Big Returns for Low-Income Students

When policymakers think of leveling the playing field between students from low-income and high-income communities, they usually focus on support for academics: more teachers, smaller class sizes, better facilities, and maybe even after-school tutoring. But what if a student can’t make it to class because their parent is sick and can’t drive them? Or what if they can get to school but can’t concentrate because they’re hungry or worried about that sick family member at home?

Image

Jamie Gracie, PhD ’25

“There’s great evidence that having a better teacher or being in a smaller class affect not only test scores, but also the likelihood of attending college, and earnings as an adult,” says economist Jamie Gracie, PhD ’25. “But if you have a higher rate of absenteeism, which is the case for low-income students, then that limits what you can learn from even the best teacher. It’s the same if you can’t focus. By solving some of these upstream issues, we might even get a bigger return from more resource-based interventions.”

Gracie, a postdoctoral fellow at Harvard’s EdRedesign Lab and an affiliate of the nonprofit policy institute Opportunity Insights, and Cornell University Professor Benny Goldman, PhD ’24, wanted to find out how much addressing these upstream problems would impact the success of low-income students. Using big data and machine learning, the two examined the effect that “personal supports” had on measures including test scores, high school graduation rates, and earnings in young adulthood. The results of their study, published in a working paper on the Social Science Research Network website in December 2025, show that investments in non-academic “navigators” can yield big returns for economically disadvantaged kids.

The Power of Personal Support

As psychologists and pediatricians will attest, there are major differences in the readiness skills of kindergarten students from low- and high-income backgrounds. Gracie says it shows up not only in measures of pre-literacy and pre-numeracy, but also in non-cognitive skills necessary for success in school, like emotional regulation, attention, and executive function. But her faculty advisor, William A. Ackman Professor of Economics Raj Chetty, PhD ’03, co-founder of Opportunity Insights, says the disparities do not reflect innate ability.

“There are large gaps in children's skills, even at the point of kindergarten entry, on many dimensions,” he says. “At earlier ages, gaps are much smaller, and they expand rapidly even from birth to kindergarten entry because of differences in the environments in which children are raised and whom they are exposed to, both within and outside the home.”

The root causes are complex and include everything from the impact of nutrition on brain development to exposure to chronic stress, to how many books are in a child’s house and whether someone reads to them regularly. “For any given student, it might be more of one thing than another, but they all come together to produce big average differences by the time these kids are five,” Gracie says.

Enter the navigator. Often a social worker, often with professional experience in school systems, navigators advocate for students, connecting them with services and supports, solving problems that lie upstream from academics. “A navigator is someone who knows the system,” Gracie says. “It’s a separate person who meets with students to identify their needs, freeing up teachers or school counselors to do their actual job. Students spend about 80 percent of their lives outside the classroom. Navigators help get them to show up ready to learn.”

Navigators connect students with a wide range of services and supports. Some are related to cognitive skills, such as after-school homework help and tutoring programs. Some focus on socio-emotional regulation, including mental health services, addressing bullying, or small-group mentoring. Some supports target basic needs outside of school entirely, including nutrition assistance. And sometimes, navigators intervene directly on behalf of a student, advocating for small changes that can make a big difference.

There are large gaps in children's skills, even at the point of kindergarten entry, on many dimensions. At earlier ages, gaps are much smaller, and they expand rapidly even from birth to kindergarten entry because of differences in the environments in which children are raised and whom they are exposed to, both within and outside the home.
—Professor Raj Chetty, PhD ’03

“In our research, we heard of one student who was doing quite well,” Gracie explains. “She was maybe on the path to go to college. Then there was a bad shock to her parents and she stopped being able to reliably get to school in the morning, so she was missing a lot of her first-period class, the only one she needed to pass to graduate. She was failing because of her attendance, even though she was a very bright student. Her navigator was able to talk to her counselor and just switch the order of the classes, so that something that was less important for graduation could be first. That gave her time to get there and still make it to the class she needed.”

Navigating to Higher Test Scores

Anecdotes are compelling, but Gracie and Goldman wanted to more precisely measure the impact navigators have on academic outcomes. “The key question is how to create lasting improvements in low-income students’ educational attainment and economic independence—not just short-term gains in school that fade by adulthood,” Goldman says.

With support from Harvard’s EdRedesign Lab, the two connected with the national nonprofit, Communities in Schools (CIS). “CIS serves about two million students per year,” Gracie says. “It’s three times bigger than Head Start. They’ve also been around for nearly 50 years, so we could track outcomes over time.”

 Through their local affiliates, CIS places navigators in schools across the US. With the organization’s data in hand, Gracie and Goldman focused on schools in the state of Texas, identifying students most likely to work with a navigator—often, those involved in a school’s case management process. Controlling for demographics and a wide range of other factors, the researchers compared students in schools with a CIS program to similarly struggling students in schools without one. 

“We used simple machine learning techniques to say, ‘Based on everything I see about you in elementary school, your test scores, your attendance every six weeks, your disciplinary infractions, things like that, how likely are you to be case managed or working closely with a navigator in middle school?’” Gracie explains. “Once we found those kids, we took every school that got a CIS program and we found a different school in Texas that didn't have one, but looked similar in the years leading up to when the program started. We show that in the five years before CIS begins, students in these two sets of schools are on similar trajectories. Outcomes are moving together. And then we look at what happens after the CIS program begins.” 

For struggling students, the data showed that a navigator made a significant difference in academic performance. “Their test scores start to improve immediately, relative to other students who were similarly struggling in the schools that didn't have CIS,” Gracie says. “By the time the program has been going for a couple years, we find that test scores for these students have gone up by 10 percent of a standard deviation. That’s evidence that, when the program begins, outcomes improve exactly for the set of students most likely to be working with a navigator.”

Moreover, Gracie and Goldman found that the benefits of having access to a navigator continued well into adult life. “Students receiving three years of CIS during middle school are 5.2 percent more likely to graduate high school, 9.1 percent more likely to enroll in two-year college, and earn about 4.3 percent more annually at age 27,” the study’s nontechnical research summary reported. “Three years of CIS—an upfront cost to CIS of about $3,000 per student—is estimated to increase a student’s lifetime earnings by $36,000 in present-day value.”

A navigator is someone who knows the system. It’s a separate person who meets with students to identify their needs, freeing up teachers or school counselors to do their actual job. Students spend about 80 percent of their lives outside the classroom. Navigators help get them to show up ready to learn.
—Jamie Gracie, PhD ’25

“These findings suggest that when a child receives CIS and earns more as an adult, the government benefits through higher tax revenue,” Goldman notes. “That means expanding programs like this can be a cost-effective way to improve long-term outcomes for low-income students.”

Chetty says that Gracie and Goldman’s research adds to a growing body of evidence showing that well‑designed interventions can generate returns far exceeding their upfront costs. “We see similar results for other interventions that pair financial investment with stronger social capital and support systems, such as recent research from Opportunity Insights on the HOPE VI neighborhood revitalization program,” he says. “A modest investment can yield tens of thousands of dollars in additional lifetime earnings.”

Image

Jamie Gracie and co-author Benny Goldman presenting.

Idiosyncratic or Systematic? 

Gracie says that the next step in understanding the impact of navigators is to explore their synergistic effects with other improvements. “I’m very interested in directly looking at the complementarities between the resource component and personal supports,” she says. “Is the combination of a navigator and a better teacher even greater than the sum of those parts?”

The researcher also hopes to drill down past the averages to see where and why navigators made a bigger difference. “I think there probably is a lot of heterogeneity across sites and schools,” she says. “You would imagine that some places were more successful than others. Some of that can be purely idiosyncratic, but it might also be more systematic. Maybe some schools and navigators are better connected to the network of local nonprofits, or government services are more readily available. We couldn’t speak to that in this paper, but it’s important for thinking as a policymaker.”

Along those lines, Gracie and Goldman hope their work can help policymakers identify some of the keys to closing the gap between students from low- and high-income backgrounds. They see their research as part of the effort to reshape the entire mindset around schools.

These findings suggest that when a child receives CIS and earns more as an adult, the government benefits through higher tax revenue. That means expanding programs like this can be a cost-effective way to improve long-term outcomes for low-income students.
—Ben Goldman, PhD ’24

"CIS is not the only type of program that does work like this," Gracie says. "In New York, California, Michigan, and beyond, there have been big expansions of full-service community schools, which are similar in nature. They might involve different measures, like bringing in health clinics onto campus. But they share the spirit of ensuring that the out-of-school issues students face are addressed so that they can succeed in the classroom and beyond. I think up to this point, there hadn't been a lot of empirical evidence on the effectiveness of these programs. We see our research as one step in that direction."

Banner image: Communities in Schools