Building of Academics’ Networks—An analysis based on Causation and Effectuation theory

Governments for many years have been keen to encourage University-Industry Collaborations (UICs). This is particularly so in those regions with a poor performance on innovation (Charles 2006). It has been acknowledged that through knowledge-based collaborations for instance, universities can assist firms in bringing forward technological innovations in their regions (Sternberg 2000; Gunasekara 2006; Agrawal 2001; Charles 2006). In Norway, as in many European countries, various reforms have increasingly led to the diminishing of the perceived boundaries between universities and their environment (Sataøen 2018). By forging new and closer local relationships, universities contribute to the social and economic development of their regions (Trippl et al. 2012; Charles 2006; Christopherson and Clark 2010).

In the pursuance of an additional or ‘third’ role of outreach activities that aims to contribute to innovation and social change (Gulbrandsen and Slipersæter 2007), universities have adopted varying models. For example, some universities have assumed the role of an ‘entrepreneurial’ university (Foss and Gibson 2015). In such universities an increased emphasis is placed on commercializing research, licensing of technology, creating university spin-offs, introducing entrepreneurship programmes, and expanding university-industry relationships (Foss and Gibson 2015; Klofsten and Jones-Evans 2000; Altmann and Ebersberger 2013; Clark 1998). Beyond this, there is an expectation that academics take on entrepreneurial characteristics within an entrepreneurial culture (Vorley and Nelles 2009). Etzkowitz (2003), illustrates these entrepreneurial traits at the sub-organisational level where research groups are described to function as ‘quasi’ firms.

Several studies have developed entrepreneurial profiles of academic scientists depending on the specific engagement mode identified. ‘Academic Entrepreneurs’ are those academics looking to commercialise academic intellectual property—these academics essentially act as entrepreneurs. On the other hand, ‘Entrepreneurial Academics’ are those academics who act entrepreneurially in order to accomplish their academic jobs (Miller et al. 2018; Nyeko and Sing 2015; Jain et al. 2009; Meyer 2003). These profiles encompass teaching and building of entrepreneurial competency as well as creating new ventures. Indeed, academic entrepreneurship is conceptualised as encompassing a wider range of engagement activities than only commercialisation (Jain et al. 2009). Further, academic scientists have been shown to form linkages across university and industry entities for the purpose of accessing and disseminating knowledge (Lam 2007; Stuck et al. 2016; Pataraia et al. 2014, 2015; Lowrie and McKnight 2004) underpinned by social networks (Granovetter 1973).

Engel et al. (2017) posit that the process of building new networks by entrepreneurs is a form of entrepreneurial activity that involves unpredictability and often, goal ambiguity. Following from the increasing entrepreneurial identity assumed by academics, we compare the decision-making process of entrepreneurial academics to that of traditional entrepreneurs. We specifically theorise that the actions of individual academics, when the outcomes of networking cannot be identified in advance, are comparable to the decision-making approaches employed by traditional entrepreneurs faced by uncertainty. The theories of effectuation (i.e. flexible and non-goal directed decision-making) and causation (i.e. goal-directed decision-making) (Sarasvathy 20052001; Sarasvathy and Dew 2011; Engel et al. 2017) can thus be employed to explore how networks of academics develop.

Knowing the approach employed by academics is key to understanding the possible outcomes of their social networks with respect to opportunity discovery (Sarasvathy 2005). Simply, by employing the entrepreneurial theories of causation and effectuation, we contribute to a potentially under-researched aspect of the literature on how academics build their networks. Subsequently, we also explore how the approach employed relates to the type of tie (i.e. industry or academia) to be established and the geography of those networks. Evidence was collected through in-depth interviews with academics from the University of Stavanger (UiS), Norway—a university with a history of co-evolution with the oil-industry, the major industry in its region.

Academic scientists advance their research through networks of multidisciplinary individuals (Stuck et al. 2016; Henry and Pinch 2000; Perkmann and Walsh 2007). For instance, academics engage with industry or government agencies (Ramos-Vielba et al. 2016), and network for the purpose of exchanging knowledge (Lam 2007; Stuck et al. 2016; Pataraia et al. 2014, 2015; Lowrie and McKnight 2004). Fernández-Pérez et al. (2015) have reported that academic scientists are important for both the dissemination of new knowledge and identifying opportunities through which knowledge could be converted into commercial form (Perkmann et al. 2011). Consequently, their experience also makes these academics likely partners for companies seeking to commercialise acquired knowledge (Siegel et al. 2007). These networks involving academics exist at both regional and extra-regional levels (Trippl 2013; Mahroum 2000).

Various modes of engagement employed by universities are largely policy-driven (Sataøen 2018). Universities have been increasingly encouraged by governments to actively participate in regional development (Charles 2006). Some universities have assumed an entrepreneurial model (Gulbrandsen and Slipersæter 2007; Foss and Gibson 2015), and their role in the generation and dissemination of information is perceived to have both tangible and intangible benefits (Shaw and Allison 1999; Goddard and Chatterton 1999; Vorley and Nelles 2009). At the regional level for instance, universities are involved in innovative projects that enhance the competitiveness of regional firms. These knowledge exchange projects comprise part of the ‘third mission’ role, which highlights the engagement of universities with their communities outside traditional research and teaching roles. Consequently, the boundaries between universities and communities have diminished (Sataøen 2018).

Given the key role of individual actors in knowledge exchange processes (Henry and Pinch 2000; Nonaka 1994; Almeida and Kogut 1999), individual academics have been reported to also assume entrepreneurial roles. These roles encompass teaching and building of entrepreneurial competency as well as creating new ventures. Academic entrepreneurship is conceptualised as encompassing a wider range of engagement activities than only commercialisation (Jain et al. 2009). Accordingly, efforts that encourage the building of entrepreneurial capacities within universities are also considered entrepreneurial (Klofsten and Jones-Evans 2000; Altmann and Ebersberger 2013). To the extent that research groups within universities have been described to function as ‘quasi’ firms (Etzkowitz 2003), it can be inferred that academics may act in an entrepreneurial manner when pursuing the development of research networks and building research teams.

The entrepreneurial profiles assumed by academics cover a host of engagement activities, as shown in Table 1. Opposed to the classical commercial ‘academic entrepreneur’, the ‘entrepreneurial academic’ adapts an entrepreneurial outlook and pursues industry partnerships with the aim to further their research objectives (Jain et al. 2009; Miller et al. 2018). While the academic entrepreneur pursues more formal partnerships with commercialisation partners, the entrepreneurial academic engages with commercial partners in a range of collaborative and less formal modes of engagement. These entrepreneurial traits adopted by academics are mostly related to teaching, administration or consulting (Nyeko and Sing 2015; Jain et al. 2009; Perkmann et al. 2013; Miller et al. 2018).

Effectuation theory, with its focus on non-goal driven logic, improvisation and leveraging contingencies (Sarasvathy 2005, 2001), has great potential to explain the undirected aspect of networking (Engel et al. 2017). Expert entrepreneurs make decisions in a non-predictive manner by employing five principles, characterised as: bird-in-hand, affordable loss, lemonade, patchwork quilt and pilot-in-the-plane (Sarasvathy 2005) as elaborated in Table 2. Causation, in contrast to effectuation rests on the logic of prediction and demands that the entrepreneur makes an analysis of the future on the basis of which a decision can be made, i.e. where knowledge thrives. The logic for using the causation processes is, to the extent that we foresee the future, we can control it whereas that for effectuation is, to the extent that we can work with things within our control, we don’t need to predict the future (Sarasvathy 2001).

The effectual and causal approaches adopted while networking have been operationalised as elaborated in Table 3. Causation employs a goal-directed approach whereas effectuation employs an emergent and unordered approach focused on co-creation (Sarasvathy 2005). Causal thinkers intend from the outset to achieve a specific goal while effectuators leverage the effect of circumstances and unexpected surprises while networking (Engel et al. 2017; Sarasvathy 2005). While effectual thinkers may not have a specific purpose for establishing a network, causators approach networking by taking deliberate actions concerning who to collaborate with and what needs to be achieved (Engel et al. 2017).

While not being goal-specific, effectuators have a broader perspective of networking and objectives in mind. The focus is placed on generating unexpected contingencies through meeting new people and discovering new facets in existing ties. Typically, effectual thinkers pursue goals based on an assessment of what is already available within their means following the ‘bird-in-hand’ principle (Sarasvathy 2005). Effectuators do not pursue random interests but what they consider to be worthwhile. In contrast, a narrow approach employed by causators is focused on meeting the right people and reaching them efficiently (Engel et al. 2017).

It has been suggested that entrepreneurs are able to shift between the use of effectuation and causation (Schreier and Senn 2018; Andersson 2011). The effectual approach to network-building has been explained to be positively associated with initial entry speed and international scope speed while a causal approach is negatively associated with initial entry speed and international scope speed (Prashantham et al. 2019). Galkina and Chetty (2015) also show that entrepreneurs of small and medium enterprises network with interested partners, instead of carefully selecting international partners according to predefined network goals. In the case of opportunity recognition it has been further highlighted that self-efficacy, entrepreneurial cognitive activities and access to specific resources (means at hand) are determinants for international new ventures to materialize (Hannibal et al. 2016; Andersson and Evers 2015). These studies show that both endogenous and exogenous factors influence the usage of either effectuation or causation.

The concept of causation and effectuation fit well with the concepts of exploitation and exploration described by (March 1991). Sarasvathy (2005) highlights this by explaining the causal approach of exploiting pre-existing knowledge as opposed to the effectual tendency to explore new ideas. While the returns of exploitation are usually positive, proximate, and predictable, the essence of exploration is experimentation with new alternatives (March 1991). This implies that exploitation, because it is action based on existing facts, most likely yields expected outcomes. In contrast, exploration results in unexpected and serendipitous outcomes.

In this study, the extent to which individual academics employ causation or effectuation when building their networks was assessed. Particular focus was placed on individual network contacts and how these links were initiated. The social network approach to building ego networks was loosely applied for obtaining specific examples in order to examine the development of academic networks. Subsequently, the theories of effectuation and causation were applied according to the characteristics of the two approaches as shown in Table 2.

When building ego networks via the personal network design approach, three distinct steps are involved; 1. Generation of names of contacts (alters) based on simple open-ended questions, 2. Interpretation of names where the respondent is asked about each name mentioned and 3. Name Interrelation, where the respondent is asked about the ties between alters. However, it is not required that a network be connected nor to have any ties at all especially at its initiation (Borgatti et al. 2013). For this reason, it is possible to have an ego network without necessarily emphasizing the connections between them. Accordingly, the third step of ‘name interrelation’ is omitted here, as connections between alters is not a question for this study which focused on the initiation of the Academics’ networks.

The university and regional contexts were both of importance for this study—the University of Stavanger presented an interesting case. The university was established at least in part in response to the discovery of oil offshore from the region of Rogaland and subsequently co-evolved with the development of the oil industry. Together, the discovery of oil and the subsequent establishment of the UiS have contributed to the economic and social transformation of the region (Ahoba-Sam 2019; Alpaydin et al. 2018).

It was expedient to select from a population of academics who were known to be engaged with industry. Consequently, we focused on engineering and applied sciences as these disciplines have been shown to report high levels of industry engagement (Perkmann et al. 2013). The data presented here reflects the views of selected academics from the Engineering Faculty of the University of Stavanger (UiS), Norway. The selected academics were specifically from the Centre for Risk Management and Societal Safety (SEROS), and the departments of Electrical Engineering and Computer Science, Energy and Petroleum Engineering, Mathematics and Physics, and Energy Resources of the UiS Engineering faculty.

Evidence was collected through in-depth questioning of 12 selected academic scientists during the period of March–May 2018. Interviews averaged about one hour and were recorded with the permission of interviewees. In the interviews, about 10 individual network contacts were discussed for each of the 12 interviewees. This made it possible to assess how a single individual established different ties. Specifically, each interviewee was engaged in an interactive session where they were asked to write down 10 of their ‘important’ network contacts¹. This resulted in the provision of 8–12 names each from interviewees. In this way, 118 observations were obtained and analysed. The definition of importance was left to each interviewee’s discretion. After generating this list, we proceeded to discuss each of the individuals’ contacts in light of how they were initiated, and the peculiarities of each relationship examined.

The interviews were recorded with the permission of the interviewees and subsequently transcribed. Coding essentially took three steps. In the first instance, the emerging themes were identified as expressed in the interviewees’ own words. Next, these themes were grouped across the various interviews in view of the suggested networking categories Engel et al. (2017) suggest as shown in Table 3. Next these emergent codes are weaved together to tell the lessons discovered, buttressed by interviewees’ quotes.

Interviews were conducted in the period March–May 2018. All interviewees were of post-graduate level and aged between 35–65 years. Four were women, with the remaining 8 being men. 3 of them were of expatriate origin whereas 9 were Norwegians. The academic scientists were selected non-purposefully, based on references obtained from the central administration of the UiS. From a list containing ~ 204 academics known by the University to be externally engaged, a selection was made based on on-going industry projects under the faculty of engineering (~ 67). From a total of 15 engaged academics who accepted to be interviewed, only the 12 interviewees² represented here consented to sharing specific examples of their personal network contacts. Overall, the questions discussed were focused on understanding how the academics’ networks were established. The data collected was transcribed and subsequently coded in order to identify common themes and patterns for analysis (Yin 1984).

A qualitative approach was employed, which presents the advantage of understanding the complexity of UICs especially as each collaboration is unique (Plewa et al. 2013). The semi-structured nature of interviews was useful for obtaining in-depth understanding of the cases reviewed (Yin 2002; Hammarberg et al. 2016; Wilson 2014). All interviewees are anonymized and will also be referred to as Interviewee A1, A2, … to A12. Further, since the data collected is personal, the names of contacts and further information leading to the possible identification of persons are excluded.

This section reviews the general nature of the networks of the individual academics analysed in this study. Since 118 observation points have been made, we therefore first present some general description of the collected data. Next, details on the approaches employed to build the academics’ networks are presented in line with the dimensions presented in Table 2.³

This study has shown that academic scientists use both effectuation and causation to build their networks. It has revealed the application of opportunity exploration and exploitation tendencies when initiating contacts across geographical (regional or international) and institutional (industry or academia) types. In a related study, Ahoba-Sam (2019) shows that the motivations for these academic scientists to collaborate are mediated by both personal and external factors. Similarly, the approach (i.e. causation or effectuation) employed by the academics studied was observed to be affected by both endogenous and exogenous factors. The choice of which approach to use was not necessarily always dependent on the academic’s personal preference. It was evident that external pressures such as the requirements of funding bodies were likely to influence a causal approach in networking rather than a more effectual approach. To this end, causation was observed more with industry networking. It can be inferred that the choice of which approach to employ is also linked to the motivation to network—if the initiation of the network is externally motivated (e.g. to access funding), there is the likelihood to employ a causal approach rather than if it is a personal motivation such as exchanging research ideas with another academic with less instrumental objectives.

The findings of this study offer some evidential support to the propositions made by Engel et al. (2017) about entrepreneurial networking. Engel et al.’s (2017, p. 44) proposition 1 states that, ‘under uncertainty entrepreneurial networking is driven by an assessment of available means within the network as carried out through repeated interactions with both existing and new network ties’. It is evident in this study that the academics’ networking was driven by the means at hand especially with regards to their competences and capabilities prior to embarking on various collaborations. However, because the focus of this study has been on the initiation stages of the networks analysed, whether they used ‘existing or new’ ties was not evident—suffice to say, in this study each network tie was analysed as if new.

Proposition 2 by Engel et al. (2017) states that, ‘under uncertainty, negotiations over pre-commitments are informed by entrepreneurs’ networking actions as driven by both collective and self-interest and as restricted by a predetermined level of affordable loss’. This was also evident in our study given that the academics’ focus was on the ability to build something together capitalising on both self and collective interests. Proposition 3 states that ‘under uncertainty, effectual networking changes the portfolio of ties who commit to co-create the venture, thereby generating unexpected contingencies and enabling the serendipitous emergence of new entrepreneurial goals’ (Engel et al. 2017, p. 47). This was also evident in the data with effectual approaches leading to exploration of new ideas. Though not explored in this study, it was suggested that as the network evolved, the tendency to be causal was more prominent as more intentional and directed decisions were made.

To the extent that most of the international networks of academics are with other academic scientists, effectuation was apparent. These networks were usually explorative in nature where intersections in research interests were sought between collaborating academics to explore new ideas (Sarasvathy 2005; March 1991). On the regional level, a mixture of the two approaches were observed. With the balance of collaborators tilted more to the industry side, a higher tendency to employ a causal approach to exploit opportunities for enhancing academics’ research objectives was observed.

The usage of a more causal approach at the regional level has some implications, and may mitigate the potentials of UICs. Whereas a causal approach to networking is projected to lead to securing the needed resources for achieving given venture goals, effectual approaches lead to more serendipitous outcomes that result in new or modified goals (Engel et al. 2017). The contributions of universities to regional innovation may be mitigated if regional networks assume a purely causal form. For the purpose of regional innovation it would appear that effectual processes that promote more emergent and unordered networking approaches focused on co-creation (Sarasvathy 2005) need to be encouraged. This is important for leveraging the effect of circumstances and unexpected surprises when networking (Engel et al. 2017) and would allow for identification of new ideas and opportunities. For the specific case of oil-rich Stavanger, more effectual processes that promote serendipitous outcomes involving resources and ideas are key to exploring new paths relevant for the oil industry.

Contrary to the finding by Schreier and Senn (2018) on expatriate entrepreneurs, the expatriate academics mostly employed a causal approach for building their industry networks. This was especially the case as proposals and research ideas were needed to get industry interested in academics’ work having moved into new territory. While describing how their various networks were initiated however, it was evident that neither causation nor effectuation were consistent with the initiation stages of a tie. It all depended on differing factors such as the type of tie (i.e. industry or academia) and at what level the tie was being initiated (i.e. project level or individual level). Further, the swap between the two reasoning approaches was portrayed in a non-linear fashion. For example, a tie could start off as effectual and become more causal when the research objectives were decided, but once objectives were achieved it was possible to revert to the more explorative approach consistent with effectual reasoning for finding additional areas to research.

As evidenced from this study, academic scientists exhibited an ambidextrous capacity to swap between effectuation and causation depending on the particular circumstance. This capacity seems to have been necessitated by the heterogeneous makeup of their network ties, being: level of ties (individual or project-based), type of ties (industry or academic) and geography of ties (local or international). We argue that, in so far as academic networks exhibit heterogeneous characteristics (of the nature of ties formed), the greater the need to possess and exhibit the ability to swap between causal and effectual tendencies. This adaptability enables academic scientists to initiate and maintain ties with different contacts.

While the study contributes to the existing literature, it is case-specific and the generalizability of the findings reported here is limited. Because this study has focused on a single case university and faculty, the networking approaches reported may not be fully representative of academics in a wider range of contexts and settings. After all, the decision to exploit an opportunity is very individual-specific (Venkataraman 1997; Hayek 1945). It would also have been worth exploring the possibility to link these findings to the wider network structure, in order to explore what happens in a dense or sparse network. A comparison between the usage of causation and effectuation and network evolution would also shed some useful insights—further research could focus on unearthing these.

A next step would be to compare the reported findings with other groups of academics from different contexts (e.g. academic backgrounds and countries). Selecting cases from different regional settings would also be important for examining the effect of the regional context on the approach taken by academics when networking. Further, because of the strong linkages of UiS to the oil industry, most of the academics’ industry networks were linked to oil firms. Would different types of firms impose a different kind of effect on the choice between effectuation and causation? Future research could thus focus on obtaining a broader range of evidence based on different contexts.