Two Major U.S. News Ranking Reforms in Three Years: What Are They Telling Us?

Through 2023, several U.S. News indicators tended to favor highly selective institutions with the resources to offer an intimate undergraduate experience. Class size accounted for 8% of the formula, faculty with terminal degrees for 3%, alumni giving participation for 3%, and admitted students’ high school class rank for 2%. Together, these four measures represented 16% of the score.

Each measure had a relatively modest weight, but their combined influence mattered. They tended to benefit private, smaller, highly selective institutions and helped shape what many families came to think of as a “good university.”

The 2024 edition removed those four measures and redistributed emphasis toward student outcomes, first-generation and Pell Grant students, research impact, and graduate debt. This changed the question being asked: not only what resources or credentials a university has, but what happens to the students it serves.

Many public universities educate large numbers of low-income and first-generation students. When those students achieve strong graduation outcomes, the new framework can recognize that success more fully. Removing class size also reduced a disadvantage for institutions with large introductory lectures. Removing alumni giving eliminated another indicator that often favored private institutions. Research measures, meanwhile, changed the balance for universities with different research profiles.

The shift was visible across familiar names. Washington University in St. Louis moved into the mid-20s, Wake Forest moved from around the top 30 toward the top 50, and NYU moved from the mid-20s to the mid-30s. Rochester, Northeastern, Case Western Reserve, American University, and Tulane also saw notable declines.

In contrast, Texas A&M rose to No. 47, Stony Brook moved into the top 60, and the University of California system was a particularly visible beneficiary. UC Merced moved into the top 60, UC Davis and UC San Diego entered the top 30, and UC Berkeley and UCLA shared No. 15.

These movements did not mean that teaching, programs, or student life had transformed overnight. They showed how changing the formula can change the ranking’s own definition of a good university. Public or private status alone does not determine quality or explain every movement.

The 2025 and 2026 editions made smaller adjustments rather than repeating the scale of the 2024 overhaul. The 2027 edition introduces another important change, alongside updates to the categories in which institutions are ranked.

First, ranking categories have been updated in response to the 2025 Carnegie Classification. Some universities now appear in a different category, such as National Universities, Regional Universities, or National Liberal Arts Colleges. Each list compares institutions within its own category. A place on one list is not directly equivalent to the same number on another, and a school that changes categories should not be compared mechanically with its previous position.

Second, Graduate Indebtedness has been removed and replaced by Earnings by Major, retaining a 5% weight. The previous indicator considered the typical federal student-loan debt of graduates who borrowed. Under that measure, a lower debt burden could provide an advantage.

The replacement evaluates earnings four years after graduation, using the U.S. Department of Education’s College Scorecard. Its population is specifically employed federal financial aid recipients whose highest degree is a bachelor’s. It does not represent every graduate or guarantee any individual student’s future income.

The comparison takes place within the same academic discipline. Computer science graduates at one institution are compared with computer science graduates at other institutions, not with history graduates. U.S. News uses statistical transformations to make results across fields comparable, reducing the reward for simply enrolling more students in high-paying majors.

The new factor therefore asks how graduates perform in the labor market relative to others in their field. U.S. News also explains that the underlying federal debt data had not been updated for several years and that future availability was uncertain. This is a change in both the question being asked and the data used to answer it.

Although the 2027 changes are not identical in scale to the 2024 overhaul, the new results show considerable movement. Wake Forest and Boston University reached No. 34; Northeastern and Villanova reached No. 42; Case Western Reserve reached No. 44; and Stevens Institute of Technology reached No. 59.

At the most selective end, MIT took No. 1, Caltech reached No. 5, Carnegie Mellon reached No. 14, and Washington University in St. Louis reached No. 16. Several public institutions, including campuses in the UC system and Purdue, moved lower, while Pittsburgh was a notable exception moving upward.

One interpretation is that removing the debt indicator reduced an advantage some public universities enjoyed, especially for in-state students. A new earnings measure can instead favor institutions whose graduates perform strongly in employment markets. But a ranking combines many indicators, and changes in a school’s own data, its peers’ data, and the methodology all matter. The 5% replacement cannot by itself prove why any individual university moved.

Institutional reputation, professional networks, career services, and the resources available to students may all contribute to graduate outcomes. Large public universities often serve a broader range of students and public missions, while some smaller, highly selective private institutions can provide more concentrated support. These are questions to investigate at the program level, not grounds for assuming that every private university outperforms every public one.

A less obvious implication concerns highly specialized institutions. To understand it, remember that the new factor does not simply compare average salaries across unrelated majors. Computer science is compared with computer science; mechanical engineering with mechanical engineering; economics with economics.

Adjusting for differences between fields does not make the structure of an institution irrelevant. It addresses the fact that some fields generally pay more, but it does not award a separate bonus merely because a university offers a particularly broad range of majors.

Consider a simplified analogy, not the actual U.S. News formula. Student A takes two subjects and earns 100 in each. Student B takes ten subjects and earns 95 in each. Assume that performance in every subject has already been compared with others studying the same subject.

A’s advantage is not that the two subjects were easier. It is that A performs at the very top in both selected fields. B has greater breadth and is excellent across all ten, but breadth itself does not automatically earn an additional reward in this example.

A highly focused institution such as Caltech concentrates its undergraduate programs in fields where it has particular strength. If graduates in those fields perform very strongly against graduates of the same subjects elsewhere, that can be favorable under an earnings-based measure. A comprehensive university may offer many strong programs, but “very good” results across many fields need not equal the very highest relative performance.

The implication is not simply that more STEM majors means a better score. Rather, a university can be narrow in scope yet highly competitive in the fields it actually teaches. The article’s specialization analogy is an interpretation of the measure, not a reconstruction of its statistical calculation.

MIT, Caltech, and Carnegie Mellon’s strong performance in the earnings measure fits this discussion: their graduates can remain highly competitive even after differences between fields are taken into account. It also raises a useful question for families. When a university offers a particular major, how well does it teach that subject and support students after graduation?

Offering more majors is not, on its own, evidence of stronger outcomes. A university that expands its programs without building corresponding depth, student support, or professional opportunities may not gain an advantage from the new measure. The meaningful comparison is the strength of the program a student will actually enter.

This is probably the question many families have been asking. Our answer is that rankings can remain a useful reference, provided we understand what they measure and what they do not. They are neither a perfect verdict nor a substitute for an individual college strategy.

Society changes, work changes, and the expectations placed on higher education change with them. Revising a ranking is an attempt to reflect those changing priorities. The goal is relative objectivity, not a timeless or completely correct definition of quality.

The earnings measure reflects families’ growing interest in career competitiveness and the return on a college education. Two reforms in three years do not automatically make a ranking untrustworthy. They reveal that rankings are not discovering one fixed “true quality” of a university; they are repeatedly updating what a good university is expected to do.

If a student prioritizes employment opportunities, industry recognition, and economic outcomes after a bachelor’s degree, this edition offers useful information. But a student who values intellectual exploration, small classes, undergraduate research, close faculty relationships, or graduate study may have priorities that are less completely captured by this formula. Some fields are not appropriately judged mainly by short-term earnings.

Overall rankings can be considered, but they should not be considered alone. Investigate the strength of the intended major, program-specific outcomes, teaching environment, professional opportunities, and the student’s long-term goals before deciding.

A student who strongly values small classes might find parts of the pre-2024 framework closer to their own preferences. That does not mean an older ranking should become a new rulebook. It means families should look directly at the experiences they care about.

Every methodology revision is asking a different version of the same question: what should a good university provide for students today? The answer built into this year’s formula may not be identical to the answer that matters most to your family.

College applications are a demanding, long-term undertaking. Read rankings as one source of information, then return to the student: their intended major, the way they learn, the opportunities they need, and the future they want to build.

ToIvy wishes every student a successful 2026–2027 application season. If you need help turning those priorities into a thoughtful college list and application plan, our advisors are here to work with you.