Mind the Gap and Bridge the Gap: Research Excellenceand Diffusionof Academic Knowledge in Sweden

Wigren, C, Gabriellson, G and Kitagawa, F

Abstract

This paper aims to highlight the changing and diversifyingnature of academic work related to various forms of knowledge production and diffusion. Focusing on the changing research policy landscape in Sweden, three interrelated questions are investigated: what academics do in terms of commercialisation and public dissemination; how they perform these activities; and why they engage in these activities. Based on data from a recent survey with over 10000 academics in Sweden, we identify and analyse “high-performing” researchers, in the context of the commercialisation and public dissemination of their academic work.The quantitative analysis is supplemented by qualitative interviews with scientists at strong research environments in Sweden. We argue that there is a virtuous cycle connecting different academic activities in strong research environments −research excellence and excellence in knowledge production on one hand, and knowledge diffusion activities, such as commercialisation and public dissemination, on the other hand.

Introduction

The need to connect knowledge from universities to wealth creation and wellbeing insociety has become an increasingly important policy agenda in many countries. University-industry collaboration, the commercialisation of research results, and the protection of intellectual property emanating from universities, have in this respect become major policy and research drivers towards the promotion of innovation and economic development (e.g., Geuna, 1999; Feldman and Bercovitz, 2006). The literature on this topic has long had a relatively narrow commercial focus, emphasising patents and the creation of spin-offs originating from academic research. Recent works have, however, pointed to the much wider channels of communication between academia and industry (Perkmann and Walsh, 2007). Furthermore, whilst knowledge transfer is usually regarded as an activity that promotes and deploys academic know-how to specific users or industry sectors, studies increasingly acknowledge that universities and individual academics also work on public engagement or engage in popular dissemination activities in order to make scientific knowledge more widely available to society at large (Calhoun, 2006; Poliakoff and Webb, 2007).

Despite growing interest among academics and policy makers worldwide in the contribution of universities to economy and society, there are a number of perceived gaps in the understanding of interactions and relationships between the academic community and wider society. For instance, there is still a lack of understanding about who in academia interacts with industry, what they do, why they do so (D’Este and Neeley, 2008), and eventually how they do it. Bridging these gaps may provide valuable practical knowledge to universities aimed at improving their activities, incentive mechanisms and institutional practices in order to support different forms of linkages and interactions fostered at different organisational levels.

The overall aim of this study is to highlight the nature and characteristics of the changing and diversifying activities of academics by examining their involvement in commercialisation and public dissemination of their research. In line with previous studies (i.e., Prevezer and Shohet, 2002; Sorenson and Fleming, 2004; Chang and Yang, 2008) we define commercialisation and public dissemination of research as forms of knowledge diffusion. More specifically, we examine the extent to which academic excellence can be related to various forms of knowledge diffusion (see Leisyte, et al., 2008). In our study, this relationship is established by focusing on high-performing researchers and strong research environments − academic units that have received external funding based on previous records of excellence in research.

Literature Review

Motivation and Institutional Structures for Knowledge Production and Diffusion in Academia

Today universities are encouraged by various policy and funding instruments to actively engage in the diffusion of research based knowledge by multiple routes, including commercial channels − licensing patents, consulting, or implementing knowledge through spin-off companies; as well as more relationship-based knowledge transfer activities (Perkmann and Walsh, 2007) − collaborative research, commissioned research, consultancy, equipment sharing, advisory roles, joint supervision, joint publication and student placements.

The literature suggests that the reasons why university researchers engage with industry is diverse, and quite often the motive is to “further their research rather than to exploit their knowledge” (D’Este and Perkmann, 2010), i.e.their collaboration is oriented towards developing research and furthering knowledge, rather than exploiting knowledge for mere financial benefits (D’Este and Neeley, 2008; D’Este and Patel, 2007; D’Este and Perkmann, 2010). Although there have been some concerns that increased interaction with industry or wider society may increase the reluctance to share information and research material (Walsh, Cho and Cohen, 2005), it seems that academics are able to cope with the potential pressure that this implies. For example, a recent study by Jain, George and Maltarich (2009) shows that academics who participate in diffusion activities develop a “hybrid role identity” where they take active steps to preserve their identity as academics. These steps were, in turn, related to establishing relationships with other actors who possess necessary skills related to commercialisation. Steps were also taken to preserve certain cherished values associated with being an academic, and making sure these were not compromised as a result of their involvement in knowledge diffusion.

A growing body of work, both in the academic literature and other public documents, examines researchers’ motivation and attitudes toward knowledge diffusion within academic contexts. Jacobson et al. (2004), for example, identifies “the reward and incentive system” in academia as the main barrier for knowledge transfer (p.248). In the field of higher education research, there is also a body of literature that examines the link between promotion, tenure and motivation to faculty performance in conducting research and/or teaching in different national contexts (e.g. Lonsdale, 1983in Australia; Macfarlane in the UK, 2007; Tien, 2000 in Taiwan; Perna, 2005 in the US). Furthermore, Kandiko and Blackmore (2009) have investigated the changing nature of research, teaching and “traditional academic boundaries”as universities increasingly engage in interdisciplinary work and third stream activities. They particularly note that the broadening of tenure and promotion criteria to include knowledge transfer and societal benefit was opening up opportunities for academics to pursue interdisciplinary work earlier in their careers (Kandiko and Blackmore, 2009, p. 91).

Popular dissemination of research, or “public engagement” (Poliakoff and Webb, 2007), is seen as one strand of “third stream” activities. It includes public lectures, exhibitions, and similar events as well as communication through media. Public engagement has only recently attracted policy attention as a form of wider knowledge dissemination, with the recognition that scientists receiving public funding for their research have a duty to communicate their research to the public (Pearson, 2001). However, a recent report in Sweden points out that those types of activities rarely receive a direct reward or recognition in the current academic reward system (see Vetenskap and Allmänhet, 2007). These findings highlight the issue of inequitable opportunities for third stream activity at different stages of academic career development (see Leisyte, et al., 2008; Stephan and Levin, 2001), but it also recognises third stream activity as an integral part of the changing nature of academic research and knowledge production (see, Gibbons et al. 1994).

There has sometimes been a tendency to view scientific inquiry and knowledge diffusion as opposite ends on the scholarly agenda. The concerns include the risk that more practically oriented dissemination activities will divert research resources (primarily time and equipment) from the exploration of fundamental long-term questions and that less time will be invested into producing scientific output in the form of publications (David, 2002; Nelson, 2004; Geuna and Nesta, 2006). However, empirical research on this issue has rather showed that faculty members that are engaged in knowledge diffusionalso are the most productive in terms of scientific output. For example, numerous studies suggest that patent counts and publication counts of academic inventors are positively correlated and that these scientists publish more papers than non-patenting faculty (Geuna and Nesta, 2006; Meyer, 2006; Breschi, Lissoni and Montobbio, 2007). Academics who start up new companiesare also more productive in terms of journal publications (Darby and Zucker, 2007; Lowe and Gonzalez-Brambila, 2007). Furthermore, there is evidence that faculty involvement in contract research encourages publication rates (Van Looy et al., 2004) and faculty members who acquire external industrial funding also publish more scientific journal articles compared to their peers (Gullbrandsen and Smeby, 2005). Hence, the results suggest that the dissemination of scientific knowledge and publishing scientific articles are complementary activities and that knowledge diffusion is largely a function of opportunities created by academic work.

The findings cited above direct attention towards academic excellence and suggest that the most highly acclaimed researchers often play central roles in both the development of science and its successful dissemination (e.g., Meyer, 2006; Darby and Zucker, 2007). However, the relatively narrow focus on patenting and start-ups in prior studieshas leftthe potential broader link between academic excellence and the involvement in third stream activities largely unexplored. Moreover, little is known about how the institutional and legal frameworks in which the commercialisation and dissemination of scientific discoveries operates shape how academics balance their multiple roles (Zucker and Darby, 1997; see also, Jain et al., 2009). Thus, whilst noting other shortcomings, the present studyrecognisesthe need to better understand both the individual academic researcher as well as the institutional context where dissemination activities take place in order to further examine this issue.

The Swedish Policy Context

Sweden has undergone major changes in its research policy during the 1990s, further accelerated by recession and global industrial changes, as well asthe new belief in universities as driving economic growth to foster knowledge-based entrepreneurship (Etzkowitz et al. 2008; Benner and Sörlin, 2007; Jacob et al. 2003).In order to promote knowledge dissemination to the general public, Swedish universities was given a third task along with teaching and research rolesin the Higher Education Act of 1997. This law holds that in addition to the two main tasks of research and teaching, universities have a responsibility to communicate their research results to the society and industry. However, due to a lack of clear understanding of what the third task means, it has often been narrowly interpreted as engaging in commercial activities (Jacob et al. 2003). Nevertheless, at smaller regional universities, which tend to have strengths and focus on social sciences and health care, departments responsible for cooperation and interaction with the local society were established.

The current IP regime in Sweden is a rather uniquefactor that shapes university research commercialisation and other academic entrepreneurship activities. Unlike many other European countries, Sweden has kept the law on the university teacher’s exemption, which allows university researchers (not the universities as organisations) to retain full rights to their discoveries. Since scientists own the IP, irrespective of the funding source, they can transfer it to an independent company, hand it over to a university organisation, or use it as the foundation for their private enterprise, as they see fit (Etzkowitz et al., 2008). The researchers can also receive support from publicly financed technology, bridging foundations and university holding companies on a voluntary basis. Thus, university researchers have full discretion about the means of knowledge dissemination. In the case of patenting, the researcher receives the entire benefits emanating from the patent, but also has to bear all costs for his or her efforts (Sellenthin, 2009).

An important policy shift in Sweden during recent years is the increased emphasis on competition and concentration of resources by rewarding “strategic areas of research”. This is exemplified by the creation of mission-oriented agencies (Benner and Sörlin, 2007) such as the Swedish Agency for Innovation System (Vinnova) and the Foundation for Strategic Research (SSF), and the most recent government call for Strategic Research Areas in early 2009. The recent policy interest in developing strategic excellent research is also manifested in the Swedish Research Council’s initiatives to foster research of the highest international quality through, for example, the creation of various types of centres of excellence. It is this recent policy interest to support “strong research environments” and “strategic research areas”, in Sweden as well as in many other countries (c.f., Heinze et al. 2007; Hollingsworth et al. 2008),that underlie the principal focus of this study on the inter-linkage between ‘academic excellence’ andvarious forms of knowledge diffusion.

Research Methodology

In order to contribute to the development at both the theoretical and empirical level,this study investigates three interrelated research questions in connection to research excellence and engagement in knowledge diffusion among academics; more specifically the questions of what they do, how they do it; and why they do it. The data analysed derives from two sources. We first analyse quantitative data collected from a large scale questionnaire survey targeting over 20000 faculty members at 31 Swedish higher education institutes (HEIs). This quantitative approach was used to make a broad overview of the involvement of faculty members in dissemination activities and to examineif excellentor “high performing” researchers differ in their characteristics and dissemination behaviourfrom the rest of the sample, i.e. who they are andwhat they do in relation to third stream activity.Secondly, we analyse qualitative data collected from 22 interviews with faculty members at Linnaeus environments at Lund University. This qualitative approach was used to illustrate processes and strategies of individual scientists to engage in enterprise and dissemination activities and to interact with industry and society, i.e. to establishhow they engage in third stream activity, and to generate an understanding of the perceptions of academics about the nature of their work in relation to wider society, i.e. why they engage in third stream activity. By employing both quantitative and qualitative research methods, we intend to drawhighly coherent and holistic conclusionsrelated to our area of research investigation.

The questionnaire survey

The questionnaire was designed as a web-based survey to be sent to faculty members at Swedish HEIs.[1] Measures were derived from a careful review of previous work on faculty involvement in dissemination activities, i.e., communication of research to the general public, cooperation with external actors, and research commercialisation (e.g., Inzelt, 2004; Schartinger et al., 2002). The questions were pilot tested at five Swedish universities in 2005-2006. This procedure led to 1948 responses, corresponding to a response rate of approximately 53%, which were used to make smaller adjustments in the questionnaire.

The finalised questionnaire was sent at the beginning of 2007 to faculty members at 31 Swedish higher education institutes, including universities and university colleges. The participants included all faculty members who were employed at least 40% of a full time position and who had a valid e-mail address. A random sample of 1000 faculty members was drawn for each of the 10 largest HEIs. At the remaining 21,all faculty members were included.The questionnaire survey yielded 10221 responses, which corresponds to a response rate of approximately 52%. The 433 incomplete responses were excluded, leaving 9788 from faculty members to be used in the analysis. We were unable to check for potential non-response bias due to promised full anonymity. However, the large number of cases and the high response rate reduce the risk of severe bias in the sample.

Creating groups for comparison of dissemination activities

Based on the final responses, we created two groups to match our interest in comparing high-performing researchers with other faculty members and examine what they do in relation to third stream activity. Academic performance was evaluated on two dimensions. The first dimension was research productivity, measured by the number of scientific publications in peer reviewed journals or authored/co-authored books over the last three years (2004-2006). Based on a review of related literature and research (e.g., Calderini, et al., 2007; Lowe and Gonzalez-Brambila, 2007; Yang and Chang, 2010) the threshold for above average academic performance was set to four or more published works in the past three years. The second dimension was the percentage of research activity within the position, measured in relation to full time employment. The threshold for above average performance was set to more than 50% research in position, as it indicates that research is the main task among the many duties of an ordinary faculty member. To be included in the group of a high performing researcher, a faculty member consequently has to fulfil both criteria. The Pearson correlation coefficient between the two performance dimensions is r. = 23 (p<.01). This signals partial, but far from full, overlap and supports the use of both dimensions.The resulting two groups were subsequently made up of 1056 high performing researchers (10.8% of total population) and 8732 other faculty members (89.2% of total population).

In the following analyses,we used univariate non-parametric tests to identify statistically significant differences between the two groups of faculty members. Compared to parametric tests, they rely on less rigorous assumptions which makesthem appropriate whenanalysing subsamples with varying number of respondents, which is the case in our study.

A careful investigation reveals some significant differences between the two groups. First, there is a difference in the age distribution,with more senior academics in the high performing group. This difference was expected, as research performance is generally found to increase with seniority (e.g., Calderini, et al., 2007) and senior staff is more likely to be included in publications as co-authors. There is also a remarkable difference in terms of academic degree, where the vast majority in the group with high performing researchers have PhDs,whereas the other group has a very large share of faculty members holding bachelor or master degrees. Moreover, there is a difference with respect to scientific areas between the two groups. Here we can see that the group with high performing researchers has a much larger share of faculty members from mathematics and natural science as well as from medicine and dentistry. There is, on the other hand, a much larger share of faculty members from humanities and from social science (including law) in the lower performing group. Finally, with respect to gender, there is an overrepresentation of male researchers in the group with high performing researchers. All differences are significant at p <.01. The full statistics are presented in Table 1 below.