Do college majors actually matter for pay?

Start with the gap that is real. Among the 162 US bachelor's majors with enough graduates to report, the highest median earnings belong to Pharmacy, Pharmaceutical Sciences, and Administration at $116,539 four years after graduating. The lowest belong to Drama/Theatre Arts and Stagecraft at $39,775. That is a 2.9x difference, and it is not an artifact of picking midpoints: pharmacy graduates at the 25th percentile ($78,609) still out-earn drama graduates at the 75th percentile ($55,399). Those two groups do not overlap at all.

Now the part the rankings leave out. Line all 162 majors up by median earnings and compare each side-by-side pair — 1st against 2nd, 2nd against 3rd, all the way down to 161st against 162nd. Every one of those 161 pairs overlaps in the middle 50% of earnings. Not a single major in the entire ranking is cleanly separated from the major ranked directly next to it.

Both facts come from the same dataset. The ranking is telling you something true at the extremes and close to nothing between neighbours.

A ranking implicitly claims that the difference between majors is the thing worth knowing. So measure it against the difference inside majors.

The typical major's middle 50% spans $36,830 — that is the distance from the 25th to the 75th percentile of its own graduates. Now take any two of the 162 majors and ask how far apart their medians are. Across all 13,041 possible pairs, the typical gap is $15,398.

The spread inside one major is more than twice the gap between two of them. And this is not one cherry-picked pair: in 80.3% of all 13,041 pairs, each of the two majors has a middle-50% range wider than the entire gap between their medians. It is easy to over-read that: it means that for most pairs of majors, which of the two someone picked tells you little about where they personally land in pay — not that the gap between majors is small or safe to ignore. That gap is an average, and an average this consistent is one of the most reliable earnings signals there is. Both are true at once: the typical gap between two majors is real, and it is still smaller than the spread among the individuals inside either one.

The extreme case makes it concrete. Across all 162 majors, the distance from the highest median ($116,539) to the lowest ($39,775) is $76,764 — the entire span a "highest-paying majors" ranking is selling. Inside computer science alone, the distance from the 25th percentile ($74,830) to the 75th percentile ($152,038) is $77,208. Two computer science graduates, both sitting squarely in the ordinary middle of their own major, can be further apart in pay than the highest-paying major is from the lowest-paying one.

Computer science is an outlier and it would be dishonest to imply otherwise: only 2 of the 162 majors have a middle-50% range wider than the whole median span (computer science, and cognitive science at $77,068). It is the vivid illustration, not the evidence. The evidence is the 80.3%.

Overlap sounds like a statistical technicality. It is actually the answer to the question you came here with.

A typical major's middle 50% overlaps with 146 of the other 161 majors. Where two majors overlap, plenty of graduates from the lower-ranked one out-earn plenty of graduates from the higher-ranked one — not as a rare exception, but as an everyday outcome.

That is why our comparison tool always reports whether the two majors you picked overlap. If they do, the ranking between them is not the signal you want. If they do not — which happens toward the extremes — the gap is worth taking seriously.

The honest one-line summary: a major shifts the odds. It does not set a salary.

Every number above comes from the U.S. Department of Education's College Scorecard field-of-study data, which is public domain. These are the Department's own national figures for bachelor's degrees, not our re-aggregation. But who they cover matters as much as what they say.

These figures cover only graduates who received federal student aid — they come from federal aid records, not a survey of everyone. In 2015–16, 54.5% of all U.S. undergraduates received aid; at the four-year colleges that grant bachelor’s degrees the rate ran higher (59.8% public, 64.4% private, 77.7% for-profit), so roughly six in ten of these graduates are counted. They are not the medians of every graduate. And because the rate differs by school type while majors differ in their mix of schools, two majors are not measured on exactly the same slice.

"Working and not enrolled" narrows it further by excluding graduates still in school, so majors that commonly lead to graduate degrees (biology, psychology) show only the graduates who went straight to work.

The cohort is 2017–18 and 2018–19 graduates, with earnings measured in 2022 and 2023 and reported in 2024 dollars. It is a snapshot of those years, not today’s pay.

None of that makes the data useless — it makes it specific. This is what federally aided bachelor's graduates from 2017–19 were earning four years out. That is a real, large, honestly measured group. It simply is not everyone, and no page, including this one, should tell you it is.

Rankings invite you to read down a list. The data supports a narrower question: what is the gap between the two specific majors I am choosing between?

Pick the two, check whether their middle 50% ranges overlap, and read the gap accordingly. If they overlap — and for most pairs they will — the difference between the majors is smaller than the differences among the people inside each one. If they do not overlap, you have found one of the few comparisons where the ranking earns its keep.

Either way the number is one input. It says nothing about what you would be good at, what you would stay with, or what the job market will look like when you graduate. It is not admissions, career, or financial advice.