Methodology
How the numbers are made.
What PSEO is
The Census Bureau's Post-Secondary Employment Outcomes (PSEO) program matches actual graduate transcript records to national wage records — not a survey, and not self-reported. The data is officially labeled experimental by Census.
What the percentages mean
Each number is the share of graduates found in wage records whose main job is located in Virginia. Roughly a quarter of graduates at year 10 aren't matched (self-employed, out of the workforce, some employer types not covered by unemployment insurance) — they're excluded from both the numerator and the denominator, not counted as "left."
Job location, not home address
A graduate living in Arlington but commuting to a DC office counts as out-of-state. This overstates leakage for schools near the DC and North Carolina borders.
Scope
Baccalaureate (bachelor's) degrees only, all fields, all graduation cohorts unless a chart says otherwise. Year-10 figures only exist for cohorts graduating by ~2012; more recent cohorts haven't reached the 10-year horizon yet within available census data.
Weren't they from out of state to begin with?
Partly, and it doesn't change the answer. Virginia law requires UVA to keep two-thirds of its students Virginia residents; William & Mary runs a similar balance, and most Virginia publics enroll far more in-state than that.
So do the math. If two out of three UVA students are Virginians, but only 34% of graduates are working in Virginia ten years later, then even if every single graduate who stayed were a Virginia native, at least half of the in-state students are gone. Out-of-state students returning home cannot explain a retention rate that sits 30 points below the in-state enrollment share. At most schools in this explorer, the leak is mostly Virginians leaving.
Where this logic doesn't apply: private schools with national draw, like Washington and Lee or Richmond, where most students were never Virginians. Their low retention reflects recruiting footprint, and this explorer flags them as such.
Caveat: enrollment residency is measured at entry and averaged across recent years, while retention spans cohorts back to 2001 — treat the comparison as directional, which is all the argument requires.
Context: national-draw private schools
Institutions like Washington and Lee or the University of Richmond naturally retain fewer graduates in-state — they recruit nationally, so many students weren't Virginians to begin with. Low retention reflects recruiting footprint, not institutional failure.