Introducing a new indicator in the U-Multirank 2019 edition

What is a university’s ‘region’?

The March 2019 report by the European University Association (EUA) “The Role of Universities in Regional Innovation Ecosystems” provides several case studies of European universities with convincing empirical evidence of their outreach and effects on the innovation system in their local city-region. However, the EUA report doesn’t actually define those ‘regions’, which leaves the ‘regional’ notion slightly opaque and in need of further clarification.

There are different ‘space/place’ ways of defining or delineating a university’s local region. Statistical offices usually apply their NUTS classification system, which provides a three-layered hierarchical system of intra-national administrative territories throughout Europe. Some academic studies define ‘functional regions’ that reflect spatial behaviour of individuals or organisations in a geographic space, such as commuter travel-to-work time. Others may, like the EUA, refrain from explicitly specifying a university’s home region, which avoids introducing a false sense of precision where the area actually starts and ends, while enabling broader or fuzzier descriptions that might be more appropriate to address specific issues or perspectives.

However, when comparisons or detailed information are required, it becomes imperative to be as precise as possible about a region’s territory and its geographical boundaries. Implementing geographical specifications becomes inevitable. NUTS-based definitions in particular are likely to misrepresent the geographic extent of a university’s outreach, especially in those cases where the university is located close to a national border and likely to generate spillover effects on an adjacent foreign region or country. In the case of small contained areas these considerations would seem to be of lesser importance: no matter how you choose to delineate or interpret the local region, a university’s impact profile will - one would think - hardly change anyway. But fresh empirical evidence, on the research cooperation patterns between UK universities and firms in the local business sector, challenges that assumption.

University-business interactions and innovation impact

A recent CWTS study entitled “Globalisation, localisation and glocalisation of university-business research cooperation: general patterns and trends in the UK university system”, funded by the United Kingdom’s Center for Global Higher Education, examines the performance profiles of 48 individual UK universities. The results indicate that driving forces of local university-business research cooperation may significantly depend on the spatial distance between the university and the partnering firm. Universities with many ‘close distance’ collaborations, involving distances less than 50 kilometer from the center of the university’s home town, have relatively large numbers of researchers with ‘dual affiliations’ (at both the target university and at nearby firm); these universities also tend to have more consultancy contracts with firms (both large and small). They are also relatively large universities in terms of research output. Several of those universities are located in the Greater London area.

Expanding that distance zone to a 100 km radius, or even further afield, reveals a shifting set of explanatory characteristics. In other words, the causes and nature of a university’s regional outreach and impact appears to be distance-dependent. So, more precise definitions that ‘local region’ are important, especially where a university’s impact on the economy is concerned: not only its impact in terms of alumni employment (in either the local public or private sector), or spending of its students and university staff, but also with regards to R&D cooperation with existing local firms, or the creation of new innovative firms.

Unfortunately, we lack comparative statistical data on those impacts. University management information systems may be able to breakdown some their regional engagement statistics by geographical origin, but such data are generally not available in the public domain. Large surveys among universities across Europe would help, but are not conducted so far. Sometimes, however, a report appears online on economic impacts of European universities, but the aggregated statistical data in such documents is not broken down to the level of local regions.

University-business research cooperation

Fortunately, one source of globally comparative empirical data exists that sheds some light on a university’s regional outreach and provides some clue of its (potential) economic impact: research publications from university-business research cooperation. Of course, this particular ‘tip of the iceberg’ only captures those successful university-business interactions with results that were sufficiently interesting and relevant to share with the outside world through those published research articles.

Nonetheless, these collaborative ties are interesting: not only do they provide empirical evidence of where exactly those partners are located, but also allow insights into research-related knowledge flows and associated ‘impact pathways’. Of particular interest are the connections and interactions with those R&D-active firms that contribute significantly to the local economy. 

New indicator for U-Multirank

U-Multirank is user-driven, multi-dimensional analytical tool for indicator-based comparisons of universities and other higher education institutions worldwide. CWTS contributes to the bibliometric indicators in U-Multirank.

Supported by our UK findings a university performance indicator was created that captures the close-distance collaboration with local industry located within a 50 km range from the university. This indicator offers, for the first time ever, comparative quantitative information on a university’s local and regional reach in terms of joint research with industry. Table 1 presents the world’s top 10 largest research universities in terms of their share of university-business co-publications that involve firms located in the local region. Universities with a strong regional outreach typically show shares of more than 30%, such as Stanford University which sits in the middle of Silicon Valley’s regional innovation ecosystem. Several universities, many in Asia, however seem to exceed Stanford’s regional strength as a research partner of local industry.

Table 1. Top 10 largest universities worldwide, sorted by declining share of university-business co-publications with firms located in the local region (2014-2017)

* Research-intensive universities with more than 20,000 research publications in 2014-2017 (CWTS Web of Science database)

This indicator is included in the latest edition of U-Multirank’s online information platform, launched on June 4^th 2019. It provides information on the performance of hundreds of universities worldwide, where the ‘UBCs with firms within 50 km (% of all UBCs)’ indicator is presented as a category score (ranging from a ‘high’ to ‘low’ score). Those scores are, by default (like any other indicator), a proxy measure of how engaged a university is in terms of outreach and impact involving local industry. This particular distance zone captures the majority of a university’s collaborative ties, and associated knowledge flows, within its home city or wider urban agglomeration.

Distance-based regions for comparative analysis

By applying a distance-based measure we are now able to produce comparative empirical evidence across universities in Europe and worldwide.  Our, slightly arbitrary, choice of a 50 km range doesn’t aim to solve the ‘region’ definition problem. Preferably, that distance zone should be customized according to a university’s outreach profile and/or geographical circumstances. Small countries would, for instance, have smaller close-distance zones.

Although we haven’t exactly pinned down the university’s local region, in any definitive or conclusive way, we can now at least reduce the analytical problem to that of choosing an appropriate distance zone - rather than avoiding this issue altogether when addressing and assessing a university’s regional impact.