Nudges Are Not Going to Transform Education
For over two decades, the field of behavioral economics has been exploring the efficacy of “nudges,” suggestions meant to subtly direct behavior in specific, presumably beneficial ways.
For example, the image of a fly in a urinal at Amsterdam’s Schipol Airport was found to reduce the clean-up costs at the urinals by 80% by giving men something to aim at.
Nudge theory was broadly popularized in Nudge: Improving Decisions About Health, Wealth, and Happiness by Richard Thaler, a Nobel Prize winning economist, and legal scholar Cass Sunstein. Thaler and Sunstein run down the ways nudges can be used to advance public policy goals, such as getting more people to vote, increasing organ donations, or saving money from their paychecks.
Education was an obvious target. What if nudges could get students to study more, or induce less likely candidates to apply for college, or pursue scholarship money? If nudging worked, we could prod entire cohorts of students to greater heights than their predecessors.
Unfortunately, recent comprehensive meta studies of nudge research have revealed that on balance, nudging doesn’t do all that much to change behaviors, and often the cost of nudging can outweigh the benefits.
Initially there was some dispute about what the meta-analysis showed in terms of the efficacy of nudging, but subsequent reviews found that the “publication bias” (studies that show a positive effect of nudging are more likely to be published than those that show no results), and small “effect sizes” (how much nudging actually achieved), suggest that nudging is no magic solution able to cure all manner of ills.
When it comes to education, much has been tried, but little has worked. It appears that in general, nudging works best with small, discrete events, like applying for scholarships, and less well with changing behavior long term. Temporary increases in certain positive behaviors tend to disappear when the nudge is removed. There’s also some evidence that nudging may increase anxiety and cause students to disengage from their work.
I don’t know about anyone else, but I feel as though I was subjected to nudge theory long before it found a home in behavioral economics, though rather than nudge theory, I would call it “nag theory,” with the chief practitioner being my mother.
Don’t get me wrong, the things my mom would get on my case about (mowing the lawn, feeding the dog, not failing math), were absolutely legit, but there was little about the nagging (I mean nudging) that actually improved overall child performance. Sure, eventually stuff would get done, but no one was really happy about the process.
When nudges become nagging, the efficacy goes down, and the costs go up. Nudging students to do their homework (say via a text) may get temporary results, but it may also prevent students from developing the kind of skills of self-regulation that will ultimately be necessary for long-term school success.
The other thing we know about nudges is that it is much easier to nudge people towards something they desire, in which case the nudge is more like encouragement, than to nag them about something they don’t want to do. As the Schipol Airport example shows, there’s no inherent desire for men to make a mess at the urinal, but it helps to have a reminder.
Similarly, in education, nudges that help connect students to past successes may ultimately have the greatest impact on long-term behavior. A study that had students writing about their core values in an act of reflective self-affirmation showed some promise in helping more students ultimately apply for college. In this case, students were being prompted to think about their goals and desires, an activity which may have kept that long-term thinking in their minds as they went about their schooling.
So it’s not that nudges never work, or that all nudging can be judged on the same criteria.
One thing the findings on the impact of nudging on education reinforce for me is the fundamental mistake we make in creating this monolithic group we call “students” and then assuming they are all the same. The research on nudging seems to continually disappoint because there is no universal response to the same stimulus. Nudging theory assumes people are largely the same, but this simply isn’t true.
A nudge that works well for one individual may be a turn off to another. The reason effect sizes for some of these studies might be small is because the nudge is working quite well for some, but is having a negative impact on others.
Education is a human enterprise, and when a nudge treats students like cogs with the same interests, same motivations, and same personalities, whether or not something works will look like a stab in the dark.
Rather than spending all this effort at nudging (and even shoving) students along the same paths, maybe we could spend more energy making the landscape conducive to individual exploration, where no nudging is required because the students are eager and well-equipped for the journey.