Analytics in Higher Education

Public colleges often recruit out-of-state students to bring in additional tuition dollars, but a new study finds that they may not be as much of a revenue generator as previously thought. 

An increase in out-of-state students at public colleges was linked to a decrease in per-student tuition revenue, according to the research, conducted by Robert Kelchen, a higher education professor at the University of Tennessee, Knoxville.

It was also often associated with declining spending on student services, meaning that in-state students usually didn't benefit from public colleges recruiting from beyond their state borders. 

The report's findings contradict common assumptions about out-of-state recruiting. 

"In general, don't expect to get a lot of money — or a lot more money — from out-of-state students," Kelchen said. 

Why out-of-state students may not bring in more revenue

Public colleges often focus on luring students from across state lines because they can charge them higher tuition rates than their in-state peers. Some researchers chalk this trend up to declining state support, which has forced these institutions to seek new ways to raise revenue. 

State flagship universities have increased their share of out-of-state students by an average of 55% since 2002, according to research from The Brookings Institution, a center-left think tank. This "student swap" squeezes out in-state students and drives up the cost of higher education, one of the Brookings researchers argued in a recent op-ed for The Wall Street Journal. 

But these students may not bring a revenue boost to public institutions, Kelchen's research found. More selective public colleges tended to see a decrease in total revenue when they enrolled more out-of-state students, while less-selective schools didn't see a difference.

"This surprising finding suggests that selective colleges are not gaining additional revenue from more out-of-state students who face far higher sticker prices than in-state students," Kelchen wrote in the paper.

More selective schools also saw decreases in their per-student spending as their out-of-state population rose. Less selective colleges, meanwhile, saw no changes. 

Kelchen offered several potential explanations for the findings. 

For one, the findings could be influenced by regional tuition exchanges, an arrangement among several neighboring states that allow many out-of-state students to receive the same or similar tuition rates as their in-state peers.

The Midwest Student Exchange Program, for instance, lets students from several states in the region receive tuition discounts at some 100 participating colleges. However, Illinois announced this year that it's ending its participation in the compact because no college in the state was opted into the program. At least four other states are inactive members. 

Individual institutions also offer tuition matches. The University of Maine offers incoming, academically accomplished out-of-state students the same rate as their home state's flagship university.

"The goal isn't to bring in students paying $40,000," Kelchen said. "The goal is to basically keep enrollment steady, to be able to offer the same programs and services that they used to for the people who live in Maine."

The link between declining per-student spending and increasing out-of-state students was also a surprising finding, Kelchen wrote in his study. He suggested public universities may pursue out-of-state students to shore up their finances with strategies that also include budget cuts.

Many public colleges are facing stiff headwinds. Public support for higher ed has wavered for years, with four in 10 states reporting that their funding for the sector declined in fiscal 2021, according to the annual Grapevine report. 

At the same time, some regions are seeing sharp enrollment declines. Those are expected to worsen around 2026, when there will be fewer high school graduates due to declining birth rates during the Great Recession. 

"Some of this is just trying to keep the same size you were in a challenging demographic situation," Kelchen said. 

How should colleges react?

The paper challenges the idea that out-of-state students are a revenue boon for colleges, Michael Harris, a higher education professor at Southern Methodist University, said in an email.  

"Public universities are understandably looking for additional — and stable — sources of funding with the challenge of state funding," Harris said. "The findings of this research suggest the emphasis on nonresident students may not be the windfall institutional leaders promised." 

However, each institution will have to look at its own unique circumstances, which a study using national data can't address. Brendan Cantwell, a higher education professor at Michigan State University, echoed those comments, saying in an email that campus officials will be more concerned with returns of their own recruiting efforts than sector-wide averages. 

Cantwell also said the study was well done and the findings were credible but noted that the data used has "serious limitations."

In the paper, Kelchen noted that his model did not directly account for regional tuition exchanges and that some of the trends in institutional spending may be due to changes in the way colleges report their data.

However, the findings suggest that many public college officials don't have many options to quickly change their financial situation, including by recruiting out-of-state students, Cantwell said. 

"Policy decisions probably matter more than campus strategy," he said.

State demographic shifts that change the demand for certain higher education programs are often why public colleges consider mergers, according to new research from Ithaka S+R.

Local and state politics also heavily influence consolidations, the research and consulting group found. Its new work delves into mergers in Wisconsin, Texas and Georgia, with separate case studies on mergers in each state.

College administrators should plan for how mergers affect racial, ethnic and socioeconomic minorities on campus — populations often not central in these arrangements, the report's authors argue.

American higher education has undergone a period of significant consolidation in the last decade, according to the report released Monday. Recent merger proposals come from Vermont, where the Vermont State Colleges System Board of Trustees voted this year to unify its three residential colleges, and New Hampshire, where the governor attempted to combine the state's two- and four-year systems until lawmakers intervened.

These discussions have also taken on new relevance amid the pandemic's economic turmoil.

Some of the most visible mergers occurred in Georgia and Wisconsin, which along with Texas were the primary focus of the Ithaka S+R research.

The University System of Georgia reduced the number of institutions under its umbrella from 35 to 26 between 2013 and 2018. Officials were attempting to find cost efficiencies, reduce duplication in programs and boost attainment levels.

In 2018, the University of Wisconsin System carried out a major restructuring of its institutions largely centered around its four-year universities absorbing its 13 two-year schools.

And the University of Texas System in 2012 began the process of merging the University of Texas Brownsville and the University of Texas Pan American into a new institution, the University of Texas Rio Grande Valley. The system billed the move as a way to save Brownsville, which had lost a partnership with a two-year school that provided it with resources and transfer students.

The report's authors stress that the circumstances were unique in each merger, but some common themes emerged.

In Georgia and Wisconsin, population changes left some rural institutions with fewer potential students than in past years, a trend that was not expected to improve. And especially among Wisconsin's two-year schools, declining state investment and enrollment challenges limited the programs they could offer.

The situation was nearly inverse in Texas, the report authors note. The area where the two combined colleges were located was long underserved, which the report attributes to "a legacy of de jure and de facto discrimination against people of Mexican ancestry in the state and the region." Merging the colleges unlocked new capital and capacity for an area with growing demand, the report states.

But politics can complicate these deals. An alternative to merging the Wisconsin two-year colleges was closing some of the most severely under-enrolled campuses. But this option was set aside to appease Republican lawmakers who controlled state government and often represented the small towns where the two-year schools were located, according to the research.

College officials can make mergers more palatable when they emphasize them as opportunities for growth rather than cutbacks. Regents in Georgia homed in on how the consolidations could improve student services, the array of programs and economic development "with efficiency only as a means to those ends."

The authors call for administrators to account for equity concerns when mapping out mergers. When these deals happen, it is often difficult to track data on how disadvantaged students are affected. But without doing so, "gains or setbacks may go under the radar" for these students.

In Georgia, for example, one of the colleges that merged was Southern Polytechnic State University, which historically was a prominent place for Black students to pursue bachelor's degrees in engineering fields. But concerns arose that if it joined with Kennesaw State University, it would lose the "cultural environment that nurtured a robust African American student body."

Officials should consider not just current and prospective students in mergers but also faculty, staff, alumni and community members, the report states.

"Higher education institutions are the sites of many personal and economic connections, and all of the people involved in those relationships will be affected," it says.

Colleges may need to lean more on artificial intelligence (AI) and data analytics to curb the effects of a declining number of traditional-age students, as well as lower levels of state support, according to a report from Educause. 

The 2020 Horizon Report pinpoints some of the challenges higher education leaders are facing and how emerging technology may be able to help. 

AI and data analytics, among other newer technologies, are already reshaping how colleges interact with students and potential recruits, the report's authors note. 

Many colleges, for instance, have rolled out automated chatbots to help answer students' questions and direct them to institutional resources. 

Northwestern University, in Illinois, developed a chatbot within the learning management system that can answer common questions and search for documents related to students' inquiries. Several colleges have improved their enrollment or graduation rates by adding such technology. 

Other institutions are using these tools to recruit new pools of students to make up for an expected drop-off in the number of high school graduates over the next decade. 

As competition over students heats up, colleges "that skillfully navigate these demographic trends may not see significant enrollment increases," Kevin Gannon, a professor and director of the Center for Excellence in Teaching and Learning at Iowa's Grand View University, wrote in an essay included in the Educause report. "[S]imply holding steady is a successful outcome," Gannon added.

Excelsior College, a private online university based in New York, is partnering with ReUp Education, an ed tech startup that uses a mix of automated messaging and human coaches to help colleges enroll students who dropped out. The company also has contracts with public university systems in Texas, Pennsylvania and New York.

Colleges are also attempting to maintain headcounts by using data analytics to boost retention and other student success measures.

The University of California, Berkeley, developed a platform that uses data on student performance to help advisers determine which students are showing signs of academic trouble. 

Other institutions have similar tools for students. At the University of Iowa, students can view a dashboard that shows them the grade they're on pace to earn in a given course and suggests areas in which they can improve. 

Students who dropped out of community college were 21% more likely to reenroll if they received a one-course tuition waiver and text messages with information about how to return, University of Florida researchers found in a study. 

Just providing information via texts, however, did not have a major impact on whether students reenrolled.

The findings add to previous research that suggests nudges, which are digital alerts designed to influence behavior, have a limited effect on college attendance when used on their own. 

The study split 27,028 former students into three groups. One received texts about the enrollment process and key deadlines, one received that information plus a tuition waiver for one course, and another was the control group.

Noting that the process of reenrolling in college could be "complex" and "challenging to navigate" for students, the researchers worked with five community colleges, all in Florida, to create custom websites intended to make it easier for students to do so. 

"One thing that was really surprising to us was the fact that the information alone was not something we could claim led to reenrollment," said Justin Ortagus, one of the study's authors. But when the nudges were paired with a financial incentive, students were not only more likely to come back but also to take more than one course. 

"From a financial revenue perspective, it also makes sense for the institution," Ortagus said. 

The campaign worked particularly well for students who had lower GPAs or had accumulated more credits than other students in the study and those who were older than traditional-aged undergraduates, the researchers found. 

The study adds to previous findings that suggest nudges alone may not influence college-going behavior, though personalizing them could help. 

It also has implications for community colleges and four-year institutions that focus on adult learners, both of which are hoping to reenroll some of the 36 million Americans who have completed some college but don't have a degree. 

This market could be especially important to community colleges, where completion rates hover around 30%, according to federal data. Boosting student outcomes by bringing back stopped-out students may also bring in more revenue for institutions in states that use performance-based funding, Ortagus said. 

Colleges' reenrollment campaigns take various forms. Some colleges contract with vendors that specialize in bringing back students.

In late 2018, the Texas A&M University System partnered with ReUp Education, a startup that uses a combination of automated messaging and human coaches to reenroll stopped-out students and support them until they graduate. The company reenrolled about 300 students across the system for the following fall semester, Inside Higher Ed reported. 

Other institutions have been using financial incentives to encourage students to return. In 2016, Colorado's Pueblo Community College promised to forgive students up to $1,000 in institutional debt after they reenrolled and finished one semester. As of April 2019, the effort had brought back more than 300 students and recouped roughly $350,000 in revenue.

Higher education often gets criticized for its silos, both within institutions and among them. So when the University Innovation Alliance launched publicly in 2014, its bold ambition of getting universities to share ideas to help them improve college access was met with skepticism. 

"This has been a fairly intractable space where every institution is doing really interesting experiments, and everybody has a narrative," Bridget Burns, UIA's executive director, told attendees at the Educause conference in Chicago in October 2019. "(But) the diffusion of innovation is glacial, and most people are not very aware of what's going on, and most people struggle with actually how to fix it."

After all, what could one public research university — in its own world of state funding, geographic enrollment trends and institution-specific growth plans — do that another could effectively implement?

A lot, it turns out. By examining and rethinking areas of the institution that could be improved with better data, more information-sharing and, in some cases, the help of technology, the UIA schools are on track to graduate an additional 94,000 students by 2025. That's ahead of their initial mark of 68,000 additional students. 

The group of 11 institutions comprises Arizona State, Georgia State, Iowa State, Michigan State, Ohio State, Oregon State and Purdue universities, as well as the University of California, Riverside, the University of Central Florida, the University of Kansas and the University of Texas at Austin. 

Together, they have examined how to collect and use data to spot students who are falling behind or may soon be — a process implemented increasingly across higher ed, though not without some criticism. Other areas of focus for UIA include redesigning the link between higher ed, career services and employers, and exploring how chatbots can be used on campuses. How colleges can become more user-friendly for adult learners is also expected to be on the table.

Through it all, the campuses have shared lessons while still approaching the issues with the unique needs of their own institutions in mind. "We know that a cookie-cutter approach is a bad idea," Burns told Higher Ed Dive in 2019.

Higher Ed Dive sat down with Burns in the fall of 2019 in Washington, D.C., to talk more about what UIA has learned about collaborations among universities, and how those lessons can be applied in other consortia and within institutions of all sizes.

This interview has been edited and condensed.

HIGHER ED DIVE: Predictive analytics was UIA's first project. Was that by design?

BURNS: We needed a scoreboard to know if our work was making a difference. If you haven't done the work to identify your indicators, and you haven't synced your data system, then you're dealing with very obtuse metrics that aren't always helpful. 

We've reported on some of the challenges UIA schools faced in that process, one of which was that many first needed to map their processes and data internally. Did you expect that to be a challenge?

I had no idea what the challenges were going to be. It's like anything where you take on a new skill or ability, in that you don't actually know what your gaps are, and you can't focus on those because you wouldn't probably start. Those were critical in moving forward.

Where were you starting from?

We had an example of an institution that had figured something out and was showing some impressive capacity and capability that other campuses wanted. Initially, we framed it as there were mentors and mentees, but we walked away from that language because nobody wants to be called a mentee. We actually found in some cases we were learning more from those in the mentee role than we were from the mentor.

How so?

Quickly, we learned that nobody wants to just emulate one institution. What we were trying to pursue was: What lessons could you learn from collective scale, what could you sift through all these different experiences and pull out as the gold?

You've said that required building trust among UIA schools. How did that happen, and what can colleges and universities looking to build trust internally learn from that?

Initially, I traveled to each campus and interviewed people about what was going wrong, what was hard, what was getting in the way. After about five of those it was so clear: Every campus had the same problems, but none of them realized it. 

In most collaborations, it's not just about what can we do together. It's about what problem can we solve together that I simply cannot handle on my own. In this case, that was around overwhelm and not really understanding what the first steps are. A lot of how institutions share involves elevating examples of work in such a high fashion that people talk about what was achieved and not about the questions they couldn't answer, or asking the dumb question, or figuring out what's the first step. Those are the things people want to talk about. 

The first priority in building a collaboration is figuring out how to create the conditions for a truly candid conversation about shared struggles.

Is higher ed different from any other industry in that way?

I don't think they are truly unique problems, but unlike other sectors, we really do somehow believe we are competing with every single other university or college, even if we are not in the same space. Part of it is the general sense of scarcity that's communicated through media, state budgets, who gets elevated as the innovator. 

Another part of it is that it is not very common — or not as common as it should be — to have worked at multiple universities due to geographic constraints. Every university has a story about why they do what they do, and if you see enough of those stories, you can see them for what they are. But if you work at one or two institutions, maybe three, maybe you're not picking up that these are sector-wide problems and we shouldn't be pushing this up a hill.

What can smaller colleges that don't have the internal economies of scale of UIA schools learn from that work?

I don't think UIA's lessons have been about the number of students we serve. They've been about fairly typical universities struggling with fairly typical challenges but going about it unusually. On the data front, in particular, the lessons we gathered carry for any size of campus. Your first job is to identify the top 10 indicators that a student is about to drop out. That seems simple, but the vast majority of campuses don't know the answer to that question, and it should keep them up at night. 

Small colleges should team up with other small colleges. What problems do you have at your institution that are consistent and you haven't been able to solve? Find others struggling with those same problems and set up a conversation that allows you to solve it together.

You said during a talk at Educause 2019 that "everyone has massive, million-dollar failures on their campus" that don't get discussed. UIA schools seem to have figured out a way to talk about some of that. What can other colleges learn from their approach?

Most of the time, I have a hard time getting institutions to acknowledge that something didn't work out. It's about creating a space where there's something else there instead of it just being the word 'failure.' I’m still looking for exceptional examples of the implementation of autopsy processes inside a campus. 

And it depends on where you sit. Some presidents tell me they have a great process, but then I talk to people on their team who will say, "I don't know what they're talking about." It's possible that isn't actually a failure. It's possible it's about leadership communication and that people don't understand what the actual goal is. It's also possible that people on that campus don't understand the perspective of the leader and don't see the bigger picture. 

In general, I would say it is worth leaning in because that demonstrates there is a need for greater understanding.