Breaking Away from Rankings

The Growing Movement to Reform Research Assessment and Rankings

By Dean Hoke, September 22, 2025: For the past fifteen years, I have been closely observing what can only be described as a worldwide fascination—if not obsession—with university rankings, whether produced by Times Higher Education, QS, or U.S. News & World Report. In countless conversations with university officials, a recurring theme emerges: while most acknowledge that rankings are often overused by students, parents, and even funders when making critical decisions, few deny their influence. Nearly everyone agrees that rankings are a “necessary evil”—flawed, yet unavoidable—and many institutions still direct significant marketing resources toward leveraging rankings as part of their recruitment strategies.

It is against this backdrop of reliance and ambivalence that recent developments, such as Sorbonne University’s decision to withdraw from THE rankings, deserve closer attention

In a move that signals a potential paradigm shift in how universities position themselves globally, Sorbonne University recently announced it will withdraw from the Times Higher Education (THE) World University Rankings starting in 2026. This decision isn’t an isolated act of defiance—Utrecht University had already left THE in 2023, and the Coalition for Advancing Research Assessment (CoARA), founded in 2022, has grown to 767 members by September 2025. Together, these milestones reflect a growing international movement that questions the very foundations of how we evaluate academic excellence.

The Sorbonne Statement: Quality Over Competition

Sorbonne’s withdrawal from THE rankings isn’t merely about rejecting a single ranking system. It appears to be a philosophical statement about what universities should stand for in the 21st century. The institution has made it clear that it refuses to be defined by its position in what it sees as commercial ranking matrices that reduce complex academic institutions to simple numerical scores.

Understanding CoARA: The Quiet Revolution

The Coalition for Advancing Research Assessment represents one of the most significant challenges to traditional academic evaluation methods in decades. Established in 2022, CoARA has grown rapidly to include 767 member organizations as of September 2025. This isn’t just a European phenomenon—though European institutions have been early and enthusiastic adopters. The geographic distribution of CoARA members tells a compelling story about where resistance to traditional ranking systems is concentrated. As the chart shows, European countries dominate participation, led by Spain and Italy, with strong engagement also from Poland, France, and several Nordic countries. This European dominance isn’t accidental—the region’s research ecosystem has long been concerned about the Anglo-American dominance of global university rankings and the way these systems can distort institutional priorities.

The Four Pillars of Reform

CoARA’s approach centers on four key commitments that directly challenge the status quo:

1. Abandoning Inappropriate Metrics The agreement explicitly calls for abandoning “inappropriate uses of journal- and publication-based metrics, in particular inappropriate uses of Journal Impact Factor (JIF) and h-index.” This represents a direct assault on the quantitative measures that have dominated academic assessment for decades.

2. Avoiding Institutional Rankings Perhaps most relevant to the Sorbonne’s decision, CoARA commits signatories to “avoid the use of rankings of research organisations in research assessment.” This doesn’t explicitly require withdrawal from ranking systems, but it does commit institutions to not using these rankings in their own evaluation processes.

3. Emphasizing Qualitative Assessment The coalition promotes qualitative assessment methods, including peer review and expert judgment, over purely quantitative metrics. This represents a return to more traditional forms of academic evaluation, albeit updated for modern needs.

4. Responsible Use of Indicators Rather than eliminating all quantitative measures, CoARA advocates for the responsible use of indicators that truly reflect research quality and impact, rather than simply output volume or citation counts.

European Leadership

Top 10 Countries by CoARA Membership:

The geographic distribution of CoARA members tells a compelling story about where resistance to traditional ranking systems is concentrated. As the chart shows, European countries dominate participation, led by Spain and Italy, with strong engagement also from Poland, France, and several Nordic countries. This European dominance isn’t accidental—the region’s research ecosystem has long been concerned about the Anglo-American dominance of global university rankings and the way these systems can distort institutional priorities.

The geographic distribution of CoARA members tells a compelling story about where

Prestigious European universities like ETH Zurich, the University of Zurich, Politecnico di Milano, and the University of Manchester are among the members, lending credibility to the movement. However, the data reveals that the majority of CoARA members (84.4%) are not ranked in major global systems like QS, which adds weight to critics’ arguments about institutional motivations.

CoARA Members Ranked vs Not Ranked in QS:

The Regional Divide: Participation Patterns Across the Globe

What’s particularly striking about the CoARA movement is the relative absence of U.S. institutions. While European universities have flocked to join the coalition, American participation remains limited. This disparity reflects fundamental differences in how higher education systems operate across regions.

American Participation: The clearest data we have on institutional cooperation with ranking systems comes from the United States. Despite some opposition to rankings, 78.1% of the nearly 1,500 ranked institutions returned their statistical information to U.S. News in 2024, showing that the vast majority of American institutions remain committed to these systems. However, there have been some notable American defections. Columbia University is among the latest institutions to withdraw from U.S. News & World Report college rankings, joining a small but growing list of American institutions questioning these systems. Yet these remain exceptions rather than the rule.

European Engagement: While we don’t have equivalent participation rate statistics for European institutions, we can observe their engagement patterns differently. 688 universities appear in the QS Europe ranking for 2024, and 162 institutions from Northern Europe alone appear in the QS World University Rankings: Europe 2025. However, European institutions have simultaneously embraced the CoARA movement in large numbers, suggesting a more complex relationship with ranking systems—continued participation alongside philosophical opposition.

Global Participation Challenges: For other regions, comprehensive participation data is harder to come by. The Arab region has 115 entries across five broad areas of study in QS rankings, but these numbers reflect institutional inclusion rather than active cooperation rates. It’s important to note that some ranking systems use publicly available data regardless of whether institutions actively participate or cooperate with the ranking organizations.

This data limitation itself is significant—the fact that we have detailed participation statistics for American institutions but not for other regions may reflect the more formalized and transparent nature of ranking participation in the U.S. system versus other global regions.

American universities, particularly those in the top tiers, have largely benefited from existing ranking systems. The global prestige and financial advantages that come with high rankings create powerful incentives to maintain the status quo. For many American institutions, rankings aren’t just about prestige—they’re about attracting international students, faculty, and research partnerships that are crucial to their business models.

Beyond Sorbonne: Other Institutional Departures

Sorbonne isn’t alone in taking action. Utrecht University withdrew from THE rankings earlier, citing concerns about the emphasis on scoring and competition. These moves suggest that some institutions are willing to sacrifice prestige benefits to align with their values. Interestingly, the Sorbonne has embraced alternative ranking systems such as the Leiden Open Rankings, which highlight its impact.

The Skeptics’ View: Sour Grapes or Principled Stand?

Not everyone sees moves like Sorbonne’s withdrawal as a noble principle. Critics argue that institutions often raise philosophical objections only after slipping in the rankings. As one university administrator put it: “If the Sorbonne were doing well in the rankings, they wouldn’t want to leave. We all know why self-assessment is preferred. ‘Stop the world, we want to get off’ is petulance, not policy.”

This critique resonates because many CoARA members are not major players in global rankings, which fuels suspicion that reform may be as much about strategic positioning as about values. For skeptics, the call for qualitative peer review and expert judgment risks becoming little more than institutions grading themselves or turning to sympathetic peers.

The Stakes: Prestige vs. Principle

At the heart of this debate is a fundamental tension: Should universities prioritize visibility and prestige in global markets, or focus on measures of excellence that reflect their mission and impact? For institutions like the Sorbonne, stepping away from THE rankings is a bet that long-term reputation will rest more on substance than on league table positions. But in a globalized higher education market, the risk is real—rankings remain influential signals to students, faculty, and research partners.
Rankings also exert practical influence in ways that reformers cannot ignore. Governments frequently use global league tables as benchmarks for research funding allocations or as part of national excellence initiatives. International students, particularly those traveling across continents, often rely on rankings to identify credible destinations, and faculty recruitment decisions are shaped by institutional prestige. In short, rankings remain a form of currency in the global higher education market.

This is why the decision to step away from them carries risk. Institutions like the Sorbonne and Utrecht may gain credibility among reform-minded peers, but they could also face disadvantages in attracting international talent or demonstrating competitiveness to funders. Whether the gamble pays off will depend on whether alternative measures like CoARA or ROI rankings achieve sufficient recognition to guide these critical decisions.

The Future of Academic Assessment

The CoARA movement and actions like Sorbonne’s withdrawal represent more than dissatisfaction with current ranking systems—they highlight deeper questions about what higher education values in the 21st century. If the movement gains further momentum, it could push institutions and regulators to diversify evaluation methods, emphasize collaboration over competition, and give greater weight to societal impact.

Yet rankings are unlikely to disappear. For students, employers, and funders, they remain a convenient—if imperfect—way to compare institutions across borders. The practical reality is that rankings will continue to coexist with newer approaches, even as reform efforts reshape how universities evaluate themselves internally.

Alternative Rankings: The Rise of Outcome-Based Assessment

While CoARA challenges traditional rankings, a parallel trend focuses on outcome-based measures such as return on investment (ROI) and career impact. Georgetown University’s Center on Education and the Workforce, for example, ranks more than 4,000 colleges on the long-term earnings of their graduates. Its findings tell a very different story than research-heavy rankings—Harvey Mudd College, which rarely appears at the top of global research lists, leads ROI tables with graduates projected to earn $4.5 million over 40 years.

Other outcome-oriented systems, such as The Princeton Review’s “Best Value” rankings, emphasize affordability, employment, and post-graduation success. These approaches highlight institutions that may be overlooked by global research rankings but deliver strong results for students. Together, they represent a pragmatic counterbalance to CoARA’s reform agenda, showing that students and employers increasingly want measures of institutional value beyond research metrics alone.

These alternative models can be seen most vividly in rankings that emphasize affordability and career outcomes. *The Princeton Review’s* “Best Value” rankings, for example, combine measures of financial aid, academic rigor, and post-graduation outcomes to highlight institutions that deliver strong returns for students relative to their costs. Public universities often rise in these rankings, as do specialized colleges that may not feature prominently in global research tables.

Institutions like the Albany College of Pharmacy and Health Sciences illustrate this point. Although virtually invisible in global rankings, Albany graduates report median salaries of $124,700 just ten years after graduation, placing the college among the best in the nation on ROI measures. For students and families making education decisions, data like this often carries more weight than a university’s position in QS or THE.

Together with Georgetown’s ROI rankings and the example of Harvey Mudd College, these cases suggest that outcome-based rankings are not marginal alternatives—they are becoming essential tools for understanding institutional value in ways that matter directly to students and employers.

Rankings as Necessary Evil: The Practical Reality

The CoARA movement and actions like Sorbonne’s withdrawal represent more than just dissatisfaction with current ranking systems. They reflect deeper questions about the values and purposes of higher education in the 21st century.

If the movement gains momentum, we could see:

Diversification of evaluation methods, with different regions and institution types developing assessment approaches that align with their specific values and goals

Reduced emphasis on competition between institutions in favor of collaboration and shared improvement

Greater focus on societal impact rather than purely academic metrics

More transparent and open assessment processes that allow for a better understanding of institutional strengths and contributions

Conclusion: Evolution, Not Revolution

The Coalition for Advancing Research Assessment and decisions like Sorbonne’s withdrawal from THE rankings represent important challenges to how we evaluate universities, but they signal evolution rather than revolution. Instead of the end of rankings, we are witnessing their diversification. ROI-based rankings, outcome-focused measures, and reform initiatives like CoARA now coexist alongside traditional global league tables, each serving different audiences.

Skeptics may dismiss reform as “sour grapes,” yet the concerns CoARA raises about distorted incentives and narrow metrics are legitimate. At the same time, American resistance reflects both philosophical differences and the pragmatic advantages U.S. institutions enjoy under current systems.

The most likely future is a pluralistic landscape: research universities adopting CoARA principles internally while maintaining a presence in global rankings for visibility; career-focused institutions highlighting ROI and student outcomes; and students, faculty, and employers learning to navigate multiple sources of information rather than relying on a single hierarchy.

In an era when universities must demonstrate their value to society, conversations about how we measure excellence are timely and necessary. Whether change comes gradually or accelerates, the one-size-fits-all approach is fading. A more complex mix of measures is emerging—and that may ultimately serve students, institutions, and society better than the systems we are leaving behind. In the end, what many once described to me as a “necessary evil” may persist—but in a more balanced landscape where rankings are just one measure among many, rather than the single obsession that has dominated higher education for so long.


Dean Hoke is Managing Partner of Edu Alliance Group, a higher education consultancy. He formerly served as President/CEO of the American Association of University Administrators (AAUA). Dean has worked with higher education institutions worldwide. With decades of experience in higher education leadership, consulting, and institutional strategy, he brings a wealth of knowledge on colleges’ challenges and opportunities. Dean is the Executive Producer and co-host for the podcast series Small College America.

Why Are So Many Smaller Independent Colleges and Universities So Similar and What Does This Mean for Their Futures?

September 8, 2025, by Dr. Chet Haskell: It is well known that many small American private non-profit academic institutions face serious financial pressures. Typically defined as having 3000 or fewer students, more than 170 of these have been forced to close in the past two decades. Numerous others have entered into various mergers or acquisitions, often with well-documented negative impacts on students, faculty, staff, alumni and local communities. Of the more than 1100 such institutions, at least 900 continue to be a risk.

The basic problems responsible for this trend are also well-known. Most institutions lack significant endowments and are thus almost totally dependent on tuition and fee revenues from enrolled students. Only 60 such small institutions have per student endowments in excess of $200,000. The remainder have far less.

The only additional potential source of revenue – gifts and donations –is generally neither large nor consistent enough to offset enrollment-related declines. While the occasional donation or bequest in the millions of dollars garners attention, most institutions raise much smaller amounts regularly.

Enrollment declines are the existential threat to many of these smaller colleges and universities. These declines are also well-documented. There simply will be fewer high school graduates in the US in the coming decade or more. This reality creates a highly competitive environment, especially in regions with many of these institutions.

Demographic worries are augmented by broad concerns about the cost of higher education and the imputed return on such an investment by students and families. Governmental policies such as limitations on international students or restrictions on immigration further add to the problem. Also, these institutions not only compete with each other for students, but they also compete with colleges and universities of the public sector and a growing number of for-profit entities.

Most of these 900 or so institutions have high quality programs, often described under the term “liberal arts”. Many are differentiated by a specialization or an emphasis. However, at their core they are very similar. The basic concept of a personal scale four-year undergraduate educational experience provided in a residential campus setting has a long history and is highly valued by many students and faculty alike. These institutions have lengthy, strong histories, loyal alumni and important roles in their local communities.

The fact is that it is difficult to differentiate among many of these institutions. Not only their scale or their general model of personalized undergraduate education are similar, but many of their basic messages sound the same. A review of the websites of these schools results in striking consistencies of stated “unique” missions, programs, facilities, faculty and even marketing materials.

Their approaches to financial challenges are also similar. There is considerable competition on price. Most of these institutions discount their formal tuition rates by 50% or more. Initiatives to grow enrollments support an industry of educational consultants whose recommended initiatives are themselves similar and, even if successful, are quickly copied, thus reducing advantages.

Some have tried to compete by raising money for new, attractive facilities through dipping into limited endowments, borrowing or securing external major gifts. These shiny new buildings – athletic facilities, science centers, student centers – are assumed to provide an edge in student recruitment. In some cases, this works. However, in many others the new facilities do not come with long term maintenance and eventually add to increased on-going institutional expense. The end result is often another demonstration of similarity.

Some institutions have tried to branch out into selected graduate programs, perhaps based on a strong group of undergraduate faculty. Success is often limited for multiple reasons. Graduate students in commonly introduced professional fields such as business or nursing do not naturally align with an undergraduate in-person academic calendar. Older students, especially those in careers, are reluctant to come to a campus for class twice a week. Even if there is sufficient interest in such a program, it is difficult to increase in scale because of the limits of distance and geography. And most of these institutions lack significant expertise and technology do conduct effective on-line operations.

Their institutional similarities extend to their governance. Typically, there is a Board of Trustees, all of whom are volunteers, often with heavy alumni representation. These boards generally lack expertise or perspective on the challenges of higher education and thus are dependent on the appointed executive leadership. They often take a short-term perspective and lack strategic foresight that may be most valuable in times of uncertainty and external changes.

Even when trustees have financial experience from other fields, their common approach to small institutions is to bemoan any lack of enrollments. Most do not make significant personal financial contributions, particularly if they think the institution is struggling to survive. The assumed budget goal is basically a balanced budget and when one does not control revenues, one focuses on the more controllable expense side, trying to balance budgets solely on cuts.  Board members serve because they want to support the institution, but many are risk adverse. For example, a fear of being associated with an institution that might generate possible legal liability for the board member means a first concern usually involves whether there is sufficient insurance.

While every institution is indeed different in its own way, they also are very similar. What explains this?

One possible way of explanation is provided by the organizational theorists Walter Powell and Paul DiMaggio who in 1983 (updated in 1991) published a seminal piece on what they called ”institutional isomorphism and collective rationality.” [1]They argued that ”institutions in the same field become more homogenous over time without become more efficient or more successful” and identified three basic reasons for such a tendency.

Coercive isomorphism – similarities imposed externally on the institutions. In higher education, good examples would be Federal government policies around student financial aid or the requirements of both regional and specialized accreditors. Every institution operates within a web of regulation and financial incentives that impose requirements on all and work to limit innovation.

Mimetic processes – similarities that arise because of standard responses to uncertainty. Prime examples in higher education are the increasingly common responses to the quest for enrollment growth. As noted, numerous consultants purport to improve enrollments, but the gains typically are limited, as other institutions mimic the same approach. In another example, recent surveys show that almost all institutions expect to be users of artificial intelligence models to promote marketing in the service of admissions, as if this is a “magic wand”. If one institution makes strides in this area, others will follow. The result will be more similarity, not less. It is a bit like the Ukrainian-Russian war, where Ukraine originally had clear advantages using drone technology until that technology was matched by the Russians, leading to a form of stalemate. As DiMaggio and Powell note, ”organizations tend to model themselves after similar organizations in their field that they perceive as more legitimate or successful.”[2]

Normative pressures – similarities that arise from common “professional” expectations. The authors identify two important aspects of professionalization: the common basis of higher education credentials and the legitimation produced by these credentials and “the growth and elaboration of professional networks.” Examples include common faculty and senior administrator qualification requirements. Another would be so-called “best practices” in support areas like student affairs. “Such mechanisms create a pool of almost interchangeable individuals who occupy similar positions.”[3] Recently, Hollis Robbins pointed out the commonalities in paths to academic leadership positions, likening these to the Soviet nomenklatura process through which a leader progresses in one’s career.[4] Evidence of this is obvious through a cursory review of the qualifications and desired qualities posted in searches for college and university presidents or other senior administrators. Most searches end up looking for and hiring individuals with very similar qualifications and experience.

The implications of such pressures and processes are several. With common values and similar personnel, “best practices” do not lead to essential changes. Innovation is quickly copied. Indeed, it becomes increasing difficult to differentiate an institution from competitors. Common regulatory structures, declining student pools, increased competition and a lack of resources for investment all combine to enhance similarity over difference. In some sense, it is almost a form of commodification where price does in fact matter, but the “product” basically the same, especially in the minds of the larger population of potential students and families.

What is to be done?

Leadership Must Confront Their Institution’s Reality

Confronting reality has many aspects, but the leaders of every institution must be clear-eyed and unsentimental about where it stands and where it is headed. This is an essential role for boards and executive leadership.

First and foremost, the mission of the institution must be understood in realistic and practical ways. What is the institution’s purpose and what is required to fulfill that purpose? Institutional mission is central as it should drive an appreciation for the current situation of the institution, provide clarity regarding longer term goals and bringing into focus the necessary means to move forward.

With clarity of mission must come a full understanding the of institution’s financial situation, its opportunities and the longer term needs required to achieve mission goals.  Building multi-year mission-oriented budgets based on surpluses (positive margins) is key. Sometimes restructuring and cuts are necessary and thus leadership must make sure all faculty and staff have a clear understanding of reality and the strategy for addressing it.

A clear understanding by all of the marginal results (positive and negative) of major components is also critical. Some elements or units return significant positive margins. Others less so. And some return negative margins, often year after year. Yet, some of these less financially productive elements may be essential to mission and must be balanced or subsidized by other elements. At the end of the day, it is the margin of the entire institution that matters. And, as the saying goes, “no margin, no mission.” However, the opposite is also true. Institutions that are unclear about their mission will be challenged to attract and motivate students, faculty, staff or major donations.

Every institution must worry about enrollments as the largest source of revenue. Declining enrollments force expense restraints. Every institution must also be concerned about growing enrollments as a key prerequisite of financial stability. Institutions operating on thin or negative margins cannot hope to achieve their mission goals without some form of growth, including having the resources to invest in growth. Without some forms of growth, an institution will either be at risk or will have to make sometimes radical changes in order to continue to pursue mission goals. The only real alternative is to amend the mission and the definition of its success.

The other important point is that all institutions are subject to unexpected external pressures that they cannot control. Examples would be 9/11, the 2008-09 Great Recession, the COVID pandemic or the advent new government policies, such as those confronting all institutions today. Coping with such events requires having some financial resiliency, strong leadership and creativity.

Yet, the combination of external pressures and the realities of small-scale institutions operating on thin margins in the face of extensive competition may mean that even the best managed and led organizations will confront existential risk.

For many institutions, merging or partnering with another institution may be the only realistic path. While there often is reluctance to cede independence to another institution, mergers are hardly new, as consolidation in US higher education is hardly a new phenomenon. There are several hundred examples of mergers, many going back a century or more. Washington and Jefferson College in Pennsylvania in 1865 is the result of such an arrangement, as is Case Western Reserve University in Ohio a century later. In addition to these mergers, hundreds of other institutions have simply closed, including at least 170 in the past twenty years.

Additionally, may institutions may be placed to take advantage of consortium relationships with other institutions. Again, there are numerous examples of institutions seeking to improve their situations through this form of collaboration. Participating institutions collaborate on such things as sharing costs or providing a wider range of student options, while remaining independent. However, this model, while valuable in many ways, rarely provides major financial advantages except at the margins. And successful consortia require a certain degree of independent sustainability for each member.

Still others may be able find opportunity in growth through symbiosis. The recent Coalition for the Common Good begun by Antioch and Otterbein universities is an example. Other variants are possible. However, again such middle ground models also assume a basic stability of the members. As stated by Coalition president, John Comerford, “we are looking for a sweet spot of resources. This is not a way to save a school on death’s door. It’s also probably not useful to a school with billions in their endowment. Institutions in the big middle ground both need to look at new business models and likely have some flexibility to invest in them.” This type of model will not work in many cases.

The point is that many of these small college will continue to be at risk as long as they are tuition dependent within a shrinking pool of potential students and insufficient external support. Fewer and fewer small institutions will be able to survive independently simply because of the financial challenges inherent in their small-scale model.

Small undergraduate institutions represent the highest ideals of higher education. They are a key source for graduate students and future professors. They are central to their communities. Their strengthening and preservation as a class is an essential element of the American higher education ecosystem with its wide range of institutional models and opportunities. But this does not mean all can survive.

The leaders of every institution need to have a clear and practical plan for the maintenance of their independence, while also being open to careful consideration of alternatives, exploring potential alternatives well before they face a crisis.

Notes:

  1. DiMaggio, Paul and Powell, Walter, The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields in DiMaggio and Powell, The New Institutionalism in Organizational Analysis, University of Chicago Press, 1991. (pp.63-82)
  2. Ibid. p. 70
  3. Ibid. p. 71
  4. Hollis Robbins, The Higher Ed Nomenklatura? Inside Higher Education, May 12, 2025

The next essay in this series will examine in some detail the steps in a process that begins with acknowledging the possible need for a partner and hopefully results in an agreement that is implemented.


 Dr. Chet Haskell is Senior Consultant and Higher Education Strategist at Edu Alliance Group. He brings over four decades of leadership and consulting experience in higher education, with a career spanning the U.S., Europe, Latin America, and Asia. He has held senior roles, including President of the Monterey Institute of International Studies and Cogswell Polytechnical College, Dean and de facto Provost at Simmons College, and 13 years of leadership positions at Harvard University and five years in senior administrative roles at the University of Southern California.

As Provost and Chief Academic Officer of Antioch University, he helped lead the creation of the Coalition for the Common Good, a groundbreaking alliance with Otterbein University. Internationally, Dr. Haskell has advised universities in Mexico, Spain, Holland, and Brazil and served as a consultant to the Council for Higher Education Accreditation (CHEA), the Western Association of Schools and Colleges (WASC) and the Council on International Quality Group.

A respected accreditation expert, he has served as a WSCUC peer reviewer and as an international advisor to ANECA (Spain) and ACAP (Madrid). He is a frequent speaker at global conferences and meetings.