The latest decision by the U.S. Board of Education and the Carnegie Foundation simplifies the classification of research universities that might be good to make sense, but it represents a serious mistake and the result is the result.
By reducing the comprehensive research indicator system to just two (in order to obtain the coveted R-1 status), the institution must spend $50 million per year on research and award 70 research doctorates annually – ACE has fundamentally changed the meaning of being a top research institution. From a more comprehensive assessment of research activities, this shift may distort the public’s understanding and perception of university excellence while motivating behavior, thus undermining the creativity and innovation of long-term research.
To most effectively appreciate the importance of this change, it is helpful to track the history of Carnegie classification system. The system was originally conceived by the Carnegie Higher Education Commission in 1973 to support research and decision-making by categorizing American universities and universities based on their tasks and outputs.
Over the past few decades, classification systems have become a trusted outline for the public, media and higher education communities. Names such as “R-1” (historically stand for “Doctoral University – Very High Research Activities”) and “R-2” (“Doctoral University – Advanced Research Activities”) have prominent features, indicating a good level of academic productivity, research funding, doctoral education and infrastructure.
The approach for the 2021 classification (until this year’s latest classification) involves a set of metrics designed to quantify research excellence and partially normalizes the size of the organization. These include total research expenditures in science and engineering, research expenditures in non-science and engineering fields, the size of science and engineering researchers (postdoctoral appointees and other non-faculty doctoral researchers), and the number of doctoral degrees awarded annually in humanities, social sciences, STEM, STEM and other fields, etc.
The primary component analysis then allows the creation of an index representing the sum of per capita research activities, thereby allowing for a careful and fair comparison between different institutions. In many ways, this approach is one of the most comprehensive and covered frameworks to date, providing statistical evaluations for American research universities built on publicly available data.
However, for the classification in 2025, the landscape has changed. Under the leadership of ACE, the Carnegie Foundation has developed a new framework that greatly simplifies the standards for meeting flagship research status. The revised criteria focus only on the two indicators mentioned above: institutions spend at least $50 million per year on research activities and award at least 70 research doctorates each year. Both institutions that meet the criteria are R-1; those who are not eligible but spend at least $5 million on research activities, and at least 20 research doctors are R-2. Now, these terms represent very high “spension and PhD production”, respectively, rather than the high and high “research activities” used previously.
This change may seem technical, but it eliminates many subtle measures of academic participation and output and represents a profound change in values. In the previous activity-based framework, institutions were rewarded for establishing a diverse research ecosystem in various disciplines. Now, the indicator has been reduced to the total amount spent and the degree, and these indicators and outputs do not necessarily equate to research excellence.
In addition, this move opens the door for institutions to “conduct tests and teach”. Instead of pursuing organic research tasks, universities may make tactical investments to achieve the magical numbers required to achieve the R-1 state. This situation is a textbook case for Guha’s Law: “When a measure becomes a target, it is no longer a good measure.”
By selecting only two indicators to evaluate the status of the country, the classification system invites institutions to play standards, not necessarily to improve research expenditures and degree outputs through improved research performance, but rather administrative and accounting shifts. This oversimplification of complex and holistic assessment tools can have unintended consequences, such as distorting institutional priorities and killing the motivation to invest in long-term, mission-driven scholarships.
Unfortunately, evidence of this phenomenon can already be seen. A rough search of the Internet will reveal several universities that recently announced an increase in R-1 status: More than 40 new institutions have obtained R-1 status based on revised standards. Despite many people’s commendable progress, it is worth noting that the rise in their “elite” research status is not due to a significant shift in academic output, but because they meet two quantitative benchmarks.
It is not that these institutions should not be proud of their growth, but that the public will now have a parity between these universities and other research footprints, which are more influential and globally. ACE effectively redefines the meaning of being a “R-1” institution without explicitly conveying that the designation no longer reflects the same type of achievement that has been done.
To prevent confusion and preserve the integrity of the classification system, the ACE and the Carnegie Foundation should consider renaming the new category to reflect its true nature. A more accurate tagging system might be RS-1 and RS-2, rather than continuing to use historically meaningful terms “R-1” and “R-2”, but rather means “research spending.” This small change will allow stakeholders to clarify that these categories are now based primarily on spending thresholds rather than overall measurements of research activities.
While simplification may make classification more politically attractive and manageable, it comes at the expense of important components such as analytical synthesis, contextual responsiveness, and assessment accuracy. To properly recognize and support a true center of research excellence, a multidimensional assessment framework must be adopted – ideally covering not only research expenditures and productivity in PhD programs, but also integrating measures of academic impact, quality of research publications, quality of research publications, development of research infrastructure, and the degree of research-based research activities.
Furthermore, to balance the structural advantages of larger institutions, appropriate normalization factors may be considered, such as cost per faculty, per capita cost, and PhD in each research active department. By contrast, the 2021 classification model could better reflect this comprehensive and fair approach, compared to, in 20225, by comparison.
To use the integrity of American research universities as engines of discovery and innovation, their assessment should be based on objective academic indicators that meaningfully reflect the institutional excellence of the research. Given the multifaceted nature of excellent research, our classification system should be equally subtle and comprehensive.
