U-Multirank is a comprehensive benchmarking tool for universities produced by CWTS. Instead of aiming to rank and find the ‘best’ university from a global list, it instead aims to offer comparative benchmarking of performance from user defined criteria based on location, size and institutional focus. This means that it preserves most of the largest benefits to be derived from rankings; increased comparability of institutional performance, greater transparency in governance and data reporting, without creating the unhealthy competitive incentives for universities.

U-Multirank depends heavily upon two other separate but related initiatives; U-Map which utilises a much more restricted set of indicators, which USP and UNICAMP still use, while UNESP has chosen to adopt the full range of indicators available to U- Multirank, and the CWTS Leiden Ranking, from which it gathers its external bibliometric information.

Like other benchmarking tools, U-Multimap is committed to the idea of overcoming the idea of a singular model of excellence, recognising that a strong higher education system requires many different forms of excellence. This approach allows U- Multirank to take a much wider variety of metrics into account than a composite ranking, without creating heavily aggregated scores of dubious descriptive power, compared to the Times Higher Teaching score, for example, which combines five different factors to produce something wherein universities with very different profiles

Compiling Team and Financing

U-Multirank is developed and implemented on the initiative of the European Commission by an independent consortium led by the Centre for Higher Education (CHE) in Germany, the Center for Higher Education Policy Studies (CHEPS) at the University of Twente and the Centre for Science and Technology Studies (CWTS) from Leiden University, both in the Netherlands. The consortium is headed by Professor Dr. Frans van Vught of CHEPS and Professor Dr. Frank Ziegele of the CHE. Associate partners include specialist organisations (e.g. on data collection, web design), national ranking organisations and stakeholder organisations.

Lead Partners

• Centre for Higher Education (CHE)
• Center for Higher Education Policy Studies (CHEPS)
• Centre for Science and Technology Studies (CWTS)

Partners

Bertelsmann Stiftung Folge 3 Push

Associate Partners

• Fundación CYD
• Perspektywy Education Foundation
• Conference of European Schools for Advanced Engineering Education and Research (CESAER)
• European Student Union (ESU)
• Universities of Applied Sciences Network
• StudyPortals
• Erasmus Student Network (ESN)

The project is financed by seed funding from the European Commission’s Erasmus+ programme.

Specificity

Given the vast range of indicators available, and the ability for universities to report as much, or as little, as they choose and the ease of comparing and benchmarking this against nearly 1000 universities worldwide, U-Multirank can provide vital insight into a university’s performance in a clear way. While it does not exert the kind of influence that the major rankings do, in part because of its failure to generate large headlines and in part because of its comparatively low level of recognition, universities would be well advised to participate in the initiative for the ease with which comparisons can be made with international partners. U-Multirank can be a useful tool for understanding other rankings, as well as giving a clearer picture of the university’s performance than the rankings themselves.

Full U-Multirank Methodology and Metrics

Data Collection and Publication Dates

U-Multirank derives its data from a mixture of publicly available citation information, patent registries, a student survey, self-reported data and prefilling.

Bibliometric

Bibliometric information gathered from the CWTS licensed edition of Clarivate Web of Science, including the SCI, SSCI and Arts and Humanities Index. Arts and humanities are excluded from the citation counts, because it is near impossible to find a consistent and reliable citation metric for these areas, and they very often depend on other forms of publication, such as books, artwork and conference proceedings. Therefore, just citations in peer-reviewed journals is not a reliable indicator of quality.

There are two groups of institutions covered by U-Multirank, the top 850 universities in the Leiden ranking, of which all three of the state universities are, and full participants in U-Multirank, which UNESP is.

CWTS processes and cleans all data ‘top down’, but accepts voluntary submission of data on commonly confused author and institutional attributions. This is especially important where attributions may be in multiple languages; variations on University of Sao Paulo, Universidade de São Paulo and University of São Paulo can appear as three separate institutions, while attribution for medical research split between hospitals and medical schools can also prove problematic.

The impact and citation measures are largely taken from the CWTS Leiden ranking, as they are produced by the same institute.

Patent Information

Information on patents is taken from the global patent index Patstat, while non-patent references (NPRs) within filed patents are cross referenced with Web of Science.

Self Reported Data

Data are reported via a questionnaire to be filled out, available here, which is for all data unrelated to citations and patents. It is then independently assessed and validated by the U-Multirank team.