Society and Culture

Test Scores and Wishful Thinking

A few days ago, Chris DeMuth emailed me this graph prepared by our colleagues at the Cato Institute. I laughed out loud. Others may just as reasonably cry.


I assure you that the graph does not exaggerate the case. It would look about the same if any other metric (for example, standard deviations) were used to express the changes in test scores and spending.

The day after Chris sent it to me, I was attending a convocation sponsored by The Atlantic magazine and the Aspen Institute, which consisted of many prominent officials being interviewed by prominent journalists. It was a fascinating conference, in which most people other than administration officials were reasonably hard-headed—about everything except education. It was as if the history represented by that graph didn’t exist. As if test scores would zoom up if we just provided the education to poor children that we know how to deliver to rich children.

I’m not saying that schools can’t be improved, just that test scores are a lot easier to depress (see the great SAT verbal score decline in the 1960s and 1970s) than they are to raise, and that such a huge proportion of a child’s educational prospects are determined by things other than school (genes and the non-school environment) that reforms of the schools can never do more than produce score improvements at the margin. There’s only one exception to that generalization: intellectually gifted children who can absorb as much as teachers can throw at them—the one group on which the federal government spends no money at all.

Listening to the eminences at the Atlantic/Aspen conference talk as if achieving equality of educational outcomes was just a matter of putting our minds to it, I tried to decide whether they are three-year-olds who really do believe in Santa Claus, five-year-olds who still want to believe so badly that they fool themselves, or seven-year-olds who know the truth but pretend otherwise. I truly don’t know.

Comments are closed.