Every acceptance rate is a fraction: admitted over applied. In 2022, the top and bottom of that fraction quietly left the federal record. The percentage stayed.
We checked the federal series for all 571 universities in our catalog that have ever published applicant and admit counts. The last year each school reported them:
| Last year with counts | Universities |
|---|---|
| 2022 | 569 |
| 2021 | 1 |
| 2020 | 1 |
| 2023 or 2024 | 0 |
Zero. Not one university in the set has applicant or admit counts in the federal series for 2023 or 2024. The acceptance rates for those years kept arriving on schedule. The counts behind them went dark, everywhere, at once — which tells you this was a change in what gets collected, not 569 separate decisions.
What a rate hides when the counts disappear
With counts, "the rate fell from 12% to 9%" is a story you can interrogate: did applications surge, or did the class shrink? Those are opposite stories for an applicant. A demand surge means the school got more popular, not harder in absolute terms; a shrinking class means fewer seats for the same crowd.
Without counts, both stories collapse into one unexplained number. And scale disappears entirely. In the last counted year, the University of Minnesota-Twin Cities logged 38,050 applications and College of the Ozarks logged 2,239 — a 17-to-1 difference in how much evidence sits behind each school''s percentage. From 2023 on, the federal record renders both as a bare rate, as if a percentage built on two thousand data points and one built on forty thousand deserved the same confidence.
A percentage without a denominator is a claim, not a count.
Where the counts still live
They did not vanish from the earth — they retreated to a less convenient place. Many colleges still publish applicant and admit counts in their own Common Data Set filings, section C1, posted school by school as PDFs and spreadsheets. Coverage is uneven: some schools post every year, some skip years, some stopped filing publicly altogether. There is no longer one federal table where you can compare everyone at once. That is the actual loss of 2022 — not that the numbers ceased to exist, but that they ceased to be comparable.
How to read admissions data after 2022
Three habits cover most of the damage. First, treat any post-2022 acceptance rate as unaccompanied — if a site shows you a 2024 rate with applicant counts attached, ask where the counts came from, because the federal series does not have them. Second, when a school''s own Common Data Set is public, prefer it; it is the only remaining primary source for scale. Third, be suspicious of blanks that have been helpfully filled. A data source that shows N/A where the record ends is telling you the truth about the record; one that shows a smooth, complete table for 2024 is estimating and may not be saying so.
Sources
The counts and their cutoff come from the federal IPEDS admissions series, cross-checked against College Scorecard; per-school counts for the last reported year are as published. Common Data Set filings are each college''s own documents, linked from their pages in our catalog. Nothing above is estimated — including the zero.