January 1, 1970

How College Rankings Work and What Actually Matters

University campus with ranking data overlay

In February 2022, a Columbia University math professor named Michael Thaddeus published a 21-page report accusing his own employer of submitting "inaccurate or misleading" data to U.S. News & World Report. Columbia had ranked #3 in the country. After the report went public and the university acknowledged errors, Columbia's ranking dropped to #12 — nine spots, overnight. The episode forces a question that millions of applicants almost never ask: if a top-ranked school can misrepresent its own numbers for years without anyone noticing, what exactly are these rankings measuring in the first place?

How the Rankings Machine Actually Works

U.S. News & World Report launched its college rankings in 1983 as a simple reader poll. By 2026, the methodology had grown into a 17-factor model evaluating more than 1,700 institutions. And U.S. News is just one of many — Princeton Review, Niche, Forbes, Washington Monthly, and The Wall Street Journal/College Pulse all run their own systems with their own priorities and their own definitions of "best."

For U.S. News, student outcomes carry the heaviest combined weight at roughly 40% of the total score. That category covers six-year graduation rates, how a school's graduation rate compares to its predicted rate (adjusted for incoming student wealth and test scores), first-year retention, and graduate debt and earnings data pulled from the federal College Scorecard.

After outcomes, the next biggest factor is peer assessment — a survey in which university presidents, provosts, and admissions deans rate other schools on a 1-to-5 scale. This alone accounts for about 20% of the score. If it strikes you as circular — elite schools rating each other highly because they've always thought of each other as elite — you're not wrong.

The remaining weight is split across faculty resources (class sizes, faculty salaries, percentage with terminal degrees), per-student financial spending, and measures of incoming student quality. Acceptance rate was quietly removed from the 2026 methodology, closing one obvious incentive to reject qualified applicants just to look more selective.

What Each Factor Actually Weighs

Here's how the major categories break down for National Universities in the 2026 rankings:

Factor	Approximate Weight
Graduation and retention rates	~22%
Outcomes (earnings, debt, graduation parity)	~17%
Peer assessment (reputation survey)	~20%
Faculty resources	~8%
Per-student financial spending	~10%
Student test scores and class rank	~7%
Graduate indebtedness	~5%

Two things stand out immediately. First, reputation makes up a fifth of the score — measured by asking rival administrators to assess one another. A school that was prestigious 40 years ago benefits from pure inertia even if its programs have quietly deteriorated. Second, per-student spending rewards schools for spending more money, which structurally advantages wealthy, endowment-heavy institutions.

A school can climb the rankings without improving a single classroom. Get richer. Admit students with higher test scores. Be better liked by peer administrators. That's not the same as being a better school for the students who actually enroll.

Gaming, Manipulation, and the Perverse Incentives

Columbia wasn't an outlier. In 1994, The Wall Street Journal first documented systematic data manipulation in college guides. The problem never went away. George Washington University was caught falsifying incoming students' high school class rank. Emory University admitted to misreporting high school GPAs. Claremont McKenna College submitted inflated SAT scores for six straight years before coming clean in 2012.

Those are the cases where someone got caught. The incentives to shade the truth were always there — and schools also learned to engineer their numbers entirely within the rules:

Reclassifying spending on administrator salaries boosted "financial resources per student"
Offering lavish merit scholarships to high-scoring applicants (who didn't need the money) kept test score averages up while crowding out lower-income students who needed aid
Reducing the share of part-time faculty improved faculty resource scores without changing who taught the actual classes
Rejecting transfer students kept freshman selectivity metrics clean, even when transfers would have contributed academically

The elephant in the room here is that rankings don't just measure schools — they shape them. When U.S. News decides to weight a factor, it becomes de facto education policy. Real spending priorities get reorganized around whatever the formula rewards this year.

When Elite Schools Said No

The boycott started with one law school and spread fast.

In November 2022, Yale Law School — ranked #1 by U.S. News every year for nearly three decades — announced it would stop cooperating with the rankings. Harvard Law followed the same day. Harvard Medical School withdrew in January 2023. Within months, dozens of other law schools had joined them.

Their objections were specific. Before the boycott forced reforms, U.S. News weighted LSAT scores at 11.25% and undergraduate GPA at 8.75% for law school rankings. High scores meant admitting students from wealthier backgrounds. Need-based financial aid was penalized because scholarship dollars given to low-income students raised a school's "indebtedness" metric. Public interest lawyers — public defenders, legal aid attorneys, nonprofit counsel — earn less than corporate associates, so schools with more of them got dinged on salary outcomes.

"The methodology disincentivizes programs that support public interest careers, champion need-based aid, and welcome working-class students into the profession." — Yale Law Dean Heather K. Gerken, November 2022

Harvard Law Dean John F. Manning added that the school "cannot reconcile our principles and commitments with the methodology and incentives the U.S. News rankings reflect."

U.S. News responded by pledging to keep ranking these schools using publicly available data — and it did. The rankings still come out, still get cited, still drive applications. But the boycott forced some genuine methodology changes: law school rankings now place less weight on LSAT scores and more on financial aid practices. That's a real outcome, even if incomplete.

What Rankings Don't Capture: The Mobility Question

Here's a number worth knowing. According to Third Way's 2023 Economic Mobility Index — which measures how quickly low-income students recoup their education costs through earnings gains — Harvard ranks #847 out of 1,320 bachelor's degree-granting institutions.

Not because Harvard graduates earn poorly. They earn extremely well. But Harvard admits so few low-income students that its impact on upward economic mobility is minimal at any meaningful scale. An institution graduating 1,800 undergraduates per year from mostly wealthy families moves far fewer people up the income ladder than a school serving 25,000 students from working-class backgrounds.

Schools that top the EMI look nothing like the U.S. News top 10. California State University campuses, City University of New York colleges, and several Historically Black Colleges and Universities dominate Third Way's list. Eight Cal State campuses appear in the top 20%. Five CUNY schools do too. CUNY's Accelerated Study in Associate Programs increased four-year graduation rates by 12 percentage points — one of the strongest documented intervention effects in U.S. higher education research.

Consider SUNY Stony Brook. It doesn't crack the U.S. News top 50. But economist Raj Chetty's Opportunity Insights research found it has one of the highest "mobility rates" in the country — more students who start in the bottom income quintile and end up in the top quintile than almost any other institution in the nation.

This isn't an argument against elite schools. It's an argument that "best college" is meaningless without answering: best for whom?

What Actually Matters When You're Choosing

If rankings can't answer that question for you, specific data can.

Net price is probably the most underused number in college research. Every school is required to publish a net price calculator. Run the numbers before you fall in love with a campus. A school with a $67,000 sticker price and strong need-based aid policies might cost less out of pocket than a lower-ranked school with a $29,000 tuition and minimal financial support. Students who start building their college list in spring of junior year have time to run the net price calculator across 15 or 20 schools before paying a single application fee — and that changes the entire calculus.

Completion rates matter more than most applicants realize. Look specifically at six-year graduation rates for students in your income bracket, not just overall institutional rates. A school with an 84% overall graduation rate and a 61% rate for Pell Grant recipients is telling you something very specific. The federal College Scorecard (collegescorecard.ed.gov) breaks this down by school.

Career outcomes in your specific field deserve a hard look. A regional state school with a direct pipeline to the state's largest hospital system may outperform an Ivy for aspiring nurses. Film programs at Chapman University place graduates into Hollywood production houses that bigger-name research universities simply don't cultivate those relationships with. No ranking captures that.

Finally, there's fit — not the vague "you'll just know" kind, but concrete questions. What's the class size for intro courses in your intended major? Does the school have research opportunities you can access as an undergraduate? What's the ratio of career counselors to students? (The national average is roughly 1 counselor per 2,000 students; some schools do much better.) Rankings don't measure any of this. Visiting, calling the financial aid office, and talking to current juniors and seniors does.

Bottom Line

Run the net price calculator for every school on your list before comparing rankings. The real cost — what you actually pay — is what matters for your financial life after graduation.
Look up graduation rates by income group on the College Scorecard. Those numbers can diverge by 25+ percentage points at the same institution.
Use rankings to build an initial list, not to make a final decision. They're useful for sorting schools into rough tiers. They're not useful for determining whether you'll thrive, graduate on time, or find work in your field.
Peer reputation scores, which account for roughly 20% of U.S. News scores, largely reflect prestige from decades past — not what's happening on campus today.
The best question to ask any admissions office: "What percentage of graduates in my intended major are employed or in graduate programs within six months?" If they can't answer quickly, keep asking.

Frequently Asked Questions

What factors does U.S. News use to rank colleges?

For 2026, U.S. News uses up to 17 factors for National Universities. The biggest categories are student outcomes (graduation rates, retention, graduate earnings and debt) at roughly 40% combined, and peer assessment — a reputation survey of administrators — at about 20%. Faculty resources, per-student spending, and student test scores round out the rest. Acceptance rate was removed from the 2026 formula after years of criticism that it rewarded schools for rejecting students.

Are college rankings reliable or do schools manipulate them?

Both things are true. Rankings capture some real signal — schools that score well on graduation rates and student outcomes often do deliver better results. But the history of data manipulation is well-documented: Emory University, George Washington University, and Claremont McKenna College all admitted to submitting false data. Columbia's ranking fell nine spots after a professor exposed its reporting errors. Rankings are a reasonable starting filter, not a reliable verdict.

Why did Harvard and Yale stop participating in U.S. News rankings?

Yale Law School withdrew in November 2022, followed immediately by Harvard Law, then Harvard Medical School and dozens of other law schools in early 2023. Their core objection was that the methodology penalized schools for prioritizing need-based financial aid, admitting lower-income students, and supporting public interest careers (which pay less than corporate law and therefore dragged down salary outcome scores). The boycott forced some methodology reforms, including reduced weighting for standardized test scores.

Do college rankings predict my earnings after graduation?

Weakly — and less than most applicants assume. Your choice of major has a substantially larger effect on early-career earnings than your school's prestige rank. The federal College Scorecard shows median earnings 10 years after enrollment by institution and field of study. Computer science graduates from mid-ranked state schools routinely out-earn liberal arts graduates from elite universities. Rankings matter most for careers where the brand name is part of the credential: consulting, investment banking, and certain competitive graduate admissions processes.

What is the Economic Mobility Index and how does it differ from U.S. News?

Third Way's Economic Mobility Index measures how effectively a school moves low-income students into higher earnings — combining the percentage of enrolled students from low-income backgrounds with how quickly those students see a financial return on their education investment. Cal State campuses, CUNY colleges, and several HBCUs score near the top. Harvard ranks #847 on this metric because it enrolls so few low-income students. The two systems ask fundamentally different questions: U.S. News asks "is this school prestigious?" while the EMI asks "does this school create economic opportunity?"

How should I actually use rankings in my college search?

Rankings work best as a coarse filter early in the search process — they help you identify categories of schools worth exploring. Past that, they should be one input among several. The net price calculator, graduation rates by income group from the College Scorecard, major-specific career placement data, and campus visits give you information that no ranking can. If two schools are within 20 spots of each other, the ranking difference is essentially noise — the fit, cost, and program quality matter far more.