The mission of higher education has long emphasized the role colleges and universities play as engines for social mobility, but a high-profile report from the Stanford University-based Equality of Opportunity Project (EoP) makes it clear that engine is stalled.
Most institutions graduate only a handful of students from the bottom of the income scale, admitting a disproportionately small number of poorer students and doing little to support those they admit through to completion.
Meanwhile, research from Brookings Institution uncovered an equally troubling trend. Over $13 billion in annual subsidies flow to students at selective public universities -- even ignoring benefits like tax exemptions for endowments -- making our nation’s most vital engines of equality more like inequality machines.
There are exceptions. Georgia State University recently made headlines for effectively applying equity-minded solutions. “The college has been reimagined -- amid a moral awakening and a raft of data-driven experimentation -- as one of the South’s more innovative engines of social mobility,” says a recent New York Times article profiling the university's success. This is what progress looks like.
But Georgia State is atypical, and that raises key questions. What will it take to spur this type of moral awakening industrywide? What role does innovation play in ensuring the development of methods and models that are equity oriented and effective at reaching low-income students? How do we recenter the higher education innovation agenda around social mobility?
Parsing the Data
The Equality of Opportunity Project offers a helpful guide. Using federal tax returns dating to the 1940s, Social Security records and data from IPEDS, EoP studied whether attainment actually is associated with upward mobility. The result of this effort has been the production of “mobility report cards,” which give us a common metric to track over time.
The analysis is based on the product of two calculations: the percentage of students an institution admits from low-income backgrounds, or access, and the proportion of graduates who increase their earnings by their mid-30s, or success rate. In other words, how many low-income students do institutions serve, and how effective are they at serving them? Each institution is then assigned a mobility rate, the percentage of students from low-income backgrounds who improve their earnings over time compared to their parents and thus manage to climb the income ladder.
Mobility rates help shed light on institutions that are moving the needle on driving social mobility and those that are not. Although EoP makes no such claims, we believe a reasonable interpretation of their work reveals a new breed of “top-tier” colleges and universities.
Institutions like California State University at Los Angeles and Pace University top the list with mobility rates of 9.9 percent and 8.4 percent, respectively. Other institutions, such as the University of Texas Rio Grande Valley (7.6 percent), City College of New York (7.2 percent), and the University of Texas El Paso (6.8 percent) also show staggering results in facilitating upward mobility.
In fact, looking through a mobility lens, these colleges, most of which are less selective, outperform their more selective peers across the board. Some, in fact, are entirely open access and offer credit-bearing online courses to help accelerate time to completion.
By comparison, elite colleges lag, with mobility rates hovering between 1 and 2 percent over all. Low-income students who attend elite colleges have a good chance of moving into a higher economic stratum -- a low-income Harvard student is as likely to enter the 1 percent after graduating as a wealthy Harvard student is -- but elite institutions graduate few low-income students in total. Their educational models are hardly accessible to the masses, let alone scalable.
Where Social Mobility Meets Innovation
Improving social mobility will require serving more and more low-income students at scale by recentering innovation around the development of equity-minded solutions. Such an approach will have to be powered by a moral awakening and transform the core of how our industry on the whole works -- what we value, where we invest and how we prioritize.
Alongside our peers working to advance innovation in higher ed, we are examining both what incremental changes (modest revisions to existing structures) and what transformative new models are needed. We’re just getting started, but we’ve curated a collection of bright-spot solutions, and outstanding questions, that shine light on where we think the rest of the industry must head to improve access and success, test new learning models, and invest in making social mobility research more actionable.
A handful of institutions, such as the University of Central Florida and Georgia State, have scaled their undergraduate programs specifically to increase the number of low-income learners they serve.
And the American Talent Initiative, which focuses on how elite institutions recruit and admit low-income students, may help lift the number of institutions that follow suit. But the jury is out on whether elite institutions will make dramatic changes either to increase the proportion of low-income students they accept or to scale their offerings to serve more.
The problem, unfortunately, is that tweaking enrollment criteria can alter an institution's acceptance rate and, by extension, its perceived prestige/selectivity. Timothy Renick from Georgia State describes the tension: “Many presidents would not be willing to make what I believe is the morally brave decision, but not always the most popular decision, which is to increase enrollment, serve your population better, but go down in rankings.”
While mobility requires access, an institution's "success" in this area is determined by how well it supports student efforts to complete a degree and land a job that actually enables social mobility.
For example, the University Innovation Alliance, a coalition of universities dedicated to increasing the number of low-income graduates, has found great success in scaling holistic advising, and the analytics and technology that enables it, in order to prevent students from dropping out. After four years of scaling advising and analytics, the 11 campuses are graduating 29 percent more low-income students each year. That translates to around 6,000 additional graduates annually.
Efforts to improve community college performance are critically important given the large number of low-income learners they serve each year. First, strides need to be made in aligning programs to labor market demand. This problem should not fall entirely on individual community colleges themselves; states should play a bigger role in aggregating labor data to continuously inform program creation.
Another area in need of scaled change is making it easier for students to combine work, school and parenting. For example, research on a program at Monroe Community College recently showed that parents of young children who used on-campus childcare had an on-time graduation rate three times higher than other single parents. The Urban Institute’s new U.S. Partnership on Mobility From Poverty champions both of these approaches.
New Learning Models
And then there are entirely new learning models that upend industry norms with fresh approaches. Learning models such as competency-based education -- spurred by the likes of Southern New Hampshire University and Western Governors University -- show promise in their ability to create faster, cheaper and more flexible pathways to completion, and are almost entirely focused on extending access for underserved learners.
We’re also keeping an eye on hybrid colleges, including Da Vinci Extension, Concourse Education, Match Beyond and PelotonU, that provide access to accredited programs while providing intensive wraparound student support, digital learning (often competency based) and funding models that mitigate student debt.
Apprenticeships, work-based learning and other “earn-and-learn” programs that lead to industry-validated credentials also show potential. These mechanisms give low-income students much-needed job skills and fast-track them out of minimum-wage jobs. Guild Education’s partnerships with Fortune 1000 companies, for instance, allow front-line employees to use their education benefits to get a degree and is a great example of where the puck is going here.
There also is more work to be done by educational researchers themselves, in partnership with higher ed institutions. Researchers will need to go beyond the question of which institutions are associated with mobility and ask what specific programs and interventions within institutions are driving mobility.
“The research we’ve done to date focuses on the association between college and mobility. Now we’re working to create a causal version of the research,” says John Friedman, a principal on the EoP research. “Our big vision is to provide data that supports action, so students can better choose colleges, and colleges can make reforms aimed at driving mobility.” This additional research, studying a portfolio of institutions from around the country, is set to be released over the next year.
Finally, data infrastructure needs to be built to allow for rapid feedback loops to vet whether and which new approaches or interventions are working. Today, outcomes take years to surface because the data behind mobility rates reflect earnings achieved when graduates reach their mid-30s. That isn’t helpful for institutions that need to experiment quickly, determine what works (or doesn’t), and change course as necessary.
We need better proxies for what near-term outcomes are associated with social mobility. These will likely be different from institution to institution, or could even vary between student populations attending the same college.
In a recent meeting, Erika Beck, president of California State University Channel Islands, reminded us why the social mobility rankings are rankings worth climbing. "We aren't just changing individual lives; we are changing family trees," she said.
No one believes they're antimobility, but increasing mobility remains a dilemma for higher education. The work ahead must focus our innovative efforts toward understanding -- and solving for -- the barriers to mobility so that we can design targeted solutions.
And while higher education serves many and often immeasurable roles in creating a vibrant democratic society, its primary purpose -- and the one that trumps all others -- is to be an equalizer of opportunity. To distribute opportunity based on talent, not zip code. To deconstruct classist barriers, not calcify them.