Carnegie Classification Methodology Change: How R1 Was Rewritten, and Which Schools Moved

How many schools gained R1 in 2025, and how many lost it?

These four figures are computed from the table below every time you change a filter, so the numbers on this page cannot drift away from the list they describe. With no filter applied they describe the whole country.

Which schools gained R1, R2 or RCU in 2025?

Every institution whose research designation is different in 2025 from what the 2021 Basic Classification gave it: 301 rows. Sort any column, and filter by direction or state. The two right-hand columns are the numbers the thresholds are actually tested against, taken from Carnegie's own 2025 public data file.

Spending and doctorate figures are the qualifying values: the higher of the FY2021–FY2023 three-year average or the single most recent year, which is the rule Carnegie applied. Spending is shown in millions, or with a B suffix for billions. Without JavaScript the full table still renders, sorted by state.

How did the 2021 Carnegie Classification method decide R1?

The rule that changed: from 2005 through 2021 there was no published number to clear. To be considered at all, an institution had to award at least 20 research doctorates and, from 2015, spend at least $5 million a year on research and development. Everything after that was relative.

Carnegie's published 2021 methodology fed ten variables into a principal components analysis: science and engineering R&D spending, non-science and engineering R&D spending, research staff with doctorates, and doctoral conferrals split across humanities, social science, STEM and other fields. That produced two indices, one for aggregate research activity and one per capita. Each institution's distance from a common reference point on that two-index plane decided which group it landed in, and raw values were converted to rank scores first so outliers would not dominate.

The consequence is the one that mattered to institutions: the groups were built to be roughly equal in size, so the boundary between "Very High" and "High" moved every cycle and was not known until the analysis was finished. Carnegie's own fact sheet says the dividing line "changed each cycle and was not determined until the data was analyzed." A provost could raise spending by 20% and still slip, because everyone else had raised theirs too.

The left-hand chart above shows what that looks like on the two numbers everyone assumed mattered. Plot the 2021 doctoral universities by spending and doctorates and the R1 and R2 dots sit on top of each other. That is not an error in the chart; it is the honest picture of a ten-variable index projected onto two axes.

How does the 2025 R1 formula work?

The rule that changed: the index is gone, replaced by two absolute thresholds applied identically to every institution in the country.

Three details decide borderline cases, and they are where most confusion comes from.

1. Both tests must pass. R1 and R2 are conjunctions, not scores. Clearing one line and missing the other puts you in the tier below, which is exactly what happened in Huntsville.
2. The measurement is the higher of two readings. Carnegie used the better of the FY2021–FY2023 three-year average or the single most recent year. William & Mary awarded 60, 84 and 66 research doctorates across the three years. Its single most recent year is 66, below the line; its three-year average is exactly 70.0, on the line. It is R1 on the average. The University of Maine tells the same story: 65, 84 and 62, a single year of 62 and an average of 70.33.
3. RCU is not ranked against R1 and R2. Institutions already designated R1 or R2 are excluded from RCU, so RCU picks up the research spenders that award few or no doctorates, including master's institutions, baccalaureate colleges and two Tribal Colleges and Universities.

Data sources are fixed too: R&D spending from the National Science Foundation HERD Survey for FY2021, FY2022 and FY2023, and research doctorates from the IPEDS Completions Survey for 2020–21, 2021–22 and 2022–23. Carnegie says the thresholds may be revised for the next release, scheduled for spring 2028.

Why did the 2025 Carnegie Classification changes grow R1 by 28%?

The rule that changed: a fixed line has no capacity limit, and a rank-ordered club does.

Under the old method the size of the R1 group was set by the method itself, because the goal was two groups of roughly equal size. If federal research funding grew across the whole sector, the boundary simply moved right and the same number of institutions stayed inside. Under the 2025 method the $50 million line stays where it is while the sector's spending grows, so every institution that crosses it is added and nobody is displaced to make room.

The 42 newcomers show what that looks like in practice. Their qualifying spending runs from $50.2 million at the low end to $2.05 billion at the high end, with a median of $97.9 million. Only 7 of the 42 sit below $60 million, so this is not a crowd of institutions squeaking over the line by a rounding error. Most of them were already spending well past $50 million in 2021 and were kept out by where the relative boundary happened to fall that year.

The composition tells the second half of the story. 32 of the 42 were R2 in the 2021 edition, which is the movement people expect. The other 10 were not in the doctoral ranking at all: they were classified as "Special Focus Four-Year: Research Institution", a category that captured medical and health science centres such as Baylor College of Medicine, the University of California San Francisco and Weill Cornell Medicine. The 2025 framework separates research designations from institution type, so a specialised medical centre is now measured on the same two numbers as a flagship public university.

What happened to R3, and is RCU the replacement?

The rule that changed: the three-level research vocabulary was retired seven years before most people noticed.

"R3" meant Doctoral Universities: Moderate Research Activity, and the last edition that used it was 2015, which had 115 R1, 107 R2 and 112 R3 institutions. In the 2018 edition Carnegie dropped it and created Doctoral/Professional Universities, a category defined by the degrees an institution awards rather than by how much research it does. By 2021 that category held 187 institutions, and there were only two research-activity levels left. If a school's website still calls itself R3, the label has been out of date since 2018.

RCU, introduced in 2025, is not a renamed R3. R3 was a doctoral category assigned by a research index; RCU is a spending category that explicitly serves institutions that "do not offer many or any doctoral degrees", including baccalaureate colleges and Tribal Colleges and Universities. Of the 216 RCU institutions, most were master's or baccalaureate institutions in 2021 with no research designation of any kind. Three former 2021 R2 universities moved to RCU rather than holding a numbered tier, which the table above lists under "other change".

What else changed in 2025 besides the research designations?

The rule that changed: one classification became three, and only one of them is about research.

The February 13, 2025 release covered the Research Activity Designations alone. The rest arrived on April 24, 2025: the Basic Classification was rebuilt as the Institutional Classification, which sorts institutions by award level focus, academic program mix and size into 31 groupings instead of by highest degree awarded, and a brand new Student Access and Earnings Classification was published alongside it.

The access and earnings framework asks two questions that have nothing to do with research volume. The access ratio compares an institution's Pell and underrepresented enrolment against the demographics of the places its students actually come from, using the top three states students migrate from. The earnings ratio compares median earnings eight years after entry, from the College Scorecard, against what people aged 22–40 with a high school diploma earn in those same places. A ratio of 1 means the institution matches its comparison area exactly.

Institutions with an access ratio of at least 1 and an earnings ratio of at least 1.5, or 1.25 for primarily associate colleges, are named Opportunity Colleges and Universities. ACE and the Carnegie Foundation designated 478 of them on April 24, 2025. If you are choosing between schools on outcomes rather than research volume, that list is the more relevant one, and it overlaps only partly with R1: plenty of Opportunity Colleges have no research designation at all.

Does a Carnegie Classification 2021 vs 2025 change affect your applications?

The rule that changed: nothing about the school. A designation released on February 13, 2025 was calculated from spending and degree data for FY2021 through FY2023. The faculty, the admit rate, the class sizes and the courses on the Monday after the announcement were the ones from the Friday before.

Carnegie states plainly that the classifications "do not consider elements like selectivity, admissions data, or other factors". So use the designation for what it measures and nothing else:

• It is a real signal about research volume. An R1 designation means at least $50 million a year of research and at least 70 doctorates. If you want laboratory work as an undergraduate, a graduate-heavy department, or a path into a PhD, that volume is a genuine input.
• It says nothing about teaching or attention. A campus with 300 doctoral students competing for lab slots is not automatically better for an undergraduate than one with 20 and a professor who answers email.
• It is not a selectivity ranking. Several R1 universities admit most applicants; several of the most selective colleges in the country have no research designation at all because they run no doctoral programs. If selectivity is what you are trying to read, look at admit rates and score bands instead, starting with what counts as a good SAT score for the schools on your list.
• Do not rebuild your list around a designation change. If a school was a good fit in January 2025 it is still a good fit. Add it to a structured list instead: the college list builder balances reach, target and likely schools on things that actually move an admission decision.

One thing a new designation does change is behaviour on the other side. Newly designated R1 institutions market the label hard, which means more email, more invitations and more campus events. That is worth tracking rather than ignoring, because at some schools it is scored: see how demonstrated interest is measured. And if you want a wider frame on why a school's public numbers move without the school moving, the same mechanics drive ranking jumps, covered in how college rankings actually work and what changed in the U.S. News methodology.

How do you check one school's designation yourself?

Three minutes, one official source, no interpretation needed.

1. Open the Carnegie Institution Search and type the campus name exactly as it appears in federal data. Branch campuses are listed separately, so "Pennsylvania State University-Main Campus" and "Penn State Harrisburg" are different records with different designations.
2. Read the line labelled Research Activity Designation. It shows one of R1, R2, Research Colleges and Universities, or no designation. Anything else you have read is either a 2021 label or marketing.
3. Scroll to the research panel for the spending and doctorate figures. Carnegie notes on every institution page that the displayed figures are 2022–23 single-year data, while some institutions were classified on their three-year average, so a school can show fewer than 70 doctorates and still be correctly R1. William & Mary is the clean example.
4. If you need the exact numbers the designation was tested against, download the 2025 Public Data File from Carnegie's resources page. It carries herd_2021, herd_2022, herd_2023 and rdoc_2021 through rdoc_2023 for all 3,927 institutions, which is where every figure in the table on this page comes from.
5. Write down the date you looked. Designations are static between releases, and the next one is scheduled for spring 2028, but records are edited between releases when an institution appeals or changes its name.

One honest caveat from working through the file: it is a published dataset, not a formula you re-run. The University of Texas at Tyler is designated R2 while its published spending figures top out at $3.5 million, below the $5 million line, and the public file records no appeal flag for it. Read the record for what a school is, rather than recomputing the record and arguing with it.

Two related pages on this site go deeper on the label itself: what R1, R2 and RCU actually mean, and the full 2025 R1 list broken out by state. If you are checking whether a school is legitimate rather than how much research it does, that is a different lookup entirely: how to check a college's accreditation.

Questions students ask about the 2025 Carnegie Classification changes