is bad policy; and the politics, which simply recognize that truth, are good.