David Dockterman has one of the coolest jobs at Scholastic (IMHO), and one of the coolest titles. As the “Chief Architect of Learning Sciences” for Scholastic Education, he designs games, apps and other technologies that help kids achieve learning goals — like becoming fluent in math facts, for example. Which means he’s an expert in motivation and the underlying science of games and in the concept of “gamification.” I’ve learned a lot from him. (Here he is over on Mashable writing about educational games.)

The other day I had this fascinating e-mail exchange with Dock (we call him “Dock”) about some new research that you might have read about in the news that showed students getting good scores on tests when they were offered cash rewards.

Here’s part of our conversation about it. I hope you’ll find it as interesting as I did.

DD:

Here’s some new research to make things more interesting: “Immediate Rewards for Good Scores Can Boost Student Performance.”

Tyler:

Hmm… So extrinsic rewards can help boost performance on tests. Does that actually motivate them to learn, or motivate them to just do what it takes to earn a reward (ie. become a good test-taker)?

DD:

This new research is all part of an evolving, complex picture. Earlier research suggests that extrinsic rewards will undermine performance and crowd out intrinsic motivation on tasks that people are already motivated to do. This study targeted low-achieving, disengaged students, a population lacking intrinsic motivation. The behavioral economics research, as summarized in the full report on this research, says: “However, on tasks where intrinsic motivation is already low or zero, external rewards are less likely to have such negative long-term effects.” At least that seems to be the case when the external rewards are seen as substantial, in this case $20 vs $10. What about the $10 reward” “But in line with previous work, we do find some evidence that low level rewards can crowd out intrinsic motivation.”

Two other key findings:

1)      Rewards must be immediate. If the money was promised to be delivered a month after the test, the benefit of the incentive nearly disappeared;

2)      Rewards are more powerful when presented as losses. Here’s $20 or a trophy, you lose it if you don’t perform well on the test. Other research has shown humans to be loss averse, we are more driven to avoid loss than win a gain. That seemed true here as well.

What’s most important, though, is that kids do better when they try. And kids who don’t feel that tests matter to their future don’t try. Consequently, their poor performance on tests can underestimate their actual ability. The gain in performance in the study doesn’t come because the students magically had more knowledge. They just tried harder. External incentives, appropriately employed with the right audience is one way to invigorate effort, but it’s not the only way. The research on stereotype threat and fixed vs growth mindsets suggests that students can also be primed to see themselves differently, including as academic achievers. There are some fun and interesting studies in this realm too. Lots to sort through.

Tyler:

You said: “What’s most important, though, is that kids do better when they try. And kids who don’t feel that tests matter to their future don’t try. Consequently, their poor performance on tests can underestimate their actual ability.”

People might say this “seems obvious,” but I do think it’s an important thing to remember! According to our Primary Sources teacher survey, only 45 percent of teachers (K-12) say their students take standardized tests seriously and perform to the best of their abilities. This is a really big deal when tests are so high stakes (as they are in education today)! Putting aside the debate about whether standardized tests are good or bad, if schools/gov/educators are going to place so much emphasis on tests, shouldn’t they think hard (and talk to these researchers) about how to motivate students in a way that gets them to perform to the best of their abilities?

DD:

You might find interesting a study that Roland Fryer did a year or two back that showed that paying students for results, for being academically successful, did not work. However, providing financial incentives for the behaviors that underlie academic success (like reading books and timely completion of assignments) did show promise.

On the other hand, psycho-social interventions, those that focus on priming a student’s identity, have also shown incredible promise in motivating student achievement.

These interventions target the caring part, giving students a sense of their own value and the value of doing well in school.

If I had to pick a path to pursue for further investigation, I’d choose caring over paying.