Written and maintained by CASRAI Editorial Board
Last updated
U-Multirank is a European Commission-funded ranking instrument that compares higher education institutions across five dimensions without ever combining the results into a single overall score. That design choice — not a missing feature, but the tool’s entire premise — is what separates it from the “big three” league tables: QS, Times Higher Education (THE), and ARWU (ShanghaiRanking), which CASRAI compares directly in its own ranking-methodology comparison. For a research administrator or institutional-research office deciding which instrument to cite in a benchmarking report, self-study, or public communication, U-Multirank is the answer to a specific complaint about the other three: that reducing dozens of unrelated indicators to one composite rank number necessarily encodes an arbitrary set of weights as if it were an objective fact.
Who runs it, and why the European Commission funds it
U-Multirank grew out of a European Commission-sponsored feasibility study in 2010 and an implementation project led by a consortium headed by Prof. Frans van Vught (Center for Higher Education Policy Studies, CHEPS, in the Netherlands) and Prof. Frank Ziegele (Centre for Higher Education, CHE, in Germany). It went live publicly on 13 May 2014, covering roughly 850 institutions in more than 70 countries in its first edition; it has since grown to cover well over 1,700 universities and colleges across some 90-plus countries. It continues to be funded through the European Commission’s Erasmus+ programme rather than by subscription fees, licensing, or advertising — a funding model CASRAI’s own audience should note when weighing U-Multirank’s independence against the commercial incentives that shape QS’s and THE’s business models.
The five dimensions, and why there’s no overall rank
U-Multirank scores institutions across five areas of activity: teaching and learning, research, knowledge transfer, international orientation, and regional engagement. Within those dimensions it reports dozens of individual indicators — things like graduation rates, external research income, patents awarded, or student mobility — sourced from institution-submitted data, bibliometric databases, and student surveys.
Crucially, U-Multirank does not aggregate those indicators into a composite score, and it does not publish a single ranked list. Its own methodology states there is no sound empirical basis for deciding how much a graduation rate should count against a patent count against a citation rate — any such weighting is a value judgment, not a measurement. Instead, for each indicator, an institution is placed into one of five performance groups: very good, good, average, below average, or weak, relative to a comparison group of institutions with a similar profile (a research university is compared against other research universities, not against a small teaching-focused college). This is the direct methodological rejection of what QS, THE, and ARWU all do: each of those three assigns fixed percentage weights to a fixed set of indicators and sums them into one number that produces one ordered list — see CASRAI’s QS weighting breakdown and THE’s five-pillar composite for exactly how that arithmetic works on the other side of this design divide. The CWTS Leiden Ranking shares U-Multirank’s discomfort with composite scores but stays purely bibliometric; U-Multirank’s five dimensions go well beyond research output alone.
User-driven, like-with-like comparison
U-Multirank is built to be queried, not read top-to-bottom. A user selects the institutions or the institutional profile group they care about (for example, research-intensive universities in a specific region), picks which of the roughly thirty-plus indicators actually matter to their own question, and the tool builds a personalized comparison table on the fly — there is no fixed “top 100” page to browse. This “like-with-like” grouping is meant to solve a real, longstanding critique of the composite league tables: that a specialized technical university and a comprehensive research university are structurally different institutions, and forcing them onto the same single scale rewards institutional type as much as it rewards performance.
When U-Multirank is the better citation
For a grant narrative, accreditation self-study, or institutional benchmarking exercise, U-Multirank is the more defensible source specifically when the claim being made is dimension-specific — “our regional knowledge-transfer performance is strong” or “our international student mobility outperforms peer institutions” — rather than a blanket prestige claim like “we are a top-100 university.” A composite rank from QS, THE, or ARWU cannot be decomposed to support a narrow, evidence-specific claim without re-deriving which sub-indicators actually drove the number; U-Multirank reports those sub-indicators directly, already broken out by dimension.
The trade-off is visibility and public familiarity: U-Multirank has never produced the kind of single celebrated “#1 university in the world” headline that QS, THE, and ARWU each generate every year, precisely because its design forbids that kind of claim. That makes it a poor fit for public-facing marketing copy chasing prestige, and a strong fit for internal reporting, funder narratives, and any context where an institution needs to make an honest, indicator-specific case rather than a single-number one.
Limitations worth knowing before citing it
Coverage is uneven: participation in U-Multirank is largely voluntary on the institution-submitted-data indicators, so an institution that has never opted in will have thinner, sometimes missing, data relative to one that actively participates — a gap that does not necessarily mean weaker performance, just non-participation. And because there is no overall score, U-Multirank cannot answer the single question league tables are built to answer (“which university is ranked higher overall”) — that is a deliberate design choice, not a shortcoming, but it means U-Multirank and the composite league tables are not interchangeable sources; they answer different questions and should be cited for the question each was actually built to answer.
Frequently asked questions
Is U-Multirank a real ranking, or just a comparison tool?
Both descriptions are accurate depending on what you mean by “ranking.” It ranks institutions on individual indicators (placing them into performance groups per indicator) but deliberately does not produce the single overall ordered list that the term “ranking” usually implies.
Does U-Multirank cover the same universities as QS, THE, and ARWU?
Largely, yes, plus many institutions the composite league tables don’t cover at all — U-Multirank’s profile-group design was specifically built to include universities of applications, technical universities, and other institution types the research-output-weighted composite tables tend to rank poorly or exclude.
Who funds U-Multirank, and is that a conflict of interest?
The European Commission, through its Erasmus+ programme — a public funding source rather than the subscription/advertising model that underlies QS’s and THE’s commercial rankings businesses. That doesn’t make U-Multirank free of institutional incentives of its own, but the funding source and the absence of a composite “winner” each year are both relevant context when weighing its independence.
