The New York Times reports that absurd numbers of college grads are applying to Teach for America — “12 percent of Yale’s graduates, 11 percent of Dartmouth’s and 8 percent of Harvard’s and Princeton’s” and “a record 17,350 recent college graduates” overall last year.

That sounds great, but the problem is that the program isn’t very good. Everyone I’ve ever known who has done it was thrown into a difficult urban classroom with little preparation and quickly became burnt out. As a result, they all left teaching at the end of their stint rather than getting the training they would need to be successful. End result: no new recruits for the nation’s teacher corps, no qualified teachers in failing urban schools.

Here’s the evidence on the effectiveness of the program reported by the Times:

It has also helped, on all campuses, that Teach for America now has a track record: An evaluation last year by Mathematica Policy Research found that Teach for America members produce slightly higher math achievement and no worse English results, than other teachers. And a June 2005 evaluation by Kane Parsons & Associates found that 63 percent of the principals in the schools where they work regarded Teach for America teachers as more effective than the overall faculty.

However, a study of Houston student achievement released this year by Linda Darling-Hammond of Stanford and others found that although Teach for America teachers performed as well as other uncertified teachers, their results did not match those of certified teachers. Teach for America officials contend that the study was flawed.

Let’s review: “slightly higher math” and “no worse English” results count as a positive finding? We’re spending all this money and human capital to barely nudge the bar? And the Darling-Hammond et al study supports the conclusions I present above — namely, unqualified teachers don’t perform well, period, no matter how smart or energetic.

That’s not to say that TFA isn’t well-intentioned or that the people in it aren’t trying their best. But couldn’t all that good will and money be put to better use? In addition, TFA falsely suggests that bad urban schools can be fixed on the cheap with low-paid, undertrained young teachers. The real problem, however, is far more deep-seated and difficult to solve. Let’s not fool ourselves.

Update 6/11/08 9:18 AM: A new Urban Institute report (PDF) finds that TFA grads are more effective teachers than previous studies have suggested. It’s an encouraging finding, but unlike the Mathematica evaluation cited above, it was not based on a randomized evaluation, so I have less confidence in the results.