Can a few text messages, a timely email or a letter increase college enrollment and student achievement? Such “nudges,” designed carefully using behavioral economics, can be effective.
But when do they work – and when not?
Barriers to success
Consider students who have just graduated high school intending to enroll in college. Even among those who have been accepted to college, 15 percent of low-income students do not enroll by the next fall. For the large share who intend to enroll in community colleges, this number can be as high as 40 percent.
There are a number of possible reasons for this attrition: many families overestimate the cost of college because the sticker price of colleges can be much higher than the net price (the sticker price minus the potentially large amount of aid a low-income student could receive); students may struggle with complex financial aid forms; there may be a lack of support to guide them through the application process. So, even when low-income students who are high-achieving do enroll in college, the majority fail to enroll in a college that is comparable to their level of achievement.
Can a few text messages or a timely email overcome these barriers? My research uses behavioral economics to design low-cost, scalable interventions aimed at improving education outcomes. Behavioral economics suggests several important features to make a nudge effective: simplify complex information, make tasks easier to complete and ensure that support is timely.
So, what makes for an effective nudge?
Improving college enrollment
In 2012, researchers Ben Castleman and Lindsay Page sent 10 text messages to nearly 2,000 college-intending students the summer after high school graduation. These messages provided just-in-time reminders on key financial aid, housing and enrollment deadlines from early July to mid August.
Instead of set meetings with counselors, students could reply to messages and receive on-demand support from college guidance counselors to complete key tasks.
The mailer simplified information by providing a list of colleges tailored to each student’s location along with information about net costs, graduation rates, and application deadlines. Moreover, the mailer included easy-to-claim application fee waivers. All these features reduced both the complexity and cost in applying to a wider range of colleges.
In both cases, researchers found that it significantly improved college outcomes. College enrollment went up by 15 percent in the intervention designed to reduce summer melt for community college students. The ECO project increased the likelihood of admission to a selective college by 78 percent.
Getting parents involved
Of course, it’s not just at college enrollment time that nudging can be helpful. Parents also face behavioral barriers while their children are in middle and high school. Many parents underestimate the number of assignments their child has not turned in as well as the number of school days their child has missed. Unfortunately, schools often do a poor job communicating this information to parents in a timely fashion.
I tested an intervention that sent text messages to parents about their child’s missed assignments and grades. The messages were frequent – sent four times more often than report cards – and provided detailed information to parents about their child’s missed assignments and grades. Each message listed page numbers and problems students needed to complete so that parents could track their child’s progress.
Parents responded by communicating with the school more often and motivating their children to do the work: students turned in 25 percent more assignments, which led to significant improvements in grades and evidence of increased math scores.
When there is no impact
While these interventions are promising, there are important caveats.
For instance, our preliminary findings from ongoing research show that information alone may not be enough. We sent emails and letters to more than one hundred thousand college applicants about financial aid and education-related tax benefits. However, we didn’t provide any additional support to help families through the process of claiming these benefits.
In other words, we didn’t provide any support to complete the tasks – no fee waivers, no connection to guidance counselors – just the email and the letter. Without this support to answer questions or help families complete forms to claim the benefits, we found no impact, even when students opened the emails.
More generally, “nudges” often lead to modest impacts and should be considered only a part of the solution. But there’s a dearth of low-cost, scalable interventions in education, and behavioral economics can help.
Identifying the crucial decision points – when applications are due, forms need to be filled out or school choices are made – and supplying the just-in-time support to families is key.