Reclaiming the Nudge

If you’re looking for a textbook example of the hype cycle, look no further than the humble nudge.

The idea of nudges -- small interventions that impact behavior -- originated in behavioral economics, where researchers found that very small, subtle interventions could increase the likelihood that someone would take a desired action or make a positive choice. In just a few years, nudging has gone from inflated expectations to trough of disillusionment. It seems that every week a new article pops up claiming that nudging doesn’t work, but is that claim itself just hype?

After three years of research and experimentation, we at Duke’s Teaching and Learning Innovation Lab are inclined to say yes. Our research suggests that the failure is not in the nudge itself, but in mistaken use of the term “nudge” to refer to virtually any intervention designed to create incremental behavior change. In fact, when designed appropriately, nudging still shows great potential to improve student outcomes and support positive decision pathways.

In 2008, the University of Chicago economist Richard H. Thaler and Harvard Law School professor Cass R. Sunstein published the book Nudge: Improving Decisions About Health, Wealth, and Happiness, and seemingly overnight, nudging was everywhere: restaurants found that people recycle more when the trash bin was labeled “landfill,” governments saw their back tax payments increase when they added a handwritten note on the collection notice and schools found that students bought more healthy foods when those options were placed at eye level along a cafeteria line.

Higher education took note, launching a flood of experiments to see if nudging could be an answer to persistent challenges large and small. Such nudge-based interventions included attempts to increase the number of underrepresented minority applicants to selective universities, reminding students to fill out financial aid forms and encouraging students at risk of failing to seek tutoring help. Yet in spite of a vast body of literature suggesting that these interventions should work, many of them didn’t.

The impact of nudging has thus far been mixed. Both the FAFSA nudge and the nudge aimed at increasing college applications from low-income qualified students found no significant effects when replicated at scale. Some nudges have even backfired: a nudge intended to motivate students to study more by showing them how they compared to their peers actually caused low-performing students to study less and even drop out.

However, some nudges have shown promise. Georgia State University found that offering customized messages, instead of form letters, reminding students of pre-matriculation tasks reduced the number of enrolled students who never showed up on campus by 21 percent. Several other universities found that slightly reducing the size of dorm rooms while offering expanded common spaces increased student retention and collaboration.

So why do some nudges work while others do not? Quite simply, many of these failed nudges aren’t nudges at all; they’re nags. And no one likes being nagged.

A nudge, as opposed to a nag, leverages cognitive heuristics to prompt slight behavior changes without people consciously reacting to the nudge. In other words, a nudge isn’t something you realize you’re getting, and the result isn’t something you think too much about. A nudge happens in the background of your daily life, and it works best when the goal is to slightly increase a positive outcome.

For example, you can nudge people to recycle their empty can instead of throwing it in the trash by labeling the trash can “landfill,” but getting a reminder to do an annoying, time-consuming task like filling out a FAFSA is more likely to cause annoyance than to motivate action.

Duke University’s Teaching and Learning Innovation Lab recently completed a pilot test of a nudge designed to help students remember more of the material they learn in lecture classes by sending them text messages with short multiple-choice questions. By prompting students to actively recall a small bit of information from a lecture, it shifts the “forgetting curve” and helps students remember more, longer. Our evaluation found that students who received these nudges had slightly higher final course grades (1.43 percentage points, significant at p<0.05) than students who did not.

So what is the future of nudging in higher education? We think it is important to be realistic about what nudges can and can’t do. This is not to say that researchers shouldn’t continue exploring the impacts of reminders and, yes, even nags.

There are likely some student behaviors that can be motivated by persistent reminders, but we should be careful not to call these nudges. Nudging should be considered when the goal is to create small, incremental changes in student choices, not effect large behavior changes.

Finally, we hope to see more collaborative, cross-university research on nudges. By working together and sharing our research, we can all advance our understanding of how to best use nudges to help students succeed.