The high value of virtual college advising for lower-income students

Bettinger’s research speaks to a longstanding challenge in education: high school students aiming too low for post-graduation opportunities — or “undermatching,” as education researchers describe it.

“[Stanford researcher] Caroline Hoxby’s research shows there’s a sizable number of students who are low- to moderate-income and qualified to go to Ivy League schools or even just to the elite schools in their state,” Bettinger says. “But because they don’t have good coaching, they are more likely to go to colleges with fewer resources, schools where completion rates are lower on average, than their high-income peers.”

A significant part of the problem is access, as underserved students are less likely to have resources to navigate the increasingly complex college admissions and financial aid processes. The goal for many organizations is simply persuading under-resourced students to apply to any college. But the problem of focus for Bettinger was getting high-achieving students from this population to consider the best schools for them — those that can position them for greater opportunity and success.

To take up the challenge, Bettinger and his collaborators partnered with the College Board to identify students at risk for “bad decisions,” as he put it. Specifically, the College Board relied on students’ responses to surveys (such as those filled out along with the SAT) to target a group who had high test scores but were in situations where “they may not have had as much success in the college admissions game” — such as coming from lower-income households.

“These were top students who could have many options for college,” Bettinger says. “They often did not realize the many options available, and the advising was designed to help them see the full array of possibilities.”

The researchers focused on virtual advising rather than the in-person variety, because a virtual adviser can serve students across a wider region — the whole world, in theory — than an in-school counselor. What’s more, other researchers like Hoxby have shown that high-achieving students who are most in need of counseling services tend to live in more geographically distant areas.

Gaining student buy-in

For the study, the College Board identified more than 16,000 high-achieving students from low- to moderate-income households, based on their standardized test scores. Three-fourths of the students were offered virtual advising through College Advising Corps; the remaining quarter formed the control group.

One of the researchers’ first hurdles was getting students to buy in to the process. “If you think about the mail and emails students get, a lot of it includes promises from people who are hoping to help them get into school,” Bettinger says. “How do students sift between the legitimate ones and the ones that just want money?”

The researchers developed carefully crafted communications delivered in two recruitment campaigns using mail, email, phone calls, and texts. Ultimately, 44% of the targeted students engaged with their virtual advisers, whose focus was on college selection and application support.

The study compared the college admissions outcomes of students who received the virtual advising with the outcomes of those who did not.

The power of virtual advising

The intervention made a measurable difference.

Specifically, while there was no impact on the overall rate of college enrollment, students who had participated in the advising were more likely to enroll in higher-quality schools, such as those with higher graduation rates: The advised group showed a 6.1% increase in likelihood of enrolling in better schools versus peers who received no counseling.

Importantly, the research also tested the benefits of matching advisers with students on demographic variables, including race and gender. “The belief among some advising groups, including College Advising Corps, has been in a ’near-peer’ model — that if they get advisers who look like the students they are advising, they are going to have more success,” Bettinger says. “The underlying question is, ‘How can somebody understand what you’re going through and help you if they have nothing in common with you?’” — an issue of particular relevance for underrepresented students.

The findings revealed no impact of matching based on gender, but race played a clear role: Nonwhite students randomly assigned to a nonwhite adviser of the same ethnicity were about 8% more likely to attend a higher-quality school than nonwhite students paired with a white adviser.

The finding of match-based benefits has implications for advising program design. “It helps us answer the question of which adviser to put in front of which students,” Bettinger says. For example, for its residential [in-school] advising program, College Advising Corps even tries to match students with advisers who’ve attended the same high school. Bettinger plans to study the impact of such design choices in future work.

The research findings, published in the April 2020 issue of the Economics of Education Review, are especially relevant in the context of the COVID-19 pandemic, when many education-focused interactions have shifted online. In fact, College Advising Corps discovered an unexpected benefit of the team’s work, as Bettinger notes: “With COVID, people who’d been trained to advise in person now had to do it virtually. Because of our research, they’d already figured out the pitfalls of advising virtually, such as how to deal with students who don’t want to interact. So they knew how to move forward in the COVID world and serve students well.”

In the end, “It’s about helping kids get into the right schools for them,” Bettinger says. “And in many cases, the kids who engage the most in the virtual world already have a strong commitment to go to school — it’s not whether, but where.”