Under the title “Should Kids Be Bribed to Do Well in School?” Amanda Ripley of Time Magazine reported about the outcomes of the large-scale study led by Roland Fryer Jr. that tested whether incentive systems affected students’ achievement. Professor Fryer, who collaborated with many others on this ambitious project (> 270 schools), found that rewards for outcomes (e.g., grades and test scores) were less effective than rewards for what he calls “educational production functions” (activities, such as reading and participating, that led to better learning).
Here is how Ms. Ripley characterizes the outcomes:
Kids may respond better to rewards for specific actions because there is less risk of failure. They can control their attendance; they cannot necessarily control their test scores. The key, then, may be to teach kids to control more overall — to encourage them to act as if they can indeed control everything, and reward that effort above and beyond the actual outcome.
To my way of thinking, this finding fits well with a behavioral view of “incentives” and conflicts with a popular, lay, view. The lay view, as I would characterize it, is that one dangles carrots before students, motivating them to try hard and, thus, achieve more. A behavioral view, as I see it, is that one reinforces the behaviors that lead to achievement, thus teaching students how to succeed. Please note the change in terminology. Although lay language equates rewards with reinforcement, informed speakers differentiate between ‘reward’ (something given in recognition of accomplishment) and ‘reinforcement’ (something that occurs contingent on a behavior).
Mayhaps I’m splitting hairs, but I think the term ‘rewards’ promotes talk about bribery (see the headline for Ms. Ripley’s article), intrinsic motivation, and the like. The latter takes us in the direction of dispassionately examining relationships between behaviors and their consequences.
Link to Ms. Ripley’s story. Download a copy of Professor Fryer’s actual report: Financial Incentives and Student Achievement:
Evidence from Randomized Trials. Read an earlier post from Behavior Mod Info about this project.
Hi John,
I agree with your points but I might split hairs even a little further just for chatting purposes – a reward can also be given contingent on a behavior, but it’s the influence on the future probability on rate, intensity or magnitude, latency, duration, etc. that determines whether it is functioning as a reinforcer. Bribery to me is in a whole other ball game – not only being the offer, but the delivery of something in advance of a behavior in the hope that it will happen much as you offer Junior a sucker in the hopes that he will stop screaming in Target.
Og Lindsley recognized that we don’t know the impact until that future event, and it was also part of moving to Plain English from technical jargon.
Lindsley, O. R. (1983). Say reward, relief, punishment, or penalty. Journal of Precision Teaching, 3(4), 100-101.
Lindsley, O.R. (1991). From technical jargon to plain English for application. Journal of Applied Behavior Analysis, 24(3), 449–458.
I find it kind of funny that people expect children who are susceptible to all kinds of competing reinforcing and punishing contingencies (whether recognized or not) to perform for the “love of learning” vs. some kind of targeted tangible consequence, and yet we also train the value of a “report card”, letter grade, or GPA which to me is a deferred consequence of some abstraction, and at least as artificial as the money which is being criticized. In the article itself, I found the children’s comments (and the children’s opinion matters to me as much as any post hoc write up) interesting…they’re kids but they know what matters to them.
Just thinking out loud.