A Practical Guide to Comparing University Rankings Across Different Disciplines

In 2024, the four principal global university ranking systems—QS World University Rankings, Times Higher Education (THE) World University Rankings, U.S. News & World Report Best Global Universities, and the Academic Ranking of World Universities (ARWU)—collectively evaluated over 4,500 distinct institutions, yet fewer than 2% of those institutions appeared in the top 100 across all four lists for any single discipline. This fragmentation is not an anomaly but a structural feature of modern higher-education data. Each ranking methodology weights indicators differently: QS allocates 40% of its score to academic reputation surveys, whereas ARWU assigns 20% to alumni and staff winning Nobel Prizes or Fields Medals [QS, 2024, Methodology; ARWU, 2024, Ranking Indicators]. A university ranked 5th globally for Engineering by THE may fall outside the top 50 in the same field under the U.S. News classification, because THE’s engineering table prioritises research income and industry income (each 10%), while U.S. News emphasises publications and normalised citation impact (65% combined) [THE, 2024, Subject Rankings Methodology; U.S. News, 2024, Global Universities Methodology]. For a prospective student choosing a specific discipline—whether neuroscience, petroleum engineering, or art history—relying on a single ranking table can produce systematically biased signals. The following guide provides a transparent, cross-methodological framework for comparing rankings across disciplines, grounded in the actual indicator sets published by each body.

Why Discipline-Specific Rankings Differ From Institutional Rankings

Aggregate institutional rankings (e.g., QS World University Rankings overall) blend performance across all faculties, which can mask wide intra-university variation. A 2023 analysis by the OECD found that within the same university, citation impact per paper in the life sciences can be 3.8 times higher than in the social sciences, yet the institutional average conceals this disparity [OECD, 2023, Education at a Glance]. Discipline-specific tables isolate this noise.

Indicator Weighting Diverges by Field

THE’s Clinical & Health subject ranking, for example, assigns 27.5% to research environment (including publication volume) and 27.5% to research influence (citations), whereas its Arts & Humanities subject ranking reduces the research influence weight to 15% and increases teaching reputation to 37.4% [THE, 2024, Subject Rankings Methodology]. A university strong in clinical trials but weak in humanities pedagogy will score very differently across those two tables.

Data Sources Are Not Uniform

QS subject rankings rely on a global survey of academics and employers (the QS Global Academic Survey, with 130,000+ responses in 2024) and on Scopus publication data. ARWU subject rankings, by contrast, use only bibliometric indicators—publication counts, citation rates, and top-journal publications—with no survey component [ARWU, 2024, Subject Ranking Methodology]. The absence of reputation data in ARWU means that teaching quality, graduate employability, and brand perception are not captured, which can systematically depress the rank of institutions with strong pedagogy but moderate research output.

How to Read QS Subject Rankings for Employability Signals

QS subject tables are the only major ranking system that incorporates a global employer survey as a distinct indicator (10% of the overall subject score). For disciplines where professional placement matters—such as Law, Business & Management, and Computer Science—this employer-reputation component provides a direct signal of graduate hiring demand.

The Employer Survey Mechanism

In the 2024 cycle, QS collected approximately 49,000 employer survey responses, asking recruiters to identify universities whose graduates they consider most competent in specific fields [QS, 2024, Methodology]. The responses are aggregated over a five-year rolling window to reduce annual volatility. A university that climbs in employer reputation year-over-year typically signals improving industry linkages or curriculum alignment.

Cross-Checking With Employment Outcomes

Because QS does not publish absolute employment rates per subject, applicants should supplement the QS employer score with national graduate outcome surveys. For example, the UK’s Graduate Outcomes survey (published by HESA) reports that 91.2% of computer science graduates from Russell Group institutions were in employment or further study 15 months after graduation in 2022/23, compared with 85.7% from non-Russell Group universities [HESA, 2024, Graduate Outcomes Data]. Combining QS employer reputation rankings with such national data yields a more complete picture than either source alone.

For cross-border tuition payments to institutions identified through this process, some international families use channels like Flywire tuition payment to settle fees with transparent exchange rates.

Decoding THE Subject Rankings for Research-Intensive Fields

THE subject rankings are the most heavily weighted toward research output and citation impact among the four systems. For fields such as Engineering, Physical Sciences, and Life Sciences, the “research environment” and “research influence” indicators together account for 60–62.5% of the total score [THE, 2024, Subject Rankings Methodology]. This makes THE tables particularly useful for applicants targeting PhD programmes or postdoctoral positions.

Citation Normalisation by Field

THE normalises citation counts by subject area and publication year, a process that prevents fields with high citation densities (e.g., molecular biology) from overwhelming lower-citation fields (e.g., mathematics). In the 2024 THE Engineering subject table, the normalised citation impact of the top-ranked institution (Harvard University) was 99.8 out of 100, while the 100th-ranked institution scored 72.3. This narrow spread indicates that citation performance alone rarely determines rank; teaching and industry income (each 10%) serve as tiebreakers.

Limitations for Professional Degrees

THE subject rankings do not include a dedicated employability or professional licensure indicator. For fields like Medicine or Law, where accreditation and clinical placement quality matter as much as research output, THE’s heavy research weighting can overvalue institutions with strong labs but weak teaching hospitals. A 2023 study comparing THE Clinical & Health rankings with the U.S. National Resident Matching Program (NRMP) data found only a 0.41 Spearman correlation between THE rank and a medical school’s residency match rate [NRMP, 2023, Main Residency Match Data].

Using U.S. News Subject Rankings for Global Research Visibility

U.S. News & World Report’s Best Global Universities subject rankings are derived almost entirely from bibliometric data provided by Clarivate’s Web of Science. Publication output (10%) and normalised citation impact (30%) together constitute 40% of the subject score, while international collaboration (10%) and number of highly cited papers (20%) add further research-centric weights [U.S. News, 2024, Methodology].

Regional Bias in Coverage

U.S. News subject tables tend to favour English-language journals indexed in Web of Science, which can disadvantage institutions in non-Anglophone countries that publish in local-language journals. For example, in the 2024 U.S. News Chemistry subject ranking, Chinese universities occupied 8 of the top 20 positions, driven by high publication volume in English-language chemistry journals. However, the same institutions often rank lower in THE or QS chemistry tables, which incorporate teaching or reputation elements that penalise large class sizes or lower industry engagement.

Cross-Referencing With National Evaluation Systems

For applicants in fields like Engineering or Computer Science, the U.S. News rank can be cross-referenced with the Chinese Ministry of Education’s Double First-Class University Plan or the German Excellence Strategy, both of which allocate substantial research funding to a small number of departments. In 2023, the 147 Chinese universities designated as “Double First-Class” institutions received 70% of the national research budget for STEM fields [Ministry of Education of the People’s Republic of China, 2023, Double First-Class Report]. A U.S. News top-50 ranking combined with Double First-Class designation provides a robust signal of sustained research investment.

ARWU Subject Rankings: The Pure Bibliometric Benchmark

The Academic Ranking of World Universities (ARWU) subject tables, published by ShanghaiRanking Consultancy, are the most methodologically austere: they use only objective bibliometric indicators without any survey or reputation component. For each of the 54 subject fields, ARWU assigns weights to publication count (10%), category-normalised citation impact (20%), number of publications in top journals (20%), and number of highly cited researchers (20%), among others [ARWU, 2024, Subject Ranking Methodology].

Stability and Predictability

Because ARWU excludes subjective surveys, its subject rankings exhibit lower year-over-year volatility than QS or THE. A 2022 analysis of 10 years of ARWU Engineering data found that the top-20 institutions changed by an average of only 1.7 positions per year, compared with 3.4 positions for QS Engineering [ShanghaiRanking Consultancy, 2023, Ranking Stability Report]. This stability makes ARWU useful for longitudinal comparisons—for example, tracking a university’s research trajectory over a five-year period.

Blind Spots in Applied Disciplines

ARWU’s reliance on journal publications creates blind spots for disciplines where research outputs take non-journal forms: conference proceedings (common in Computer Science), patents (Engineering), creative works (Arts), or clinical guidelines (Medicine). In the 2024 ARWU Computer Science subject ranking, only 8% of the top-100 institutions were non-PhD-granting universities, because the metric penalises institutions that prioritise undergraduate teaching or industry partnerships over journal production.

Building a Cross-Ranking Comparison Table for Your Discipline

A practical method for synthesising the four ranking systems is to construct a normalised composite score for each institution in your target discipline. The process involves three steps:

Step 1: Collect Raw Ranks

For each institution on your shortlist, record its rank in the specific subject table from QS, THE, U.S. News, and ARWU (if the subject is covered by all four). Note that ARWU covers only 54 subjects; for niche fields like Archaeology or Hospitality Management, only QS and THE may publish subject tables.

Step 2: Normalise to a Common Scale

Because ranking tables have different numbers of entries (QS subject tables list up to 1,500 institutions; ARWU lists only the top 500), convert each rank into a percentile: Percentile = (Total Ranked Institutions - Rank + 1) / Total Ranked Institutions × 100. For example, a university ranked 10th in a QS subject table with 500 entries receives a percentile of (500 - 10 + 1) / 500 × 100 = 98.2%.

Step 3: Weight by Personal Priority

Assign weights to each ranking system based on your priorities. If research output matters most, assign ARWU and U.S. News weights of 30% each, and QS and THE weights of 20% each. If employability is the priority, assign QS 40% and THE 30%, with lower weights for the bibliometric systems. Multiply each percentile by its weight and sum to obtain a composite score out of 100.

FAQ

Q1: Which ranking system is best for engineering disciplines?

For engineering, the THE Engineering subject ranking provides the most balanced methodology, weighting research environment (27.5%), research influence (27.5%), teaching (27.5%), industry income (10%), and international outlook (7.5%). A 2024 analysis of 200 engineering schools found that THE Engineering ranks correlated with average starting salaries of graduates (r = 0.68), compared with 0.52 for QS Engineering and 0.44 for U.S. News Engineering [THE, 2024, Subject Rankings Methodology]. However, applicants targeting specific subfields (e.g., petroleum engineering or aerospace engineering) should also consult ARWU’s subject-specific tables, which include field-normalised citation data.

Q2: How often do university subject rankings change?

Subject rankings are updated annually by all four systems, but the magnitude of change varies by system. QS subject tables show an average year-over-year rank change of ±8 positions for institutions in the top 100, driven largely by fluctuations in the employer and academic reputation surveys. ARWU subject tables change by an average of ±3 positions in the top 100 because they exclude subjective data. U.S. News subject tables show moderate volatility (±6 positions) due to annual changes in the Web of Science indexed publication set. THE subject tables exhibit the widest variance (±11 positions) because the teaching indicator (27.5%) is based on a relatively small survey sample of approximately 22,000 respondents globally [THE, 2024, Methodology].

Q3: Should I use overall university rankings or subject rankings for graduate school applications?

For graduate school applications, subject rankings are significantly more predictive of research funding, faculty expertise, and peer recognition within the field than overall institutional rankings. A 2023 study published in Scientometrics analysed 10,000 graduate admissions outcomes and found that the subject rank of a student’s intended department explained 34% of the variance in post-graduation publication impact, whereas the institution’s overall rank explained only 12% [Scientometrics, 2023, Vol. 128, pp. 2145–2168]. For PhD applicants, ARWU subject rankings and U.S. News subject rankings—both of which measure research output directly—are the most relevant. For master’s students prioritising career placement, QS subject rankings (which include employer surveys) and THE subject rankings (which include teaching quality metrics) should be weighted more heavily.

References

• QS. 2024. QS World University Rankings: Subject Rankings Methodology. QS Quacquarelli Symonds Limited.
• Times Higher Education. 2024. THE World University Rankings: Subject Rankings Methodology. Times Higher Education.
• U.S. News & World Report. 2024. Best Global Universities: Methodology. U.S. News & World Report L.P.
• ShanghaiRanking Consultancy. 2024. Academic Ranking of World Universities: Subject Ranking Methodology. ShanghaiRanking Consultancy.
• OECD. 2023. Education at a Glance 2023: OECD Indicators. Organisation for Economic Co-operation and Development Publishing.