Reimagining an American Standard: A Deep Dive into the 2025 Carnegie Classifications

2025 AIR Forum attendees packed the room for the session “Unpacking the 2025 Carnegie Classifications,” an update on the most significant overhaul of the Carnegie Classification system in its 50-year history. Led by Sara Gast (American Council on Education), Mushtaq Gunja (American Council on Education), and Kyle Whitman (American Council on Education), the session offered a comprehensive look into the goals, process, structure, and implications of the 2025 reclassification. 

For decades, the Carnegie Classifications have served as the higher education sector’s “common language,” sorting colleges and universities into categories used in research, funding models, accreditation, rankings, and internal benchmarking. But until now, the underlying logic of the system—sorting institutions primarily by the highest degree awarded—had changed little since its inception in 1973. Gunja opened the session by framing the reclassification effort as both a necessary and overdue modernization, driven by widespread shifts in institutional missions, student demographics, academic program mixes, and how classification data are used externally. 

“Higher education has evolved dramatically over the past five decades,” Gunja said. “And yet our classification system had not kept pace. It was time to think differently.” 

Why a Redesign? The Problem with Legacy Classifications 

In its original form, the Carnegie Classification was intended as a descriptive tool for researchers—a way to make sense of institutional variation and to support peer comparisons. But as Gunja explained, the system became a victim of its own success. Over time, the classifications were adopted by third parties for purposes never intended: performance funding, faculty pay differentials, prestige hierarchies, and—perhaps most consequentially—national rankings. 

Most concerning to the team was the way institutions were grouped. In the legacy “basic classification,” an institution that awarded 75 master’s degrees and 7,500 bachelor’s degrees would be classified as a master’s institution. While that logic may have worked in 1973, it no longer reflected how institutions operate—or how students experience those institutions today. 

As the classification categories became further decoupled from on-the-ground reality, their influence grew. “We started to worry that we were no longer grouping like institutions together,” said Gunja. “And that has real-world consequences.” 

A Three-Part Framework for 2025 and Beyond 

The 2025 redesign introduces a new three-part framework to replace the single “basic” classification: 

1. Research Activity Designation 

Most familiar to the field as the R1 and R2 classifications, the research activity designation has been simplified and clarified. In the past, the process of determining whether an institution was R1 or R2 was opaque, relying on weighted indices and unclear thresholds. 

In the new system, for the 2025 release: 

• R1 institutions must report ≥ $50M in annual research expenditures and ≥ 70 research doctorates. 
• R2 institutions must report ≥ $5M in research expenditures and ≥ 20 research doctorates.
• A new category, Research Colleges and Universities, includes institutions with ≥ $2.5M in expenditures, even if they do not award many or any research doctorates.

This threshold-based model enhances transparency and recognizes important research activity at non-doctoral institutions, expanding the scope of acknowledged research excellence. 

2. Institutional Classification (formerly "Basic") 

Sara Gast detailed how the new institutional classification moves beyond the outdated highest-degree logic to a multidimensional model built on three core dimensions: 

• Degree Award Focus: capturing the mix of undergraduate and graduate awards 
• Fields of Study: reflecting the academic programs that most (≥50% of students) study 
• Size: measured via 12-month unduplicated headcount, rather than traditional fall FTE 

These elements produce 31 new groupings, with deliberately descriptive (and intentionally awkward) naming conventions. The aim: resist implied hierarchies and create peer groups that reflect students’ actual experiences. 

“We’ve made the labels more descriptive and less normative,” Gast noted. “R1s took off not because of institutional quality, but because of the prestige attached to a simple name. That’s exactly what we’re trying to avoid.” 

She emphasized that while some legacy labels remain, their definitions have changed. For instance, an institution may still be called a “baccalaureate college,” but the designation now considers the overall student experience, not just the number of degrees conferred at the bachelor’s level. 

3. Student Access and Earnings Classification 

Perhaps the most innovative component of the 2025 redesign is the entirely new Student Access and Earnings Classification, presented by Kyle Whitman. It introduces a dual-metric evaluation: 

• Access: Are Pell recipients and underrepresented minority students equitably represented, relative to the demographics of the communities an institution serves? 
• Earnings: How do students’ post-attendance earnings (eight years after entry) compare to similar peers in the labor market?

Each institution receives a score on both metrics, scaled relative to 1.0: 

• Above 1.0 = outperforming expectations 
• At 1.0 = meeting expectations
• Below 1.0 = underperforming

This produces a 2x3 matrix with six resulting classifications. The most celebrated category, Opportunity Colleges and Universities, reflects institutions providing both high access and high earnings outcomes. Of the 3,055 institutions analyzed, 479 (16%) earned this designation. 

Thresholds for earnings scores vary by institutional type—for example, associate-degree institutions are evaluated differently than institutions that award a baccalaureate or higher degree to reflect labor market realities. Whitman noted that 76% of institutions scored above 1.0 on access, and 81% above 1.0 on earnings, suggesting that higher education, on average, is performing well across these dimensions. 

Supporting Use and Interpretation 

Throughout the session, the presenters emphasized transparency and engagement. All classification data and methodologies are publicly available at carnegieclassifications.acenet.edu. Institutions can explore: 

• Their placement within the three classification systems 
• Side-by-side comparisons with similar peers
• Visualizations and underlying data
• Detailed manuals explaining definitions and thresholds

Gunja also described the extensive consultation process: multiple technical review panels, data experts, and a roundtable of institutional presidents were involved in shaping the redesign. Former Carnegie steward Vic Borden was also credited for his foundational work and advisory role.

Implications and Next Steps 

The reimagined classifications are designed not just to describe institutions, but to support more accurate research, benchmarking, and policy evaluation. With broader, more nuanced peer groupings, institutions can now better contextualize their student outcomes and educational missions. 

Still, the team acknowledged that this is the beginning—not the end—of a new era. They invited feedback and committed to regular updates and educational sessions to help stakeholders make sense of the changes. 

“We’re not trying to redefine what’s valuable,” Gunja concluded. “We’re trying to reflect what’s real. And to do that well, we need the higher ed community to engage with the data—and help us keep improving.”

2026 AIR Forum

Mark your calendars to join us May 26–29 in Washington, D.C., for the 2026 AIR Forum! Pre-conference educational opportunities will be offered May 25–26 for an additional fee. 

With more than 200 crowd-sourced sessions by colleagues and thought leaders from around the globe, AIR Forum is the “must-attend” event for higher education professionals who build and support data-informed decision cultures.