A program whose coordinators connect struggling students with academic and social services quickly improves test scores, attendance and disciplinary issues, and later college enrollment and earnings, finds new research co-authored by a Cornell economist.

Acting as navigators and coaches in high-poverty schools, site coordinators from the nonprofit Communities in Schools (CIS) help students access resources both within and outside of school, including tutoring, food aid, housing and health services. CIS is the nation’s largest program of its kind, serving nearly 2 million students in more than 3,000 schools – nearly three times bigger than Head Start.

Analyzing outcomes in Texas, where more than half of CIS schools are located, the researchers determined that the program’s personalized approach, including weekly meetings with students and action plans tailored to their needs, delivered short- and long-term benefits at relatively low cost compared with other well-regarded reform strategies, such as reducing class sizes. As a result, they said, there’s room for the program to expand significantly.

The researchers said the findings add to growing evidence that pairing navigators with traditional support systems can amplify their impact.

“The results are very promising – struggling students saw real improvements across the board,” said Benjamin Goldman, assistant professor in the Cornell Jeb E. Brooks School of Public Policy and the Department of Economics. “It makes sense: When families face multiple, overlapping challenges, accessing available resources isn’t always straightforward. Our results show that pairing students with a dedicated navigator and mentor – an approach that has proven effective for adults in other settings – can meaningfully improve outcomes in high-poverty schools.”

Goldman is the co-author with Jamie Gracie, a postdoctoral fellow at the Harvard University Graduate School of Education’s EdRedesign Lab, of “When Resources Meet Relationships: The Returns to Personalized Supports for Low-Income Students,” a working paper released Dec. 15.

Drawing upon Texas education records and federal census and tax records, the research focused on middle schools (grades 6-8) where CIS programs were introduced between 1998 and 2016, comparing student performance before and after the programs began, compared to a control group of similar students in non-CIS schools. Highlights include:

• Immediate improvement in math and reading test scores, particularly for students who were struggling academically.
• Students participating in CIS for three years of middle school are 5.2% more likely to graduate high school, 9.1% more likely to enroll in a two-year college, and earn about 4.3% more annually at age 27.
• Students with different needs receive different services – some more academic, some more social and behavioral – but long-term outcomes are nearly identical across these different types of students.
• Return on investment compares favorably to other education reforms. The $3,000 cost to provide three years of CIS support increases a student’s lifetime earnings by an estimated $36,000 in present-day value. The estimated annual impact on earnings is about 10 times that of class-size reductions. The earnings analysis was based on schools outside Texas where CIS had been operating longer.

Importantly, the researchers found that short-term gains in test scores, attendance and behavior – but not test scores alone – accurately predicted longer-term outcomes in graduation. Social scientists increasingly seek these types of “statistical surrogates” to gain earlier insight into programs’ effectiveness without waiting decades to measure longer-run effects.

“When thinking about economic opportunity targeted at kids, we want to know if an intervention is going to make them better off economically in the long run, but ideally we don’t want to wait 15 years to figure that out,” Goldman said. “We show that the immediate effects in the same year CIS enters schools can be used to quite accurately forecast the future effect on high school graduation.”

Goldman and Gracie said CIS’ reach in Texas could be doubled to achieve the maximum possible impact on graduation rates, focusing on the highest-need schools not yet served. Opportunities in other states might vary based on the specific resources each state provides, Goldman said.

More broadly, Goldman said, the findings point to the need address the enormous barriers some students face outside school that may place them at high risk of dropping out. A recent example, he said, is the rise in absenteeism prompted by the pandemic, especially among lower-income students.

“You could have the best teachers and counselors, but if kids don’t come to school, those supports can only go so far,” Goldman said. “Sometimes what’s needed is a trusted adult – a mentor or coach – who can ask, ‘What do you need? What’s going on at home?’ and then help connect students to the right supports to get them back on track.”

This research was supported by the Harvard-based nonprofit Opportunity Insights, where Goldman is a research principal and Gracie an affiliate; EdRedesign; the Harvard Kennedy School’s Stone Program; the National Science Foundation; and Cornell.