Does Bribing Your Kids Help Get Better Grades?

One of my favourite "pop" economists is Steven Levitt. He is a professor of economics at the University of Chicago Booth School of Business, but is probably better known for his book, Freakonomics.

Recently, he posted a blog and podcast discussing his field research into bribing kids to try harder on tests at school. Financial incentives are nothing new, but a working paper he published in September, 2011, revealed some interesting results.

The Summary.
A total of 6,500 students in the Chicago area were involved. The students, prior to a standardized low stake test, were offered a reward (either financial, or non-financial, like a trophy or toy), just prior to taking the test. The reward would be given out immediately following the test, if improvement was shown from previous baseline testing. The control group was told nothing. In an added twist, a "loss condition" group was actually handed the reward (ie. cash or trophy)  before writing the test and had it sit on their desk while they wrote. It was taken away from them if they did not improve their mark, versus the reward group receiving their reward upon completion and improvement.

The experiment also was conducted with a delay arm, giving the reward out 1 month after testing versus no  reward. Finally, the results were analysed based on age of student, arbitrarily divided into younger children (grades 2-4) and older kids (grade 5-8).

Results were interesting. For younger children, non-financial rewards were just as effective as cash. However, in the older group, cash was king, with no benefit seen for trinkets and trophies, or even lower dollar amounts ($10 vs $20).

Most interesting though, was the "loss" group had the most significant impact. Students who were given cash prior to testing with the potential to having it taken away saw the biggest improvement.

Conclusions for Real Life.
This is pretty intriguing. Levitt's study seems to show that getting the most bang for your buck is to actually give your child a reward for doing their best on a test (ie. getting an "A" or whatever goal you designate), but adding the caveat that if they don't achieve it, the reward gets taken away.

I don't think I could be that cold-hearted, but it makes me rethink offering a little something for working a little harder.

Well, this should open up a can of worms.
What do you think?
Do you offer any rewards for tests at school? How about sport?
Does this research make you want to change your tactics?
Let me know!