Methodology

Where the data comes from

All figures come from the U.S. Department of Education College Scorecard: the institution-level file (May 19, 2025 release) and the field-of-study file (April 17, 2025 release). Scorecard itself draws on IPEDS, federal student aid records, and tax data. Airport distance is the only computed figure — a straight-line calculation from each campus’s IPEDS coordinates to the nearest FAA large-hub airport.

Which colleges are included

The universe is four-year, main-campus institutions awarding bachelor's degrees or higher (2,406 schools). Institutions must report latitude and longitude to be included at all. Branch campuses are excluded, so a state system appears once per main campus rather than once per site.

Missing data is never filled in

Colleges report unevenly. Acceptance rate is available for about two-thirds of institutions, and SAT averages for fewer than half, as test-optional admissions has spread. Where a college does not report the measure a list ranks on, it is left out of that list entirely. We do not estimate, impute, or substitute a similar figure. Each list states how many institutions reported the measure, so you can see how much of the field a ranking actually covers.

The Department of Education also suppresses figures derived from very small groups of students to protect privacy. Those cells arrive blank and are treated the same way — excluded, not guessed at.

Small-college adjustment

Outcome measures such as earnings, graduation rate and debt are medians over the students a college happens to enrol. At very small institutions those medians swing wildly year to year and would otherwise dominate every superlative list. Lists ranking outcomes therefore exclude institutions enrolling fewer than 500 undergraduates. Lists about enrolment size do not apply that floor, since it would defeat their purpose. The national and Virginia acceptance-rate lists use the same 500-undergraduate floor for consistency with those rankings — that size cut is not a median-stability adjustment for admission rate.

State-scoped lists

Every state (plus D.C.) has a full-set selectivity list ranked by published admission rate. A handful — California, Georgia, Massachusetts, and Virginia — also appear on the main lists hub; the rest are linked in context when you filter the rankings engine or open a school profile in that state. Coverage counts on those pages use that state’s institutions in the product universe, not the national 2,406. Geographic full-set slices are omitted from the derived “most ranked” meta tally so a regional roster cannot inflate that count.

Ties and ordering

Where colleges report identical values — many report admitting 100% of applicants — the order is resolved by a stated secondary rule and finally by institution ID, so the same data always produces the same list rather than shuffling between builds.

Earnings by major

Program-level lists use median earnings one year after completing, counting only graduates whose highest credential is that bachelor’s degree. The cohort is limited to students who received federal financial aid and who were working and not enrolled in further study during the measurement year. Where a college offers several programs in one field, the figure shown is the median across them.

Curated shortlists

A small set of lists — Resume Signal, the Ivy League, Little Ivies and Hidden Ivies — fix their membership up front. The Ivy League roster is the eight-school athletic conference, so nothing about who belongs is a judgment call. Little Ivies and Hidden Ivies are human-chosen and committed in the source tree (elite LACs vs. Ivy-adjacent private universities, with hard exclusions); within those sets, schools are ranked by admission rate like any other selectivity list. Resume Signal is different: both membership and the 1–100 order are editorial judgments about résumé brand recognition in elite hiring and selective graduate admissions. Scorecard admission rates appear only as context there. Overlap across curated lists is allowed. Curated lists are excluded from the derived “most ranked” meta tally so a fixed roster cannot inflate that count.

What these numbers do not tell you

Earnings figures reflect who a college admits as much as what it teaches — graduates’ field of study, family background and the regional job market all move these numbers, and none of that is controlled for here. Selectivity measures demand and class size, not teaching quality. Net price is an average across aided students and may look nothing like your own offer. Cost of attendance is a sticker figure almost nobody pays in full. Nothing here measures teaching, advising, safety, accessibility, or whether you would be happy at a place for four years.

Treat these lists as a starting point for questions, not as a verdict. Confirm anything that will affect a real decision with the institution’s own admissions and financial aid offices.

Corrections and rebuilds

Lists are regenerated whenever the underlying dataset is rebuilt, so rankings shift as the Department of Education publishes new cohorts. Each list is stamped with the date its data was compiled.