Keywords

1 Introduction

Given the complexity and rapid change that characterizes work and working environments in our advanced knowledge society, personal capacities for professional growth and continuous learning are crucial for professionals. Yet, in many cases a traditional cognitive approach to professional learning does not suffice any more, that is, cumulative acquisition of knowledge and augmentation of expertise by an individual . In other words, professionals in and around the workplace can no longer solely rely on their individual competencies (Tynjälä, 2008). To cope with changing requirements and complicated professional problems, professionals must increasingly share their knowledge and engage in collaborative activities. As a response, both practice and research are paying increasing attention to the social and relational side of professional learning, reflecting the urge for professionals to continually interact and connect (Boshuizen et al., 2004; Hakkarainen et al., 2004). As such, professionals’ learning is not only shaped by their know-what (i.e., declarative knowledge) and know-how (i.e., procedural knowledge), but also by their know-who (i.e., relational knowledge about who knows what) (Borgatti & Cross, 2003).

In this chapter, we set forward a social network perspective to demonstrate how this know-who or the relational side of professional learning can be unraveled further. Network research embraces a distinct perspective that focuses on relationships among actors, which can be individuals, work units or organizations (Brass et al., 2004). According to a social network perspective, actors are embedded within networks of interconnected relationships that provide opportunities for and constraints on learning and development. A social network perspective attempts to capture interactions in a more straightforward, visual and fine-grained way (Borgatti et al., 2013). Namely, the key assumption underlying a network perspective is that the patterns and quality of social relationships (i.e. networks) offer a valuable framework to examine how, whether and to what degree interaction takes place. As such, taking a network perspective on professional learning entails that how individuals and organizations learn and develop is significantly affected by the way they are tied into a larger web of social connections (Wasserman & Faust, 1994). This perspective differs from traditional perspectives in that it focuses on the web of interactions that surrounds actors, rather than on individual actors in isolation.

In specific, the chapter is set up as follows: First, we demonstrate how a social network perspective offers a theoretical and methodological framework, and various tools for an in-depth examination of interactions. Second, we reflect on the added value a network perspective brings to the existing body of professional learning and development research. We hereby build on the cumulative body of research adopting a relational or social perspective on professional learning and development in and around the workplace. Then, we elaborate on social network theory and its most central ideas and approaches. Building on this framework, we look back at the existing body of studies relating networks and professional learning. We conclude by looking forward, and set up several avenues for future research taking a network perspective on professional learning.

1.1 The Added Value of a Social Network Perspective on Workplace Learning and Development

Now, what does a social network perspective add? More specifically, what does a social network perspective have to offer to study professionals ’ learning and development in and around the workplace? We argue that a social network perspective may contribute at least in two important ways (Borgatti & Cross, 2003; Coburn & Russell, 2008; Daly, 2010; Moolenaar, 2012).

First, social network theory provides a powerful, analytical framework and mechanisms that allow for a detailed investigation of the nature, antecedents, and outcomes of interactions (for reviews see Borgatti & Foster, 2003; Brass et al., 2014). Through webs of relationships or ‘networks’, professionals and organizations can exchange knowledge, information, materials and other resources regarding their practice. A social network perspective foregrounds the importance of social interactions for achieving individual and collective learning. To date, a solid framework has developed, comprising theoretical concepts such as structural holes (Burt, 1992), closeness centrality (Freeman, 1979), structural equivalence (Lorrain & White, 1971), and the strength of ties (Granovetter, 1973). Using this framework, social network studies have related professionals’ relationships or network position to significant outcomes such as leadership (e.g., Carter et al., 2015), employability (e.g., Gerken et al., 2016), development (e.g., Dobrow et al., 2012), performance (e.g., Mehra et al., 2001; Sparrowe et al., 2001), and innovation (e.g., Baer et al., 2015).

Second, research on social networks builds on a long tradition of advanced and rigorous methodology and visualization to study interactions. Social network research is multilevel by nature as it allows simultaneous investigation of different levels of analysis (e.g., teachers in schools, or employees in teams). It thereby takes into account the nested structure of data, and includes a level of analysis that is often overlooked, namely the relational level. The levels of analysis can concern, for instance, interpersonal, team , interunit, and (inter)organization level interaction (Brass et al., 2004). In other words: “by embedding individual behaviors in the pattern of their interpersonal relationships, social network analysis can capture the multilevel nature of interaction to an extent that conventional methods and measures cannot” (Moolenaar, 2012, p. 9).

A major challenge for workplace research focusing on interaction and collaboration is that it has been interpreted in a very broad sense. We will now discuss how social network research attempts to capture interactions in a more straightforward and fine-grained way. And, as such, meets several conceptual and methodological challenges posed by the existing body research; such as studies on communities of practice , organizational (shared, collaborative) learning, and professional (learning) communities (Stoll et al., 2006; Wenger et al., 2002).

First, the growing body of research focusing on the social aspect of workplace learning has mostly concentrated on interactions in general. However, a gap in the extant literature is that most studies fail to measure professional interactions with much precision (Coburn et al., 2012). They describe interactions as a whole by providing descriptions, for example, using frequency indications of how often they reported a certain type of interaction . Yet, they do not actually report on differences and nuances of interactions in detail (e.g., the strength and quality of different relationships). Nor do they explore the nature and constellation of interactions (e.g., the diversity and spread of interactions). Adopting a social network perspective provides a more fine-grained exploration of professional interactions, yielding a better understanding of professional learning. This more fine-grained or in-depth insight into interactions is obtained by precisely measuring e.g. the strength, frequency or quality of each relationship in a network; instead of offering an overall description of ‘the relationship in general’. We will further illustrate our point by discussing and visualizing specific research examples in the next section.

Second, professional learning research typically assumes that the locus of professional communities is set by formal boundaries (Coburn & Russell, 2008), focusing on formal organizational boundaries such as teams, departments or workplaces. Yet, a professional is often embedded in a network of relationships that span subgroups and include individuals inside and outside organizational boundaries. Professionals increasingly face a need to engage in knowledge sharing and collaboration through multi-professional networks and teams. Consequently, scholars increasingly argue to not only pay attention to bounded communities but to also include professional interactions across boundaries of communities (Hodkinson et al., 2008; Wenger et al., 2011). A social network perspective allows simultaneous examination of individuals and the (sub)units they are nested in (e.g., professionals in functional teams), within and across organizational boundaries.

Third, traditional professional learning research has few techniques or tools at its disposal to visualize interactions in detail. Social network analysis provides a variety of tools and techniques to reveal partially hidden or informal social structures and relationships (de Laat & Schreurs, 2013; Hakkarainen et al., 2017). Recent work has extensively demonstrated the use of straightforward network visuals to promote and support professional learning processes of individuals and organizations (Hogan et al., 2007; Van Waes & Van den Bossche, 2020). Visualization of interactions have not only proven useful for scientific purposes, but are also a valuable tool to translate findings to practice; for example, to design interventions or when giving feedback on interactional data to practitioners, policy makers or managers (Cross et al., 2010).

Fourth, a network perspective enables to examine interactions taking into account the multiple levels at stake. Network analysis may concurrently consider the individual level, the dyadic or relational level, and the (sub)group or organizational level. For example, research questions may simultaneously address characteristics of an individual professional, the relationships s/he has with colleagues, within or between teams in the organization.

1.2 Social Network Research and Approaches

Now that we have argued why a social network perspective offers an added value to research on workplace learning and development, we move more deeply into social network research itself and the two major research approaches that can be adopted.

In recent years social network research has been firmly established as a major research area, and the number of publications referencing social network research is exploding (Borgatti et al., 2014). Networks consist of relationships, which are termed ties or links, between actors which are called nodes. Actors can be individuals or collectivities, such as teams, organizations or countries. The central focus of social network theory is on relationships and interactions as an explanation of actor and network outcomes. This in contrast to traditional or individualist explanations that focus on attributes of actors that are treated as astructural and independent cases. This reflects a shift from attributes to relations; or from monadic variables (attributes of individuals) to dyadic variables (attributes of pairs of individuals), which consist of social relations and recurring interactions. The fundamental unit of analysis is the pair of actors rather than the individual (Borgatti et al., 2014).

In social network research two fundamental kinds of network research approaches can be discerned, a whole network and a personal network approach (see Fig. 8.1). Both approaches have their specific focus and merits (Borgatti et al., 2013), and offer different insights into professional learning processes. We illustrate this point with an example: Fig. 8.1 illustrates Holly’s personal network (encircled in red), extracted from the whole network. Holly will receive different information from her professional network, compared to Lee, as she occupies a bridging position between two groups. Her boundary-crossing interactions with two different groups may offer her more learning opportunities.

Fig. 8.1
figure 1

Example of a personal network (encircled) extracted from a whole network. (Halgin & Borgatti, 2012, p. 38)

Full size image

In whole network or socio-centric research, the ties among all pairs of nodes in a bounded group are studied (e.g., all teachers within a school). A whole network approach allows researchers to analyze patterns of connections, including structural features such as centrality, density, and betweenness (cf. infra). A personal network or ego-network approach involves systematically mapping social relationships of focal individuals, termed ego’s; and determining the set of nodes that ego has ties with, ‘alters’ (Crossley et al., 2015) (e.g., a focal teacher and his/her contacts). A personal network approach allows participants to define their own network boundaries, as it intends to investigate the ties of individuals across boundaries of communities, practices and locations (e.g., all ties of a focal teacher within or outside the classroom, grade, or school).

It is important to highlight that personal network studies answer different research questions compared to whole network studies, and as such may offer different insights into how professionals learn and develop. For example, Hytönen et al. (2014a, b) used a whole network approach to examine the development of energy efficient experts’ professional networks in the context of a continuing professional education in an emerging field. They aimed to understand networking processes and activities at different levels, that is, among all participants, at a small-group level and at an individual level. The results revealed differences in networking activity at the different levels. Their study showed that even though the intensity of professional knowledge exchange might be low among all professionals of a multi-professional network, intensive networking activities can take place among smaller groups and between individual actors and, thus, provide important resources for participants. Later, Hytönen et al. (2014a, b) adopted a personal network approach to identify the key actors (or cognitively central actors) of the energy efficiency experts’ professional networks to understand why certain people achieved essential roles in knowledge exchange in multi-professional networks (Hytönen et al., 2014a). The study demonstrated that in multi-professional networks, the cognitive centrality of an actor is for the most part related to social context, that is, how the expert ’s profile fits into the wider professional context.

Van Waes (2017) adopted a personal network approach to study university teachers’ networks. A personal network approach uncovered how features of a unique university teacher ’s network related to variables at the individual level of analysis, such as their professional development and expertise. For instance, their work showed how experienced experts developed larger and more diverse networks, compared to experienced non-expert colleagues whose small networks showed little diversity. This shows how a network perspective allows a deeper, more fine-grained exploration of interactions. The use of a personal network perspective sheds light on the fact that professional development is not a time-age effect as experienced experts seem to lapse into arrested development, linked to limited network input. This may in turn cause isolation (Bakkenes et al., 1999; Ericsson, 2006), resorting to interactions that require low interdependence. This arrested or stagnated development is associated with automaticity, i.e., their behavior becomes routine and reaches a stable plateau without further improvement (Ericsson, 2006).

2 Looking Back: Extant Research on Networks and Professional Learning

We further underpin our thinking by discussing exemplary research using a social network perspective to shed light on professional learning. We discuss how structural network features may affect the flow of resources between people, and how relational network features influence which resources are available from what kind of people.

2.1 Structural Network Features

A basic structural concept used in social network studies is density. Density characterizes the general cohesion of the network, that is, the number of existing networking ties in relation to all possible ties. This implies that the greater the proportion of ties in the network, the more dense the network is. Studies show that density increases the rate, extent and fidelity of knowledge diffusion in networks (Singh, 2005). Therefore, people with denser networks might have more diverse access to resources as they have a higher number of connections. Density is often used to examine changes taking place in networks, such as increasing or decreasing number of ties in different contexts. For instance, it is often taken for granted that professional education and training supports the development of networking ties among participants. However, recent studies have shown that the emergence of professional learning networks is not always straightforward (Hytönen et al., 2014b; Rienties et al., 2014). The development of professional learning ties does not take place automatically or without careful planning. Deliberate efforts as well as well-developed operating models are required to support tie development (Rienties et al., 2014). This specifically seems to be the case if participants come from different backgrounds and represent heterogeneous expertise (Hytönen et al., 2014b).

Another basic structural concept is centralization in social networks. Centralization can be studied by focusing on centrality that characterizes an individual actor’s position in a network, or centralization of a network structure. Centrality values indicate the amount of information that a person provides to other network members. Therefore, it has been used as indicator for actors’ importance or popularity in the network (Sparrowe et al., 2001). Degree centrality is probably the best known and straightforward form of centrality. It is measured by calculating ‘in-degree’ and ‘out-degree’. In-degree captures the amount of people who seek an individual out for resources (by peer-evaluation). The more someone is nominated as a valuable resource in the network, the higher the in-degree. Out-degree stands for the number of times an individual reaches out for resources (by self-evaluation). In professional learning studies centrality measures have been used in searching for, for example, key persons in professional networks or identifying actors’ different knowledge mediating roles (Hytönen et al., 2014a; Palonen et al., 2004). Network research on newcomers’ networking roles has demonstrated that newcomers and young workers can very quickly achieve a central networking position in a professional community and become important knowledge-mediating actors (Hytönen et al., 2011).

The concept of brokerage refers to persons who are positioned in between people who themselves are not directly connected (Burt et al., 2013). These brokers are considered valuable networking partners as they have access to versatile repositories of knowledge through their connections (Palonen et al., 2004). Studies have revealed that in professional communities these key persons are sought for professional help, advice and support more often than other professionals (Hytönen et al., 2014a). Therefore, the key persons with strong brokerage roles or knowledge mediating roles are often described as ‘stars’, ‘hubs’, ‘gatekeepers of knowledge’ or ‘cognitive central participants’. They bridge structural holes, i.e. holes in the social structure that result from absent or weaker connections, by building connections and mediating knowledge across different people and different knowledge cultures (Burt, 1992). In professional communities and networks, these persons are seen to connect people facing similar professional problems; to translate knowledge across different knowledge cultures and disciplines as well as facilitating innovations, new operational models and professional practices (Sverrisson, 2001). Consequently, they have influential roles in professional learning processes for individuals and organizations. Their role as knowledge mediators seems to be especially important in emerging and developing fields in which the knowledge base is not yet stable or consolidated (Hytönen et al., 2014a).

2.2 Relational Network Features

Most extant network research focuses on the patterns or structure of networks (e.g., density, centrality). While that is important, often questions about the content, meaning and significance of relationships are less examined (Bellotti, 2014; Borgatti et al., 2014). Consequently, studies increasingly focus on examining what kinds of relational features are related to structural network qualities (Froehlich et al., 2020). Many network studies have analyzed how professional connections with different qualities assist in sharing knowledge and competence.

A central concept used in examining the exchange of resources is tie strength. Tie strength indicates the closeness or strength of relationships by measuring for instance the frequency, intensity, reciprocity, depth, or time spent in a relationship (Marsden & Campbell, 1984). Strong ties connect to people that are close, whereas weak ties are looser contacts. Both weak and strong ties provide access to different kinds of resources and, thus, have different roles in professional conduct and learning. For example, strong ties are instrumental in the diffusion of innovation , the transfer of tacit or complex information, as well as solving complicated problems and transferring knowledge between organizational units (Reagans & McEvily, 2003; Uzzi, 1996). Palonen et al. (2004) demonstrated that novel and complex knowledge is not easily transmitted without strong reciprocal ties. Strong social networks are also associated with increased individual and organizational performance (Burt, 1992; Hansen, 1999). In contrast, weak ties are more likely to bridge socially distant parts of a network, and thus more likely to gain access to new resources (Granovetter, 1973). They play an important role in the formation of novel ideas and non-redundant information (Levin & Cross, 2004)

An important principle guiding network formation is homophily. Social networks are often homogeneous in nature meaning that people tend to interact and create strong connections with people who have characteristics similar to their own, such as age, gender, educational background, equal work status or occupational group (McPherson et al., 2001). Homophily influences the information people receive, the attitudes they form, and the interactions they experience . For example, people with more diverse networks demonstrate more innovation (Kilduff & Krackhardt, 2008). Whereas homophily or similarity between people may enhance the decay of their networks, as information or knowledge may become redundant (Burt, 2000). Networks including a rich variety of people and reaching over the borders of professionals’ immediate working environments and communities are especially important for coping in changing working life. Diversity in professional networks has been associated with expertise development with experienced professionals, whereas experienced non-experts display more relationships with people with similar characteristics (Van Waes et al., 2015).

Scholars increasingly emphasize the quality of the content that flows through network ties, or the ‘stories’ behind networks (Baker-Doyle, 2015). Researchers have adopted qualitative network techniques as they offer extensive explorative powers to examine the nature, the meaning, intensity, and depth of interactions (Fuhse & Mützel, 2011; Hollstein, 2011). Qualitative network data enable us to examine whether interactions between people involve, for example, swapping entertaining stories, exchanging basic information, or collaborating intensively on shared products. They also allow us to investigate the in- or interdependency between people, the depth of their exchanges, and the impact of their interactions on professional learning. Work by Coburn (Coburn & Russell, 2008; Coburn et al., 2012) demonstrated how the depth of interactions in networks determined the extent to which innovative learning processes succeeded. Recent research by Van Waes et al. (2016) showed that experienced experts had more high interdependent interactions (joint work, sharing), compared to experienced non-experts, who described more independent (practical, organizing) talk (see Figs. 8.2 and 8.3).Footnote 1 The quality of ties also differed in that the experienced experts had more high interdependent relationships in their networks and created more value , in comparison to their experienced colleagues with lower expertise. This shows how a network perspective allows a deeper, more fine-grained exploration of interactions. A promising venue to further uncover the social side of professional learning in its totality, is a mixed method network approach, using visuals such as the concentric circle method (Van Waes & Van den Bossche, 2020). Mixed method network research is gaining increasing terrain (Domínguez & Hollstein, 2014; Froehlich et al., 2020), and allows to address research questions that interact both structural and relational network features.

Fig. 8.2
figure 2

Personal network map of experienced expert teacher

Full size image
Fig. 8.3
figure 3

Personal network map of experienced non-expert teacher

Full size image

3 Looking Forward: Future Research on Networks and Professional Learning

The existing body of network studies examining professional learning opens up new avenues for research. In this section we identify several areas ripe for further exploration.

3.1 Further Uncovering Relations Between Professional Learning and Network Development

Network ties are often implicitly regarded as ‘learning ties’. Yet, installing networks or communities does not necessarily mean that the professionals within are actually learning. Although it is often assumed indirectly, an increasing number of ties in the network does not automatically imply that new knowledge is created, that learning has taken place, or that professionals’ development is impacted. A dropped tie does not always imply that knowledge is lost, and a kept tie may not add anything or become redundant. The research designs of most studies are not designed to examine this causality between network development (social capital) and professionals’ learning (human capital) directly. Causality between the relationships under study is often funded indirectly by suggestions from the reviewed organizational literature on network development and professional learning. As such, future studies designed to examine directional links between network development and professional learning will be important. Future studies should aim to understand the nature of a learning tie, that is, when and how learning occurs in a network or a network connection. Moreover, setting up lines of longitudinal research would allow to study patterns of development of learning ties in professional networks. For instance, tracking networks over multiple measurement moments may help to discern profiles or patterns in the development of human and social capital. Longitudinal network data and performance measures could be collected at several time points to shed further light on, for example, how professional networks of experts unfold, and which the crucial elements are in this development over time.

3.2 Paying Attention to the Interface of Education and Workplaces

Professional learning research has often focused on studying social structures, communication and knowledge flows within a single unit, such as an organization or team (Phelps et al., 2012). Even though workplaces are important places for facilitating professional learning, many professionals -acting in rapidly changing working environments and dealing with complex multifaceted professional problems- need to rely on their networks reaching beyond the boundaries of their immediate workplace organizations and traditional institutional resources in order to support the development of skills and competencies (Nardi et al., 2000). In this regard, recent studies emphasize the importance of deeper interactions and more strategic cooperation between educational institutions and workplaces that are often seen as separate and distant from each other (Harteis et al., 2014; Hytönen, 2016; Hytönen & Kovalainen, 2020). New kinds of efforts to bridge and combine expertise of and stronger professional connections between educational institutions are needed to meet future educational challenges and to provide flexible possibilities for professionals to update and expand their expertise. These could generate new environments, to cultivate skills, and to share and receive critical knowledge between people with different types of expertise and professional competencies (Roxå et al., 2011).

As earlier network studies have shown, the development of comprehensive professional social networks and occupational knowledge exchange forums do not take place automatically at the learning environments organized by educational institutions and workplaces (Hytönen et al., 2014a, b). Future network studies could help to understand better the interfaces of education and working life by providing knowledge on whether and how professional learning networks and connections are constructed and how their development could be supported in sustainable and productive way. Furthermore, future network studies should focus on examining how can learning in educational institutions and workplaces be better integrated and communication across organizational boundaries, different professional cultures and multi-professional networks facilitated. Future research could also help to overcome the question of how are individuals able to connect expertise from one specific field with the diverse expertise of their multi-professional networks. Future network studies should focus on examining the interface of education and workplaces, and more broadly multi-professional networks crossing the boundaries of workplace communities and organizations. Potential questions to be addressed in future research are: How can learning in educational institutions and workplaces be better integrated, and the development of effective learning networks more supported? How are individuals able to connect expertise from one specific field with the diverse expertise of their multi-professional networks? Future research could also help to overcome the obstacles that are related to developing actual interconnections and relations between theoretical and practical knowledge cultures.

3.3 Not ‘Just’ Structure… Making Room for Network Agency in Professional Learning

Traditionally, network research considers changes in networks as resulting from an interplay between self-organizing properties of networks, that is, networks develop because of the properties they have and the way they are structurally embedded in the larger network (Agneessens & Wittek, 2012; Brennecke & Rank, 2016). For example, if someone offers you help, you are likely to reciprocate this tie to maintain the structural balance. Social theorists have long been discussing the relative contributions of structure and human agency to social interactions and network dynamics (Bourdieu, 1986; Giddens, 1984). Some scholars have recently questioned whether structure has overwhelmed agency in empirical network studies (Gulati & Srivastava, 2014). If actors can intentionally affect their network, one may wonder whether a causal focus on structure and self-organizing properties of networks can be justified.

Few studies have examined the relation between network agency and professional learning. Professional learning in changing working environments is to great degree embedded in deliberate creation and cultivation of network relations. Exemplary, is the increased value attributed to networked expertise or relational expertise in and around workplaces (Hakkarainen et al., 2017). Research by Van Waes et al. (2015, 2016) demonstrated how experts displayed higher agency, as they described to frequently re-evaluate their networks and to act intentionally on them. Apparently, they somehow ‘learned’ to manage their network. The underlying assumption in this type of studies is that individuals, who are aware of their networks and the resources and expertise residing in it, are more likely to reach out to the ‘right’ people at the ‘right’ time when presented with challenges or opportunities (Borgatti & Cross, 2003). Professionals who consciously act to strengthen their network, display what is recently coined as ‘network intentionality’ (Moolenaar et al., 2014), that is, agency in forming, maintaining, activating, and dissolving relations to gain access to resources for the mutual benefit of oneself and others, given their own cognitions of what makes for a ‘good’ network (Nardi et al., 2002).

Future research should challenge traditional network research by further uncovering the role of network agency in professional learning processes. For instance, it would be valuable to link the existing body of research around information and feedback seeking at the workplace with the concept of network agency. Information and feedback seeking are often regarded as individual undertakings and the role of network or relational agency is often underexposed (Ashford & Cummings, 1983; Van den Bossche et al., 2014). This also holds for information seeking in newcomer socialization processes (Morrison, 2002; Saks et al., 2011). Setting up a line of research considering a network (agency) perspective would help us to answer questions like: Which are potential barriers to the development of network agency in professional contexts? Can we support feedback and information seeking by supporting professionals’ network agency? What are good ‘beginners’ networks’ for newcomers in workplaces in terms of the pre-existing properties of networks? This sheds light on another aspect of learning: How to learn to become a professional networker, and what does this entail? Which knowledge, skills or attitudes are necessary in enhancing network agency in professionals and companies?

3.4 Designing Network Interventions and Using Network Visualizations as Feedback Tools

In recent years, both practitioners and researchers have also started to consider the design of effective initiatives to enhance the value of collaboration (Cross et al., 2010; Cullen-Lester et al., 2016). These ‘network interventions’ may include (research-based) coaching or consulting activities, or organizational development activities in general. Network interventions are purposeful efforts to use social network data to accelerate behavior change , to improve performance, or diffuse innovations (Valente, 2012). They are designed to support professionals and organizations to intentionally act on their networks (Cross & Thomas, 2009; Parise, 2007). In intervention research, social network methodology is used as a mapping tool to render professionals’ networks visible (Jaspersen & Stein, 2019). Network visualizations can make the characteristics of professional networks available for assessment . For instance, scholars have provided evidence that professionals who learned the properties of an effective network (‘teaching to see social capital’), achieved greater performance and career advancement (Burt & Ronchi, 2007).

However, the mere mapping of networks does not necessarily provide a clear path to intervention. More information is needed on how to encourage the development of strong networks when they do not exist or how to sustain them when they do (Coburn et al., 2010). Preliminary research has shown that mapping informal networks using social network analysis can detect multiple (isolated) networks in organizations, connect ideas, and facilitate value creation (Cross et al., 2010; de Laat & Schreurs, 2013). Studies also showed how network agency may constitute a supporting mechanism for network change . This work suggests that network agency can be fostered through intervention by raising network awareness (Van Waes et al., 2018a, b). Future research into the design and timing of network interventions could yield further insight into how to foster learning through interventions in different workplaces.

3.5 Exploring ‘The Dark Side’ of Professional Networks

Social network research strongly emphasizes its positive consequences. However, one should be careful to interpret all ties as prosocial and favorable (Portes, 1998). Several scholars have argued that negative or challenging relationships may be even more consequential for professional learning and may outweigh the effects of positive ties (Everett & Borgatti, 2014). Existing research sheds light on questions about how less favorable network constellations, and negative or so-called ‘difficult ties’ develop. These concern relationships in which you have to exert significant extra effort to communicate, share perspectives, or come to a common understanding about important topics (Daly et al., 2015), e.g. disliking ties, difficult collaboration ties, no-friend ties. These negative relationships would have greater power than positive relationships to explain workplace outcomes, which is termed ‘negative asymmetry’ (Labianca & Brass, 2006). Negative relationships are also related to organizationally relevant outcomes such as lower individual performance, decreased satisfaction with one’s group, and lower organizational attachment (Sparrowe et al., 2001; Venkataramani et al., 2013). For instance, individuals who dislike someone are unlikely to seek advice from the person they dislike, even if that person is highly competent (Casciaro & Lobo, 2008). Evidence is mounting that negative relationship ties can create liabilities for individuals in organizations both because resources are sometimes withheld from them, but also because negative flows are directed toward them (Marineau et al., 2016). Researchers further suggests how professional culture may hinder interactions (Roxå et al., 2011), or how lack of physical proximity can make for very isolated professionals (Spillane et al., 2017). Studies have shown that small networks lacking diversity in composition relate to arrested development (Van Waes et al., 2015), and that perceiving little value in one’s personal network may be detrimental for expertise development (Van Waes et al., 2016).

To date, few studies have provided in-depth examinations of this less favorable sides of networks, as network surveys generally probe for positive relationships (such as friendship, trust, presence of professional ties). Future research increasing insight into the formation of negative or difficult ties may enhance our knowledge around less favorable constellations of networks for professional learning in workplaces and professional communities. For example, why some people are able to sustain joint work interactions, and while others mostly resort to superficial interactions at the workplace and stagnate in their development. Such a line of research will also inform organizations on preventing isolation and development of negative silo’s, and in supporting the development of favorable network configurations.