Why we can’t always be “nudged” into changing our behavior

Are we more likely to click on the first result on Google than the second?

Are we more likely to eat a big meal if we use a big bowl?

Are we more likely to apply to a top college if we get a personalized admissions packet?

All of these questions have been explored in the research literature on behavioral “nudges,” or methods for slightly changing the environment to change people’s behavior.

The term was popularized in a 2008 book by University of Chicago economist Richard Thaler and Harvard Law School professor Cass Sunstein, Nudge: Improving Decisions about Health, Weight, and Happiness. Nudges became particularly popular in nutrition — experts are eager to find easy ways to change people’s eating habits — and in education, where researchers are casting a wide net for cheap ways to improve outcomes for students.

Unfortunately, it’s really hard to change things in those two areas — at least that’s my takeaway from a bunch of disappointing “nudging” results in the past few years.

Early research in nutrition and education suggested that humans are very suggestible. Packaging sizes, plate sizes, location on a buffet table, and other small things affect what we eat; sending a $6-per-student information packet to high-achieving low-income students substantially increased the number who wound up enrolling in top colleges.

But last year, we learned that if things sound too good to be true, they probably are. Much of the “nudge” research on nutrition came from Brian Wansink, a former Cornell researcher who had 15 studies retracted after he was found to have engaged in academic misconduct (and after other researchers couldn’t get the same results).

While there are no allegations of academic misconduct in studies evaluating the effectiveness of nudges for educational interventions, those efforts have ultimately been disappointing too. A larger-scale attempt at replicating the information packet intervention found that it had no effects on getting low-income students into top colleges. “Sometimes it takes more than a nudge,” the research group MDRC concluded.

Another study sent text and email reminders to 700,000 high school seniors and incoming college students encouraging them to apply for financial aid. The hope was that the reminders would get more students to fill out aid applications. It didn’t work.

The candid, if disappointing, summary of their results: “no impacts on financial aid receipt or college enrollment overall or for any student subgroups. We find no evidence that different approaches to message framing, delivery, or timing, or access to one-on-one advising affected campaign efficacy.”

“It didn’t seem to matter how we framed the message or how we sent the message; we weren’t finding differences between them,” one of the study’s authors said.

A different study tried “nudging” students to study more by giving them accurate estimates of how much harder they’d need to work for their desired grades in the class. The effort didn’t make the students work harder; it just made them accurately expect lower grades. None of the interventions they studied produced any significant academic benefits — not for at-risk students or for the college population as a whole.

As a recent college graduate with mediocre grades, that didn’t surprise me at all. Students might not have had access to the accurate estimates, but they already knew that studying more would mean they got better grades. No one at college is going to be surprised by this information. Similarly, it’s not surprising that information packets alone aren’t enough to get students to make a decision about a topic as fraught and complex as where to attend college, or that text message reminders aren’t enough to get them to apply for financial aid.

But is the right takeaway that nudges don’t work at all? Probably not. The very first result I mentioned — that people are more likely to click on the first Google result than the second — is absolutely true. People also buy things at eye level in grocery stores more often than things that are harder to see. And maybe some of the education interventions that have shown promising results will replicate, even if most don’t.

But we should expect modest effect sizes, and smaller effects on any goal that’s already highly valued and that people already have lots of reason to have thought about and worked on. Frustratingly, nudging might have the smallest effects on things we care about the most.


Sign up for the Future Perfect newsletter and we’ll send you a roundup of ideas and solutions for tackling the world’s biggest challenges — and how to get better at doing good.

Future Perfect is funded in part by individual contributions, grants, and sponsorships. Learn more here.