Tag Archives: rewards

Punished by praise

New York Magazine ran a feature story about a year ago, by Po Bronson, warning that telling your kids they’re smart is bad for them. The scientific peg is a series of studies by social psychologist Carol Dweck, in which one group of children was praised for their abilities on a task (“You sure are good at this!”) and the other group for their efforts (“You sure did a good job on this!”) The former group, at follow-up, was less inclined to take on challenging tasks, and more prone to give up after experiencing minor setbacks. Dweck’s explanation: the “abilities” group is getting the message that success is explained by your inherent characteristics. If that’s the case, then any task that presents difficulties must be one you’re inherently bad at, and there’s no point trying to get past the difficulties.

I’m reminded of a very strange argument I had with a tearful first-year calculus student, back when I was teaching at Princeton.

“This class is really hard!” she said.

“Math is supposed to be hard!”

“No, math isn’t supposed to be hard!”

“Yes, I’m the professor, and I say the course is supposed to be hard!”

“But it isn’t supposed to be hard!”

And so on.

When I started teaching I really liked Punished by Rewards, by Alfie Kohn. Kohn isn’t just against praising students’ abilities — he’s against praise of any kind. And he doesn’t like grades, good or bad. His contention is that all these things instill in students the pernicious idea that the goal of learning is to get the reward (the A+, the compliment from the teacher, the approval of the parent) And that the praise rewires you so badly that once you get hooked, you can never really go back to learning for its own sake.

And now? I’m not the radical teacher I used to be. But we do try to avoid imputing to CJ too many inherent attributes. “That was a good song!” not “You’re a good singer!” “We like it when you share,” not “You’re a nice boy.” It’s not as hard as Po Bronson makes it sound.

Tagged , , ,
%d bloggers like this: