Search "which colleges send the most students to top medical schools" and you'll find confident-looking ranked lists -- Harvard undergrad sends X students to elite med schools, Stanford sends Y. Here's what's actually behind those numbers, and what real data exists instead.

Where the "elite feeder list" numbers actually come from

The rankings that circulate -- and get repeated, secondhand, across admissions-consulting blogs -- trace back to a single source: a dataset built by scraping LinkedIn profile data for people whose "education" field lists both an undergraduate school and one of a defined set of "elite" medical schools, matched against IPEDS enrollment figures for a per-capita cut. It's undated in terms of when the underlying scrape was run, doesn't disclose how it handles duplicate or incomplete LinkedIn entries, and can't distinguish "enrolled" from "graduated" -- LinkedIn users self-report their own education history, with no verification against actual matriculation records. Every other site repeating a "top feeder to elite med schools" ranking is citing this same dataset, once removed, not an independent count.

That's not a source AAMC, a medical school, or an undergraduate institution stands behind -- it's a self-reported social-media scrape presented with ranking-table confidence.

What top medical schools actually publish about their entering class

Real, official numbers exist -- they just don't answer "which schools send the most." Elite medical schools' own class-profile pages report how many different undergraduate institutions are represented, as a breadth statistic, not a per-school breakdown:

  • University of Pennsylvania (Perelman): 67 undergraduate schools represented in the entering class.
  • University of Chicago (Pritzker): 50 different colleges and universities.
  • University of Miami (Miller): 68 undergraduate institutions.
  • Columbia (Vagelos): 60-63 colleges across recent entering classes.
  • Duke: 54 universities in a recent entering class.
  • Yale: roughly 81-86 undergraduate institutions per class in recent years.

Read that pattern for what it actually shows: entering classes at these schools are drawn from dozens of different colleges each year, not concentrated in a small, named group of "feeders." No official page names which schools contributed how many students -- these institutions report breadth, not a ranked source list.

The real AAMC data -- and it answers a different, more interesting question

AAMC does track undergraduate-origin patterns, but at a different cut than "which schools feed elite programs." Its Feeder/Nonfeeder Data Snapshot defines a "feeder school" as one of the 50 undergraduate institutions supplying the most matriculants to all U.S. MD-granting medical schools combined, each year -- not specifically to top-20 or elite-ranked programs. (Our companion piece on medical school acceptance-rate marketing covers AAMC's related Table A-2, the named list of schools by applicant volume -- this Snapshot uses matriculant volume instead, but doesn't republish school names itself.)

The real findings are worth knowing on their own terms:

  • Almost half of all medical school matriculants nationally -- 49.7% in the most recent year -- came from just 50 undergraduate schools, out of roughly 900 undergraduate institutions that supply any matriculants at all. 39 of those 50 schools were feeder schools in every single year of AAMC's 10-year study window.
  • In the most recent year, 29 of the 50 feeder schools were public, supplying 7,369 matriculants; 21 were private, supplying 4,044. Public feeder schools aren't a small share of the picture -- they supply nearly two-thirds of feeder-school matriculants.
  • Feeder-school matriculants are meaningfully less socioeconomically diverse than nonfeeder-school matriculants. In the most recent year, 40.6% of nonfeeder-school matriculants reported parental income under $100,000, versus 35.9% from feeder schools; 15.4% of feeder-school matriculants reported parental income of $250,000+, versus 12.0% from nonfeeder schools. Nonfeeder-school matriculants were also more likely to be first-generation students, Pell Grant recipients, from a medically underserved area growing up, or born in a rural county.
  • Once admitted, feeder-school and nonfeeder-school matriculants graduate medical school at comparable rates -- between roughly 95% and 97% at six years for both groups across every cohort AAMC studied, with nonfeeder schools running very slightly ahead on four-year graduation rates specifically.

That last point is the one worth sitting with: where a matriculant went to undergrad doesn't meaningfully predict whether they finish medical school. The measurable difference between feeder and nonfeeder schools is in who gets in -- a socioeconomic access pattern -- not in who succeeds once admitted.

What this means for you

  • Don't plan your undergraduate choice around an "elite med school feeder" ranking. The ones you'll find online trace to an unverified LinkedIn scrape, not official data from AAMC or any medical school.
  • A high volume of matriculants from a given undergraduate school reflects size and applicant pool as much as anything else -- the same caution our piece on general pre-med acceptance-rate claims makes about "our students get in at 85%" marketing applies here too.
  • If you're weighing a large public flagship against a smaller private school for pre-med, AAMC's real data cuts against assuming the flagship is a lesser path -- public schools supply the majority of feeder-school matriculants, and matriculants from nonfeeder schools (many of them public or less-selective) graduate medical school at rates statistically indistinguishable from feeder-school matriculants.
  • Undergraduate prestige correlates with getting in, not with succeeding once you're there. Once you're admitted, AAMC's own outcome data shows feeder-school origin stops predicting anything that matters. What actually moves the acceptance-rate needle is your GPA and MCAT score, not where you went undergrad -- see AAMC's real, published MCAT/GPA acceptance-rate grid for exactly how much each one matters.

Sources