Way back when “Return of the Mack” was on regular rotation in my off-campus apartment and Randy Quaid saved the planet from aliens, I first learned about Alfie Kohn. I was in an undergraduate teacher prep class and we read an article of Kohn’s (it might have been this one) where he argued that rewarding kids at school for things they did well wasn’t any better than punishing them for things they did poorly. Kohn expands on this idea in his book, Punished By Rewards, which made a big splash in the 90s because, while society had moved away from the draconian punishments of yesteryear and state laws now forbade corporal punishment, rewards were passed out like, well, candy. Or colorful pencils. Or those awesome scratch-and-sniff stickers. Or gold stars. Or promises of ice cream parties. Or erasers. Or, well, you get the point. And now here came Kohn scolding teachers all over again.
And so I started my teaching career as most naive, just-released-from-college kids do. With the proper amount of self-righteousness and arrogance, I marched into my classroom determined to offer no rewards. Students would learn for knowledge’s sake. We would build a community and have respect for each other. We would talk about our problems and address underlying causes of misbehavior.
Then the real world hit and doing all of those things was really, really hard.
Some kids were just plain jerks who needed to be taught a few hard lessons, if only so the rest of the class would see that you can’t go through life treating people like dirt and get nothing harsher than a counseling session, a behavior plan, and rewards for doing the very things every other kid in the class was doing as a matter of course. And so I started rewarding some kids, punishing others, and playing that whole game.
And not long after that, I learned first-hand what I had read in a boring old classroom. Alfie was right. Rewards don’t really work. They’re manipulative, frequently arbitrary, and basically no different than punishments (they just feel nicer).
Fast-forward to 2011 and Daniel Pink’s book, Drive, made many of the same arguments. Citing some of the same research as Kohn, Pink concluded that extrinsic rewards are usually a bad idea. Motivation is largely intrinsic and the way to tap into that motivation is through autonomy, a slow and steady march toward mastery, and by doing meaningful things in service to something larger than the self.
All of that is well and good. I accept that it’s generally a bad idea to reward students for their performance and to bribe them to behave better. Make the work interesting. Offer choice. Don’t be such a dictator. Provide feedback so students understand their progress toward mastery. Assign meaningful work. Do all that.
The testing window opened in my state this week. Over the next two months, students from third graders to high school juniors will take The Big Test. And big it is. Schools will be judged on the results. They’ll be labeled on some silly statewide reporting system. Some will face consequences. Teachers will be evaluated based on the results. Some may lose their jobs. The scores will influence public opinion of American education as a whole and either burnish or tarnish the reputations of districts, schools, and even entire state’s education systems and policies.
There are plenty of problems with The Big Test, (one of which might be the questionable timing of asking students to take it after they’ve just had 10 days off for spring break, as my wife’s students did this week) but perhaps none are bigger than this:
There is no reason students should try hard on it.
In my state, students get nothing for doing well (it’s kind of like being a teacher in that regard).
No scholarship money.
No name in the local paper.
Not even a pat on the back.
Students suffer no negative consequences for doing poorly.* Nothing will happen to a student who decides to treat the entire enterprise exactly how it deserves to be treated, as a joke. Their scores won’t be reflected on their report cards. Grade point averages will be unaffected. Graduation is not at risk. Students’ parents won’t even learn the results for a number of months after the test is over (and by then, most won’t care). Students won’t be retained or asked to leave school. The only thing they lose is time, and they lose more of it the harder they try.
Subjecting the exams to Pink’s criteria, the tests offer its takers no autonomy. Because it’s a one-time event for which they receive no useful feedback, students cannot progress toward mastery. As for meaning, there is no purpose that students give a hoot about. It is, for almost every student, the very definition of drudgery. It’s busy work. By the state’s own declared aims, it’s got nothing to do with them. For students, it’s as low-stakes as you can get.
All of which is why you should unabashedly bribe your students to take their time and do their best.
In the adult world, we offer money. In the classroom, we offer pizza, ice cream, a dance party, video game time, or anything that will make students think twice before just clicking on answers so they can be done with the thing. When there is no expectation of intrinsic motivation, we have to find other ways to get people to try.
And here’s the thing: Bribery works! I have proof!
Every three years, 15-year-olds from around the world take the PISA exam. The results of this test are reported breathlessly in education circles and often lead to huge policy changes in the countries of the students who struggle. A group of researchers wondered an obvious thing. Did kids actually try on these tests? They had reason to be skeptical. There are no stakes for students who take the PISA; they never even get to see their results. And student effort matters. As I tell a handful of parents every year, it’s hard to report on a student’s abilities when they don’t try on their work.
American students traditionally fall in the middle of the pack on the PISA, but perhaps they underperform because they just don’t see the point in doing their best. The researchers decided to test motivation by paying students for their performance. So they pulled 25 math questions off previous PISA exams and they split students into two groups. One group’s participants received $25 and then handed over a buck for every question they missed. Students in the other group got nothing. Here’s what researchers found:
- Students from Shanghai, who ranked first on the 2012 PISA, did just as well whether they were paid or not.
- With the exception of low-ability students, U.S. students did better if they were paid.
- When paid, U.S. students attempted more questions in the second half of the test and were more likely to answer those which they did attempt correctly.
- Researchers predicted that if the U.S. had used financial incentives during the 2012 PISA test, the country’s math ranking would have risen to 19th, from 36th. (And to 32nd if all other countries also paid their students.)
Here’s a graph:
And here’s more about the study if you want the dirty deets.
Steven Levitt, the economist famous for co-writing the Freakonomics books, performed similar experiments in three Chicago schools. Bribery worked there, too. While there was some variation, Levitt and colleagues concluded:
“The magnitude of the impact of the incentives on that day’s test are quite large: approximately 0.12−0.22 standard deviations, which is similar to effect sizes achieved through a one-standard deviation increase in teacher quality or 20% reductions in class size.”
“Overall, we conclude that both financial and non-financial incentives can serve as useful tools to increase student effort and motivation on otherwise low-stakes assessment tests.”
To bribe effectively, Levitt’s research suggests you do the following:
Offer immediate rewards
If students have to wait, bribery doesn’t work. So you won’t be able to bribe students for improved performance on the state test because the results take too long. But you can bribe them on their effort, and the research suggests that you should.
Have established credibility
Levitt had the most success bribing students at the school where he had done previous experiments. Students there believed him when he said they would get money for doing well. He had less success at less familiar schools. Levitt surmised that those students, having never been paid to perform in a school setting, probably didn’t believe he would deliver and so the proffered bribe had little impact on motivation.
Leverage the power of loss aversion
Bribery worked better when students were given the reward at the start and knew they would have to give it back if they failed. So if you really want to be effective (and yes, maybe a little cruel), buy your class donuts before the test, place one on the corner of each desk, and threaten to take it away if you think they aren’t trying their hardest. (Hey, quit looking at me like that. I’m just reporting the science.)
Consider the age of your students
Smaller awards work with smaller kids, but you’ll need better stuff for high schoolers. Cheap little trophies worked just as well with elementary students as did the promise of ten bucks. However, it took a larger dollar amount ($20) to get older kids to give a damn.
You can read the whole study here. But if you would rather not, I understand. And I’m not going to bribe you to do so.
I will, however, attempt to entice you to join my subscriber list. By signing on to the Teacher Habits blog, you will be the first to know about newly released books. You’ll get discounts on those books. You’ll also get new articles emailed directly to your inbox. And you’ll be the first people I ask for advice on book covers and titles. Now aren’t those things better than a trophy?
* I am aware that there are stakes for certain students. Those with third-grade reading laws that require retention (my state of Michigan joined that merry bandwagon last year) and students who have to pay to retake the SAT may have all the motivation they need to try hard.