Study 3
Shrinking colleges
How many colleges are smaller than they were ten years ago, by how much, and which kinds grew instead?
down 10%+ from ten years ago
0%51%
Fall 2014 → fall 2024 · share of colleges
Up 10%+ from ten years ago
0%23%
Fall 2014 → fall 2024 · share of colleges
Median college's change
0%-11%
Fall 2014 → fall 2024 · median college
The national picture
51% of 1,537 colleges have at least 10% fewer undergraduates than in fall 2014. The share that was smaller than fall 2014 each year since, and how the ten-year change is distributed.
64% in fall 2024; it peaked during the pandemic.
Where it changed
The same measures split by group, on one scale. Switch to Students to weight by how many undergraduates each group has, or to the five-year window to isolate the pandemic.
fall 2014 to fall 2024. Share of colleges that shrank (left) or grew (right) by 10+ points, fall 2014 to fall 2024.
- Shrank 10%+
- Grew 10%+
Northeast
419 colleges
Shrank 49% · Grew 20%
Southeast
388 colleges
Shrank 46% · Grew 29%
Midwest
392 colleges
Shrank 62% · Grew 15%
Southwest
119 colleges
Shrank 41% · Grew 35%
West
187 colleges
Shrank 42% · Grew 30%
Territories
32 colleges
Shrank 91% · Grew 0%
Did they lose applicants, or enroll fewer of them?
At the 784 colleges that shrank 10%+ over the ten-year window: median and median , each fall.
What it means
Across 1,537 colleges, 51% have at least 10% fewer undergraduates than in fall 2014, making a decline the median experience (11% smaller at the typical college), even though the total panel is about the same size (1% smaller: students concentrated rather than disappeared.
The decline is concentrated: in the Midwest, 62% of colleges shrank 10%+, and the region's total undergraduates fell 13%. Puerto Rico's colleges lost the most of any region, 39% of their students in total.
Growth went the other way: R1 universities' median college grew 10%, while for-profit colleges grew 8% at the median but shrank 5% in total — a few large closures against many small gains, the reason this page shows colleges and students separately.
Some of that growth is online, not campus: among 10,000+ colleges, the campus-based cut (198 of 213) trims the total change from 10% to 7%. For-profit colleges drop out of the campus-based view entirely, by definition.
These breakdowns show where undergraduate counts changed, not why; a college shrinking could mean fewer applicants, lower yield, or less retention.
How this was measured
- Which colleges. Colleges with 300 or more undergraduates in both the window's first and last fall: 1,537 colleges in fall 2014 and fall 2024. It's a , so colleges entering or leaving the data don't look like change.
- Groups. Colleges are grouped by . A group needs 30 or more colleges in the panel; smaller ones show .
- Shrank / grew. A change of 10 or more percentage points in undergraduates, either way, over the window.
- 300-undergraduate floor. A college needs 300+ undergraduates at both ends of a window to join its panel (the same bar as the site's diversity-change indicator): below it, a handful of students can swing a percentage a lot. A college that fell below 2,000 undergraduates counts as "under 2,000" at both ends of the size grouping.
- Two windows. The ten-year window (fall 2014 to fall 2024) is the default; the five-year window (fall 2019 to fall 2024, 1,552 colleges) isolates the pandemic drop and the partial recovery since. Each has its own fixed panel.
- The distribution chart clamps changes beyond ±60% into the end bins, so a few extreme closures or mergers don't stretch the scale.
- The companion chart covers the 784 colleges that shrank 10%+ over the ten-year window (a sub-panel), median applicants and enrolled first-years reported each fall.
- "Campus-based only." Leaves out for-profits, colleges where more than half of undergraduates took every course online at either end of the window, and campuses that merged or split; the headline figures above are all colleges.
- The universe. The site's four-year colleges, not all of U.S. higher education. Breakdowns show where a change happened, not why.
Sources:College Scorecard (enrollment), Fall 2014 to Fall 2024;IPEDS Admissions survey, Fall 2014 to Fall 2024About the data(built 2026-10-12; CPI retrieved 2026-10-12)