Sunday, February 15, 2015

Rewards and Motivation: What Does the Research REALLY Say?

Dissecting Research Icon; photo source Dissecting Apple by Andres Nieto Porres (Creative Commons Share Alike)
Often it seems like you can find research to support virtually any argument you want to make concerning education policy and practice. While this is the case in many fields, education research is rife with conflicting research findings on a wide range of topics. However, once you go beyond the blog headlines and the abstracts of research papers, you find that there really are often consistent answers to be found. Unfortunately, just like the students we teach, those answers are often nuanced and incompatible with pronouncements of "the one right way."

If you are a teacher, you almost certainly have been exposed to the decades-long debate about the proper use of rewards and incentives in the classroom. Do rewards help or hurt? Does the use of extrinsic motivation harm intrinsic motivation? Today, we take a dive into the research on rewards and motivation. It's Edward Deci and Richard Ryan versus Judy Cameron and David Pierce; Alfie Kohn versus Daniel Willingham; PBIS versus ... well, not-PBIS. Read on as I try to dissect what the research really says, and its implications for educational practice.

SPOILER ALERT: The research on rewards is not as clear cut as either side makes it out to be. But, if you use rewards in your school, there's a really good chance you're using them poorly.

The current debate over the use of rewards largely dates back to the 1970s and Edward Deci's research that found that, in general, the use of rewards (extrinsic motivation) created significant and lasting damage to people's internal (intrinsic) motivation. While Deci's work gained the most notoriety, there were certainly studies that contradicted his results. In 1994 Judy Cameron and David Pierce published a meta-analysis [subscription required for access] of much of this research and came to the conclusion that, in general, rewards were not detrimental to intrinsic motivation.

Since then, we have seen a series of volleys back and forth between the two camps, with articles such as: "When Paradigms Clash: Comments on Cameron and Pierce's Claim that Rewards do not Undermine Intrinsic Motivation [pdf]," "Effects of Reward on Intrinsic Motivation- Negative, Neutral, and Positive: Comment on Deci, Koestner, and Ryan [pdf]," "Pervasive Negative Effects of Rewards on Intrinsic Motivation: The Myth Continues [pdf]," and "Extrinsic Rewards and Intrinsic Motivation in Education: Reconsidered Once Again [pdf that includes further comments and counter comments]." In a nutshell, both sides are looking at largely the same set of research data, synthesizing it in somewhat different ways, drawing different conclusions, and then attacking the other side's methodology.

That's a very small snippet of the debate going on among researchers and theorists. What is shocking, however, is that when you read the details of their findings, there really is considerable common ground. The problem is, just like political parties in America today, both sides have been radicalized in our public discourse. The result is that very often, the nuances are being lost and teachers believe that students are being completely "Punished by Rewards" or that "careful arrangement of rewards enhances motivation, performance, and interest."

The real findings of this research are not nearly as controversial. Let's take a look first at what Judy Cameron and crew actually found in their 2001 paper "Pervasive Negative Effects of Rewards on Intrinsic Motivation: The Myth Continues." Again, this is a paper that clearly is meant to be arguing that the negative effect on intrinsic motivation is a myth. However, here are the effects on free choice intrinsic motivation that the analysis actually found [click to enlarge]:

Effects of Rewards on Free Choice Intrinsic Motivation
Reconstruction of graphic presented by Cameron et al. (2001) with color and layout modified for clarity.
So, here we have one of the leading analyses in support of rewards and it only finds that they have a positive effect on intrinsic motivation when the student has little interest in the task, when the rewards are verbal, or when a tangible, expected reward is offered for doing a task better than their peers. More significantly, she still finds negative impacts on intrinsic motivation when tangible rewards are promised in advance for working on a task, doing a task well, or for completing units of a task.

Are we ready to see just how different the results turn out under Deci's analysis? Here we go [click to enlarge]:
Effects of Rewards on Free Choice Intrinsic Motivation as reported by Edward Deci (1999)

And, for those that want to get even wonkier, compare Deci's effect table (left) with Cameron's rebuttal (right), both of which can be found here [pdf] [click to enlarge].

So, in the end, the research really does give teachers guidance in using rewards in the classroom. Here's what you need to know:

  1. Rewards can help individual students who struggle with interest in specific areas. As they begin to gain competence and interest in the task, rewards can be gradually phased out.
  2. Positive verbal rewards (provided they are viewed as "informational" rather than "controlling" by the student- a good topic for a future post) can increase intrinsic motivation (though the effect is smaller in younger students).
  3. Providing students with tangible rewards that were not expected by the student have no effect on intrinsic motivation.
  4. Extreme care needs to be taken when it comes to using tangible rewards that students expect to receive. Deci's research finds no ways to structure these rewards that do not undermine intrinsic motivation. Cameron's research finds that if carefully structured, it is possible to employ rewards that do not have a significantly negative impact.
  5. If you are a football coach and you tell Jones he can take Smith's starting wide receiver spot if he can beat Smith in the 100 meter, then Jones' intrinsic motivation may increase according to Cameron. While interesting, if you can think of ethical ways to use this finding in the elementary school classroom, please leave me a comment to let me know.
I will be the first to admit that there are many more issues at stake in the rewards debate, and that the research community as a whole lacks good studies on the longevity of these impacts to intrinsic motivation. However, the differences between these two camps are greatly exaggerated.

The bottom line for me is that a good rewards system needs to be carefully thought out and tailored to individual students. Classroom-wide and school-wide tangible rewards programs have a real potential to harm intrinsic motivation and should be avoided. For most students, teachers should focus on positive verbal rewards. They should carefully select an individualized reward system only for students who need it, and gradually fade these rewards over time. 

No comments:

Post a Comment