· 4 min read · worthmydegree.com
What a college ROI ranking cannot tell you
Every year a new ranking says which colleges are the best investment. The San Francisco Chronicle's interactive is a good one, built on Georgetown University's ROI analysis of the federal College Scorecard: it reports, for each school, the median earnings of students ten years after they enrolled, minus the average net price they paid. By that measure Caltech leads California at $627,000 and a typical UC or CSU returns about $247,000 over ten years.
Those are real numbers, honestly computed, and worth a look. They are also answering a different question than the one you are asking. A ranking asks which schools' past students earned the most. You are asking whether one specific plan, your major at your school on your loan in your city, will be worth it.
The bigger names measure even less of that question, just in different ways. U.S. News weighs earnings five years after graduation at 5% of its score, with another 5% for debt, and says a new earnings metric arrives in its 2027 edition. The Wall Street Journal's ranking went the other way, putting 70% of its score on student outcomes, led by salary impact and the years needed to pay off the net price. Times Higher Education and QS rank reputation and research and measure earnings hardly at all. Wherever a ranking sits on that spectrum, the same five gaps apply, because each gap is a fact about you rather than about a school. Here they are, and where each one gets answered instead.
Money figures below from this site's own federal wage file, the May 2025 release, are rounded to the nearest thousand dollars; figures from the Chronicle and Georgetown are quoted as published.
It pools every major into one number
A school's ROI is one median across everyone who enrolled, whatever they studied. But among the 177 careers that typically require a bachelor's degree, typical pay runs from about $42,000 to about $232,000, more than five times end to end. An engineering graduate and a social work graduate from the same campus are one data point each in the same median, and the ranking cannot tell you which one you will be. That is most of why technical schools top every list of this kind.
It does not know where you will work
Registered Nurses typically earn about $187,000 in San Francisco and about $84,000 in Columbus, Ohio. Same degree, same license, more than double the pay, and very different rent. The Chronicle's own methodology note says it plainly: none of these data control for cost of living. A school whose graduates cluster in expensive coastal cities will look better than one whose graduates stay in cheaper places, without either school doing anything differently.
It leaves the debt out entirely
Georgetown's methodology excludes student debt from the calculation. The ranking subtracts what students paid out of pocket, not what they borrowed or what the borrowing cost them. Two students with identical earnings and identical net prices get the same ROI whether one graduated debt-free and the other owes $60,000 on a plan that runs thirty years. For most families the debt is the decision, and the 2026 repayment rules changed it substantially.
It measures who enrolls as much as what the school adds
The earnings behind the number describe students who received federal aid, a bit over half of all students, and they count everyone who enrolled, including those who never finished. More importantly, schools that admit students who were already likely to earn a lot will show high earnings whether or not the school caused any of it. A ranking cannot separate what a college adds from who it selects, and the most famous names select hardest.
There is no compared to what
The ranking's ROI is earnings minus price, against nothing. It does not ask what the same person would have earned without the degree, which is the only comparison that makes worth it mean something. A path can clear a ranking's bar and still leave you behind someone who skipped the degree, kept working, and never made a loan payment.
Price your own plan instead
Each of those five gaps is a question about you rather than about a school, which is why a ranking cannot answer them and a calculator built around your own plan can. Pick your major or career, name your school, and it prices the actual loan under the actual 2026 federal repayment plans, adjusts pay and costs for your city, and measures the whole thing against the person who went straight to work: the compared-to-what the rankings skip. If you do not have a career in mind yet, start from your Interest Profiler results and price the list it gave you.
Use the rankings for what they are good at: a rough national map, and a reminder that college outcomes differ enormously. Then stop reading about other people's medians and run your own numbers.