Journal of Knowledge Management Practice, June 2005

The Knowledge Network: A Fundamentally New (Relational) Approach

To Knowledge Management & The Study Of Complex Co-Dependent Organizations

Eugene G. Kowch, University of Calgary


Knowledge Management (KM) thought has evolved from combined management, information technology (IT) and social science research, and the addition of relational network theory to the field could add important analytical tools for both KM scholars and practitioners. First, this author surveys the dominant ontological stances behind much of the KM literature. Next, the paper proposes that we inform and augment current KM thought by including more subjective neo-institutional, relational social science models (relational networks) to broaden and deepen KM theory and praxis. Knowledge network analytical elements, based on a collection of social and policy network theory give us the ability to render complex pan-institutional relations more simply, making these techniques powerful leadership tools. Presenting findings and examples from a multiple case study of complex knowledge organization, this author demonstrates that knowledge network characteristics such as capacity (to organize) can be described, and used as inputs to powerful relational networks designs across large institutions. Knowledge network relational frameworks must be further developed and tested to help institutional leadership in our increasingly co-dependent, technology integrated businesses, corporations and large organizations in the knowledge era.

1. Introduction

Easterby-Smith & Lyles (2003) summarize in the Blackwell Handbook of Organization Learning and Knowledge Management to tell us that the future major trends in KM research seem to be “an increasing emphasis on social capital… networks and communities, and increased emphasis on research design” (p. 645). These authors call for more research on knowledge creation and transfer, and for a better understanding of how communities develop their organizational learning and knowledge management capabilities/capacities (p. 650). The author of this paper suggests that by combining social, policy, organization and knowledge management concepts, knowledge networks can be identified, characterized and designed.

This contribution made by this paper is to some connect social and policy theory with management theory to suggest some key theoretical and practical knowledge networks constructs. First, the author presents an analysis of the approach to organizational knowledge taken by the management discipline. This is followed by a review of literature from other disciplines that contribute to knowledge management literature to discover the dominant ontologies or “world views” behind these foundational works. As such, a rather uncommon blend of social, political and organization science literature, as well as institutional network theory literature offers a wider view of how we come to know knowledge management and relational networks today. With this broad multidisciplinary background, the author then categorizes the sweep of research approaches to this literature along the basic Cartesian ontological continuum (ranging from objective to subjective stances), for combining fields of thought requires attention to ideological perspectives as well. Network study has deep roots in complexity theory, and today we must render the complex simple so as to manage it well. Hence, network theory is more useful today than ever before as we consider how organizations, people and knowledge are really connected (Barabasi, 2003).

Networked knowledge is then presented along with examples from recent research (Kowch, 2003). Networked knowledge can be studied by the use of various analytical lenses to reveal key knowledge network characteristics. This technique is common methodology used in the analysis of policy in complex situations (Lane, 2000). This understanding is achieved through levels of descriptions and analyses of network actors or nodes (at the micro level) network interest organization, (at the meso level) and environmental conditions (at the macro level). This meso or institutional level analysis is particularly important work because modern organizational theorists and practitioners urge that today post functionalist organizations are better suited to be sustainable as the organizations of the future (Collins, 2001; Drucker, 1996; Alvesson & Deetz, 1996). In conclusion, the author offer presents research study findings from a multiple case study that successfully, and for the first time, defined key characteristics of complex knowledge networks at the institutional level.

2. Knowledge, Networks And Social Capital: Multi Disciplinary Literature

2.1. Management And Information Technology: Surveying Mainstream KM And IT Field Perspectives.

Organizational knowledge, learning and capabilities form a triangle: the ongoing development of organizational knowledge is, or can be, a dynamic capability that leads to continuous organizational learning and further development of knowledge assets (Tsoukas & Mylonopoulos, 2004). In the last decade, management and business school literature has led to renewed interest in the description and analysis of organizations as a collection of knowledge systems which are bundles of knowledge assets, the effective management of which affords firms competitive advantage (Choo & Bontis, 2002). Knowledge management (KM) is primarily defined as collections of knowledge assets where the system of integration, updating, maintenance and management of those assets is of great importance (Snowden, 2002; Nahapiet & Ghoshal, 1998). This paper suggests that those assets are relational, complex and that they can be understood better. Indeed, knowledge management case study research continues to describe, explore and model both the theory and practice of KM, often as “systemic, complex processes of knowledge creation, transmission, storage and retrieval” (Vince et al, 2002). Recently, specific networked knowledge-involved processes have been studied by researchers taking a “human relations” organization perspective. In this research, the unit of analysis is the organization as a relational network (collection) of people who interact and work to generate or manage knowledge (Van Wijk et al., 2003; Burt, 1995).

Other management scholars take a different approach to research networked knowledge processes, adopting a common “structural perspective” (Ghoshal & Bartlett, 1999). These scholars assume that the way organizations work is both systematic and complex, yet the models for structural organizations are quite simple patterns. As a reaction to criticisms that these models are less descriptive and discriminating than we need for the modern complex organization because too often they do not describe what is ‘really’ happening, the concept of social capital and post structural organization theories have evolved (Putnam, 2001; Alvesson & Deetz, 1996). Human or “social” relational networks are key components in the macro concept of social capital, and some management and political/public administration scholars now use this type of conceptual framework to study the functions and outputs of (knowledge) happening via networks to define relational and structural capital (Putnam, 2001; Nahapiet & Ghoshal, 1998; Coleman, 1988). Processes do have value. By combining network perspectives from different disciplines we can define networks by their aggregate functions (output), and by the kinds of the transactions or relations found (for example, to track influence patterns, reciprocal communication, trust, respect for expertise). So while networks are structures of sorts, their function and character can tell us a lot about incredibly complex organizing and knowledge flow in today's large institutions (Barabasi, 2003). These knowledge structures may not be stable or even planned, but they occur happen frequently.

From policy network theory, we know a lot about how policy networks function, and by this framework we can also describe the reason(s) for the emergence of a network. Certain kinds of network structures are somewhat predictable, and sometimes they can be designed to create high capacity and autonomous networks (Atkinson & Coleman, 1996). The same could be true for knowledge networks. We can map and interpret the shape of relational patterns, the nature of the relations in the network – it makes sense that more work could be done in the KM field using this theory to tell us more about how sticky or non-sticky knowledge flows occur. Indeed, as we will see from the research examples at the end of this paper, describing networks structures v may be the first step toward determining the key parameters for designing complex relational knowledge network systems.

Today, some business and IT scholars consider intellectual capital as knowledge that is being commodified and held at the person or micro level (Kogut & Zander, 2003; Polanyi, 1967). Does the sum of actor held intellectual capital add up to the capital held by a relational network? No studies on this have been found, but such work could tell us a lot about how to account and inventory these kinds of capital as they exist in dynamic relational (network) knowledge based organizations. So much KM research is practical or functional in focus, fixed on understanding organizational knowledge as a commodity that is static. This author agrees with Easterby-Smith & Lyles (2003) that knowledge is not static, and it flows in ways that likely add value to it, from the organizational perspective, depending on the capacity of the network to create or use the knowledge for the bottom line. Large sets of instrumental or objective behavioral strategies for organizations or companies have been generated using such older functionalist thought, usually by consultants who are assessing knowledge and its flows in the organizations understood as closed, bureaucratic hierarchical systems (Alvi & Tiwana, 2003; Van Wijk et al, 2003; Wenger, 2002). We know that the organizations of the future will be less hierarchical and closed than those of the past, so KM theory and the related models need to be adjusted for this foundational understanding.

Information technology (IT) scholarly research on the subject of knowledge management is also plentiful today and, if we wish to collect social, business and IT theory as suggested in this paper, we must also ask if the dominant world view in this KM literature is subjective or objective.

IT is understood as support for company based knowledge creation, codification and retrieval, transfer, integration and application of knowledge assets (Alvi & Tiwana, 2003). This body of KM literature provides us with models that can allow some predictive or at least descriptive capability if we accept, at the macro level, that organizations today exist within the knowledge economy context. One example of such IT influenced KM thought is the proposition that the communication capabilities offered by the World Wide Web enables knowledge management processes (knowledge flow) better than possible before (Davenport & Prusak, 2000). In KM scholarship, information technologies are seen as enablers for communication and information handling, so knowledge sharing can be designed and supported by artificial intelligence systems, expert systems, knowledge repositories and careful work flow enhancements. IT literature focused on e-phenomena tends to reveal knowledge as forms of data, where the use of mechanical systems and processes can enhance knowledge handling –this view demonstrates a common IT field perspective on KM as a physical network (computer networks) with appropriate mechanical structures that are to transport “hard” (quantifiable) knowledge data (Robey & Boudreau, 1999; Nonaka & Takeuchi, 1995) according to the intents or needs of the organization for KM. In the body of KM related info systems or IT literature, the concept of “network” is presented as a metaphor for person to person contact and communication / exchange, and information systems or IT professionals have played a significant role in promoting IT support for KM (Scarbrough & Swan, 2003). This has become a thriving industry which is as effective as its abilities to characterize and predict knowledge flows and relations at the individual and organizational level. The IT models depend on accurate organizational, work flow and relational network conceptualizations – one can ask the question “how well do the IT KM models simplify complexity”?

Various management and IT epistemologies converge within the knowledge management field, and both subjective and objective approaches exist. This could be because the metaphor of a network fits intuitively with our views of organizational, social and computer technology interactions. With the convergence of fields in KM, academics can take unique advantage of a potential to better describe the new knowledge management processes required in widely held, dislocated, partnered and global companies as they prepare to survive rapid-change knowledge economy (Wenger, 2002; Drucker, 1996; Lave & Wenger, 1991). We know that relations are important, but they can be messy. By combining policy and social network constructs, we can characterize whoever is doing what with whomever or what firm, to discovering why they come together and why they stay together or come apart. While this kind of less structured model is not mainstream, it is becoming more acceptable in IT schools.

In their excellent qualitative study of KM discourse to 2003, Scarbrough and Swan (2003) claim that the management (literature) focus has also moved to professional (bureaucracy) model as a compromise between economic and business environment pressures. Here again, the KM field exhibits a reaction to models that do not describe complexity well enough. Indeed, since 1980, more management scholars claim declare the limitations of the professional model to accurately describe business, and this has created an imperative for scholars to merge some IT and management thought because of “the growth of so-called ‘knowledge-worker’ occupations (Drucker, 1994) and [because of] technological advances created by the convergence of computing and communications technology (Easterby-Smith, 1998). Overall, the body of management literature in KM draws on a predominantly systems based, functionalist, hierarchical and “professional bureaucracy” view of the processes of organization and human interaction processes, presenting managers as instruments of institutional owners (Scarbrough & Swan, 2003).

With regard to post structural ideas about organizational relations, KM scholars know that we must research and develop more theory about how and why knowledge flows, so our ontological position might need to be a little more subjective and a little less structural to more accurately model complex business realities today (Easterby-Smith & Lyles, 2003;. Scarbrough and Swan, 2003) conclude that IT and HR consultants still tend to commodify knowledge and the KM discourses, and KM practitioners have a propensity to use ‘less academic” KM literature compared to earlier organizational learning literature that preceded it (p. 510). They foist harsh criticism on IT professional discourse as well because a combination of IT field and HRD consultants today are “taking advantage” of the ambiguous literature for profit (p. 510). These researchers go on to claim that HR consultants are “commidifying knowledge itself” in companies, slowing the growth of KM praxis and theory development. This is the same kind of utilitarian thinking that led to Enron’s debacle. These are also the same kinds of theoretical and praxis trends that led to a decline of social and policy network theory in the policy field during the last decade (Thatcher, 1998; Atkinson & Coleman, 1996). More recent developments in policy networks help us solve some of these earlier problems by allowing us to consider what knowledge and influences are drawing people together across structural features (groups, organizations, governments), and by characterizing people and network organization capacity to organize or manage themselves to get work done (Kowch, 2003). This author suggests that a bridge between similarly (too descriptive, not predictive enough) KM discourses can occur by extending KM to include political science (network) theory, to account for increasing degrees of institutional codependence and neo-institutional realities (Lane, 2000; Kowch, 2003).

Modern sociologists and organizational theorists alike concur that IT is an integral part of the processes of organizing in organizations, and they also agree that the success and shape of that integration is a function of the world view (ontology) that both IT and KM theorists hold about information, technology, sociology and organizations in our inescapably linked world depending more on knowledge and the good management of it (Barabasi, 2003; Roberts & Grabowski, 1996; Franklin, 1993). By studying the processes that knowledge network studies can describe (how people organize knowledge), and by focusing on the content of these networks (what is organized or flows between actors) as well as by focusing upon the character of the actors (who has trust, position), we can study how human networks get things done – or how knowledge networks perform (Ibarra, 1993). We need to generate theoretical models (and practical KM designs) that allow us to create high capacity networks and actors, dynamic work and knowledge flow patterns and predictable relational processes by merging IT, management and social sciences in less structural Cartesian ways. Network models might be part of this reconceptualization for KM. This sort of integration of process and product thinking considering post structural social, political and organization / management thinking is essential for us as we should be able to generate architectures or designs for KM networks that exist in the necessarily (linked) organization of the future (Barabasi, 2003; Kowch, 2003 ). Without this development, we may see KM decline as a field of study much as policy network and social networks declined - because the theoretical models were not as descriptive and predictive as the real world required (Lindquist, 1996).

2.2. Social Network, Policy Network (and Policy Community) Literature: A focus on Relations, Organization and Interdependence

2.2.1 Social Networks

Social scientists have taken philosophical and epistemological approaches to make sense of complexity. Social network theory, enjoying rapid development once again with affordable computing, is one school of thought where the complexity of human and organizational relations can be rendered simple so that we can map, describe, interpret and model relation dependent phenomena better. When combined with social policy theory (the study of issues and issue organizations) we can also define the autonomy and capacity of people and institutions to organize themselves. This author posits that this knowledge can help us design new KM architectures because knowledge flows are dependent on a complex interplay between organizations, human interactions and technologies. Next in this paper, we review old social network theory and introduce the elements of policy network theory.

Based on empirical research, Mark Granovetter (1973) proposed a social science model he named the “social network” and proved the functional importance of weak and strong ties that exist in between individuals in patterns between people in society. Contrasting our intuitive sense of ‘social networks’ as cocktail party friends, Granovetter found that weak link networks are more effective than strong link networks, because they allow more knowledge to be accessed, when needed, than tighter, closed networks. In 2004, this work is part of the foundation for resurgence in emerging science focused on how all things in our world, recognizing what reductionist thinking may not have let us contemplate as we ponder knowledge exchange – that people, institutions and governments are related or linked together in different ways at the same second in time (Barabasi, 2003). How do we simplify this complexity? In simple terms, social network study focused originally on the characteristics of the network content - the “what” in the relations that linked people in networks (nodes), and more specifically upon the network actor characteristics or the “who” in networks at the micro (actor) and meso (aggregate network) levels of analysis. The work also focused on the maps or patterns of relations as definition of the structureof a network as a snapshot in time. Wasserman et al suggested (1994) that instead of concentrating as much on communication contact frequency in social networks, scholars should focus as well on both the individual (“who”) micro level, on the meso (institution/organization) level, and on the macro (environment) levels at the same time. “Individual attitudes, beliefs, behaviors create a greater context for social network study, bridging the micro (actor) and macro (social worlds)”. Those interpretive models were primarily functionalist frameworks however, still limiting us to “study the resulting structural creations” from interactions (Wasserman & Galaskiewicz, 1994). To explain this by metaphor, social networks can be understood as the framework in the “wedding tent”, which shapes interaction, but these nets are also a part of the “wedding” itself, – an inevitable part of the social capital creation happening. The main point here is that older social network theory, focused on frequency of contact and impossibly complex maps was not enough to help us characterize how people organized themselves – and why.

By the early 1990s, social network models were adapted as interpretive frameworks for the management field. Burt (1995) studied formal and informal networks, DiMaggio & Powell (1991) situated formal and informal networks within institutions, and Ibarra (1993) proposed a framework to understand institutional network potential for action and innovation by using organization theory concepts (Ibarra, 1993). Krackhardt (1992) further developed metrics to describe actor relation ties, allowing network researchers a way to quantify key structural network components such as centrality (highly central actors can be gatekeepers) and density (a ratio of the number of actors to other number of potential ties) to characterize how “tightly knit” a network may be. While social network theory permits us to understand who is related to whom, and what the nature of that relation is within a tight or loose structure, the foundational concepts originate from social science (sociology) terms and those inherent delimitations.

A comprehensive review of social network literature is beyond the scope of this paper, but one can see that the social roots of the theory remain, today, in political science, management and sociological applications. Van Wijk et al (2003) provide a direct association between organizations and networks in KM field: “Any organization is a social network” (p. 430). He adds that a research focus on ties (structure) and the knowledge (content) shared between those ties (the “what” in networks) could tell us about our knowledge assets, and about the actor status in the firm, adding that the knowledge transfer between actors or network nodes occurs through ties (the net structure) and that structure itself is a strategic asset too. So by studying the ties (structure), much as Granovetter studied ties (studying communication and frequency of contact to define tie strength), we in the KM field may be able to better define knowledge content that is flowing “in the pipe”. By studying also the type of pattern of the ties (the net structure or the “pipeline”), we can identify opportunities for (knowledge) exchange. This is an important point, because by considering network relations and structure, KM scholars may gain abilities to understand the potential for knowledge flows through specific systems,. At the moment, KM and business school literature seems fixated on newer versions of the foundational social network theory, applied generally to business cases (Cross, Parker & Sasson 2003). As I mention, social network theory and policy network theory fell by the wayside a decade ago – has it re-emerged because we can generate fantastically complex computer models quickly that interest folk, or because the theory has advanced? Advanced network and game theorists warn that complex empirical methodologies may never allow us to describe and discriminate such complexity (Sharpf, 1997).

As mentioned, our study of knowledge networks will be limited by our own ontological stance, so our research methods need to fit that which we study. Scholars have proven that ontology is an important causal factor in the process of understanding knowledge management when we use social network concepts particularly when we consider networks as organizations where knowledge management occurs (Scarbrough & Swan, 2003; Kowch, 2003; Ghoshal and Bartlett, 1999). How we interpret (social network) findings using social network theory depends in part on our ontological or world view within the discipline of sociology (Alvesson & Deetz, 1996; Burrell & Morgan, 1979). In a post functionalist world, both practitioners and academics need to approach a problem of understanding. As such the maps we create to describe these networks (social networks) could be predetermined so that we end up studying our own idea of representations of knowledge flows and management in any context. This prevents one of the most powerful use of network theory – to describe complex linkages of people making things happen across ‘expected’ structural or political boundaries, for example. Since social network analyses and study methodologies are predicated on distinct sociological theoretical parameters, it follows that as we integrate network theory with KM theory to model knowledge networks, we should know and consider the interpretation stance we take for any derivative application of the concepts – and derivations are possible if we consider this philosophical dimension (Kowch, 2003). For example, if we take a subjective stance to interpret a social network (organization), we work from the interpretive sociology domain; using phenomenology, hermeneutics, or phenomenological organization theory to design and interpret our results (Burrell & Morgan, 1979). Corresponding research methodologies include ethnomethdoloogy and phenomenological symbolic interactionism, or action research, for example. By contrast, if we take an objective social organization stance to interpret social organization, we interpret within the functionalist sociology domain using social action, social systems, structural functionalist and objectivist theory. Corresponding research methodologies include quantitative empirical study, focusing on inputs, outputs, structures, contingency theory and power models (Alvesson & Deetz, 1996).

An earlier focus on network contacts or nodes (who) and structures remains central to the application of social network concepts within knowledge management, but new interpretive tools allow us to consider collections of networks as well. It has already been discussed that KM scholars who write about social networks interpret their findings from objectivist, political, economic, information systems and organization and management stances (Davenport & Prusak, 2000; Van Wijk et al, 2003; Alvi & Tiwana 2003; Scarbrough & Swan, 2003; Ghoshal & Bartlett 1999; Allen, 1997; Burt, 1995). Beyond this small problem, we can understand collections of networks. There is a macro concept that allows us more latitude to model collections of networks. The macro concept of social capital may allow KM scholars to think about the aggregate outputs (assets) and dynamics (work flows) of networks that get things done. Positive social capital occurs when knowledge or learning networks bond, bridge or link across boundaries, interests or people, and negative social capital occurs when networks do not form, or when they do not function well to flow knowledge (Bourdieu, 1986). As such, social capital is a concept for understanding collections of networks and the outputs they create.

Social Capital: This necessarily brief review of social capital literature begins with cross-discipline definitions of social capital, then narrows to management and political science definitions of the concept. Social capital has been called an “umbrella” or macro concept as it is applied by the management, social science and political science domains (Adler & Kwon, 2002). In an excellent review of source literature, Adler and Kwon (2002) conclude that social capital falls squarely within the broad heterogeneous family of resources commonly called “capital”, describing the emerging plethora of metaphorical uses of the term as non-harmful, but confusing. I suggest that this loose use of terms is harmful to our KM academic understanding about how to integrate social capital with other fields, a process that will be a necessary action for KM scholars of the future. Adler and Kwon define social capital as a concept that defined by three categories of relations: (1) market relations, (2) hierarchical relations and (3) social relations (2002). Their definition is based upon the economic principles of product exchange, authority exchange, and gift exchange for each category, respectively (Adler & Kwon, 2002).

Adler and Kwon’s relation categories define social capital as type of capital that existing “either internally or externally to organizations” (p. 20). This internal/external organization relationship definition is another example of a structural functionalist stance that social capital depends on networks found only within organizations. That academic stance limits our ability to define network relations because networks are defined as existing only within a closed system view of institutions with a bounded rationality representation of knowledge era institutions (Alvesson & Deetz, 1996). The definition is slightly ironic. Social network models were created was that they could model open or closed organization structures in a knowledge economy, to provide scholars with a capacity for designing and understanding the proliferation of intra-organizational relationships (Drucker, 1996; Bourdieu,1986; Barabasi, 2003). This said, it is important to note that most KM authors do see the organization as bounded (Alvi & Tiwana, 2003; Kogut & Zander, 2003). From a management perspective, that idea of social capital is useful and common to any objective study of organizations and management, but we must use it by understanding the sociology roots of the theory (Snowden, 2002; Atkinson & Coleman, 1996). Nahapiet and Ghoshal (1998) presented social capital in terms of organizational advantage in 1998, albeit (again) with the same pluralist ontological stance. Yet, an aggregate descriptor of network knowledge transactions, as an asset, is attractive to KM scholars like this one and social capital may be a good framework for doing so.

Adopting a political science perspective, the Canadian Government’s Policy Research Initiative (PRI) presents two definitions of social capital (Judge, 2004). Their first definition is that SC is a function, where that function consists of those elements of social structure that facilitate the actions of actors within the structure (focusing on network enabling variables like trust, expectations, norms, and sanctions – features that link knowledge and networks more precisely. This again is a functional (objectivist) conceptualization, (p. 9) based on the earlier work of Coleman (1988) and more recently Putnam (2001). This social capital definition is also focused on the contacts or the “who” nodes in social networks. The second definition of SC is that social capital is really just a network: “SC is the aggregate of the actual or potential resources which are linked to the possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition or, in other words, to membership of a group” (Bourdieu, 1986, p. 242). Bourdieu defines social capital as social relations that can be determined by studying content (what) or by studying membership (who) in networks, and also by focusing on the process or dynamics of the network – studying the (how) of network activities to characterize social capital – quite a different, more subjective approach to understanding SC. We will see later how one study of network processes may help us understand this potential for KM network theory development,

In the social sciences, there is a complimentary, widely accepted definition of the three forms of network based social capital that describe the capacity of networks to organize knowledge in terms of what the links do, not what they carry as content. These three forms are:

1.      Bonding SC: relations between homogeneous or similar groups/actors/entities (i.e.: engineering design groups in a multinational engineering firm

2.      Bridging SC: relations between diverse social cleavages (i.e.: environmental groups and engineering design teams)

3.      Linking SC: relations between homogeneous groups, diverse social cleavages and different social or power strata (i.e.: relations between the multinational firm and the federal government or with competitors)

These forms of social capital have been adopted by the OECD and the PRI as well as the UK government and the World Bank as they create policy for the knowledge era (Judge after Woolcock, 2004;, Putnam, 2001). A survey of business, political science and sociology literature implies to us that social capital can also be a (broad) descriptor of (corporate or institutional) relations when applied at a lower level of granularity than (social or policy) network description to provide, an aggregate study of the same systems of relations. If either strong bonds, links or bridges are found in this system, we can determine that these processes are positive social capital assets – weak assets would be determined if poor bonds, links or bridges are found across the relational networks.

A recent KM paper brings this into focus. By demonstrating a sound argument about how networks facilitate learning, knowledge creation and knowledge integration in the KM field, Van Wijk et al. propose that a social network is also a form of knowledge, and that networks in a firm create the conditions for knowledge transfer creation, transfer and integration (2003). We have already reviewed literature claiming that organizations are social networks. Van Wijk et al (2003) equates the structural embededness of an actor (node) within a knowledge network to Nahapiet and Ghoshal’s (1998) structural social capital, and Van Wijk also equates the relational embededness of a network actor to Nahapiet and Ghoshal’s relational social capital. So Van Wijk offers us a knowledge network concept, suggesting that (strong) structural and relational parameters (high social capital) are indicators of a high capacity knowledge transfer (network) organization. This is important when we study organizations that emerge to do work via non workflow or non-hierarchical patterns. Although Van Wijk goes on to model network organization as external and internally bounded relations (once again limiting us to an isolated organization model), he is saying that high (capacity) social capital networks offer the potential for higher knowledge transference (2003). This concept will become valuable in this author’s argument to augment KM and knowledge transfer analytical tool sets whereby a determination of the capacity of a network to organize its knowledge (and to get work done) is possible, even if the human network spans institutional boundaries. Political science offers just the model (from policy network and policy community literature) to describe network capacity and autonomy. These are important characteristics that describe well how relational networks organize themselves. In recent research, these ideas have been converted and tested for use at the institutional or meso level (Kowch, 2003).

2.2.2. Policy Networks and Policy Communities

This section will present modifications to social network theory made by policy network theorists in the mid 1990s. Originally, policy network analysis developed as a response to the inability of public choice (competitive) hierarchical policy models to describe late 20th century collaborative government/sector organizations, and to limitations in social network theory, so as to better explain how people organize their interests as they collect to react and to get policy (making) work done (Lane, 2000; Wilks & Wright, 1987). Policy is defined in this context as a reaction to a challenge or issue, not as a set of dry documents or rules (Pal, 1997). Once again, because older linear, rational choice institution-based frameworks could not describe the reality of government / industry relations in an increasingly partnered government and institutional world, a new frame of analysis was needed (Lane, 2000). The early research in this area was on complex relational systems: governments, international banks, satellite air rights, chemical companies, and market regulatory boards. Rather than conceptualizing government as a benevolent distributor of resources to competitors, network policy analysts accept that (disaggregated) government/industry or sectoral partnerships/relations today are the norm (Lane, 1999; Pal, 1997). On could easily argue that multinational and small business today also exists in an increasingly disaggregated, partnered form. The diminished utility of traditional public choice, public policy or public administration models to describe how such interwoven, codependent governments, industries and sectors organize their interests was the main reason that policy network research emerged (Atkinson & Coleman, 1996). Power and structure models could not describe such complexity. Policy issues or problems are no different than strategic or operational problems in a company, for policy is defined by the policy network framework theorists as a process – a reaction (or inaction) to an issue by a group of people (Coleman & Skogstad, 1990). These actors coalesce from a larger (interested, but not active) constellation of individuals (called a policy community) to act upon the problem (to solve it). Because networks are organizations, and since organizations are knowledge (Van Wijk et al, 2003), this author suggest that the process of interest organization is not different from knowledge management (knowledge transfer and flow), and that policy network conceptual frameworks are very similar to knowledge management networks. In other words, knowledge management theorists may use policy network capacity models to define (or design) high capacity knowledge networks.

In this light, we see a collection of actors as people who are reacting to a problem or challenge (this could be a market challenge or any challenge) as they organize what they think is important (about the problem), to go about the process of developing tangible, tactical or strategic responses to that problem (solutions). Institutional actors do not identify, rank and solve problems in isolation, neither do they work in necessarily hierarchical or bureaucratic structures (Kowch, 2003). They can coalesce or come together from a constellation (community) of actors who have similar interests – in effect, forming a knowledgeable group of actors who create a network to get work done (Alvesson & Deetz, 1996; Garcea, 1997). This is powerful knowledge for network designers and corporate leaders in a knowledge age.

For example, if water rights become an issue (or problem) for a collection of farmers, industries and governments in a region, you can safely bet that a collection of interested individuals will form (with management knowledge or perhaps without it) to generate a response or to push for solutions. This is an example of the identification and ranking of a key policy issue (problem) shared across institutions and organized by people within a network or pattern of relations (that is not bound to the organization). These people may (or may not) successfully create a solution to the problem by working with actors across departments in a company, from various farms, and from across the government agency responsible for the sector, depending on their motivations and common interest, and on their collective capacity to organize (manage) their interests in order to act (to solve the problem). This is complex activity to model, and policy network analysis allows us to render these types of emergent, pressing and sometimes quick-forming pan-institutional processes more simply (Lane, 1999).

Business exists in the same situation today. For example, say two different compensation schedules need to be negotiated because of a new partnership formed by merger. Some people in the new organization will be motivated to achieve a certain solution over other possible solutions, to work together to define, rank and organize their key interests or issues, and to exchange knowledge, as a collection of people across the new entity work to create a solution or response. This can occur no matter how far apart the various divisions or regions are geographically if the issue is important enough to some people (Coleman & Skogstad, 1990). Because labor laws and government regulations may differ in different states, government may be involved in the solution as well. Knowing (or designing) the capacity of this network to find and organize its interests is important for all concerned, so that the process serves some end or solution (work). Such knowledge is extremely valuable to the architects of new (relational) organizations where positive social capital is the desire. Al Qaeda is an example of a powerful network created to generate negative capital, as are the common gossip and corporate sabotage networks. Leadership of good relational networks demands knowledge of these models. Wise managers should be able to craft such a network for success. As a minimum, leaders need to know how to characterize a high capacity relational network.

This paper demonstrates next the findings from a study of three universities in two states and the processes they exhibit as they organized interests to set the educational technology policy directions (and budgets). These institutions and governments collectively invest 10s of millions of dollars annually, affecting hundreds of thousands of students and many faculties. Policy network study is a study about; what issues draw people to an action network, why they were drawn to the issue, and about how they organize their interests in patterns or structures (called networks) to make things happen. It is a study of the how of networks, not only of the whom and the what of networks. In policy network study, individuals or actors are analyzed at the micro level, and the network (pattern) is analyzed at the meso or neo-institutional level. Because interest (knowledge) organization processes are studied and interpreted using an extension of mostly the functionalist policy network canon, a post hoc analysis can be done to interpret the nature of the interest organization process itself – and that information can inform network design for managers engaged in creating high capacity or high social capital generating networks in similar situations at the institutional level of analysis or at other levels (Kowch, 2003).

From an ontological perspective, the policy network canon still is offered mostly from functionalist sociology and organization theory stances. The concept of neo-institutionalism is helpful in breaking the bounded rationality proposed by internal/external institutional analysis (Lane, 2000). The neo institutional construct is particularly useful when designing or interpreting partnered organization processes (networks) that do not function entirely bureaucratically, hierarchically or as closed systems (Kowch, 2003). Neo-institutionalism is the condition where, in a disaggregated state (where government and industry share in responding to issues), institutions have considerable autonomy to organize interests and to create strategies for problem solving (Mawhinney, 1996). That autonomy is of course a function of the capacity of the institution to exchange ideas, and upon the pattern of relations by they choose to exchange the flow of ideas (Atkinson & Coleman, 1996). The basic unit of analysis at this institutional (meso) level is the pattern of relations between individuals who depend, on each other to exchange and to generate new knowledge while they organize their interests or problems (Howlett & Ramesh, 1995). Neo institutional interpretive frameworks allow us to characterize both internal and external (pan institutional and community invested) representations as one network (Mawhinney, 1996), making social capital accounting possible.

According to Coleman & Skogstad (1990), high capacity issue organizing policy network actors, collectively possess the following characteristics:

      a clear concept of role in the process

      a supporting value system

      a unique, professional ethos

      an ability to generate information to answer unanswerable questions

      an ability to maintain cohesion

      an ability to organize and manage complex tasks, leading to a work output (result)

      an ability to rise above the (near term) self interest of the group (network).

In addition, Garcea (1997) notes that high capacity actors have three characteristics that affect capacity in networks:


      institutional contexts (programmatic or political, & managerial and financial management)


Subjective or objective ideologies or stances to the organization task are important determinants of response network capacity (Kowch, 2003). Linking management models describing the capacity of networks to get things done (and to handle change), Kowch adds the work of Ibarra (1993) to describe the dynamic potential of networks to get issue organization (knowledge) work done. Ibarra’s model was based on previous work in social networks (Krackhardt, 1992; Granovetter, 1973) and in institutional level network research, Kowch used these models to add analytical validity at the meso level (network) analysis of policy networks and to provide a second description of loose or tight network ties. Both methods yielded the same descriptions for network change capacity and innovation capacity. So these are the criteria by which the process of organization, can be characterized. The type of organizing the network does (i.e.: pluralist, corporatist, concertist, statist) can also be identified by doing an autonomy analysis (Lindquist, 1996). In this paper, only the network capacity determination will be demonstrated for parsimony reasons. It is clear that high capacity networks also contribute to high social capital generation networks. When we link social and policy network theory to management theory to describe knowledge flow (network transactions and processes), we have the foundations for knowledge network theory – a model for knowledge flows in complex relational structures.

Having confessed to drawing the reader through a detailed collection of (new) conceptual tools from knowledge management, social network, social capital and policy network schools of study , this paper offers a brief explanation of the research methodology that was used to generate network (doctoral dissertation) study results from three case studies. From these three policy network case studies, performed at the micro (actor), meso (network) and macro (environment) levels, the author finally presents findings and a detailed example of high capacity policy issue (knowledge) organization networks at the neo-institutional level. These cases demonstrate how combined mapping, organization and network theory help us to understand more about who, what and how knowledge is organized today.

3. Research Methodology – Network Cases

The subject qualitative multiple case study was designed to apply policy network descriptive theory at the institutional level. By using referential sampling techniques, data was gathered to allow actors across three universities and two governments to self-identify thereby imposing no particular network organization structure. To begin network description, the researcher offered a common problem or issue at hand (managing educational technology across the university today) by asking each actor to identify and rank three influential other (leaders) in the problem solution (knowledge organization) process. In business, a problem like dropping sales could serve as a similar issue. No actor was told who nominated them, so no actors knew who else was in the emergent network. Interestingly, actors were surprised, post study, to find that the influence network was nothing like the bureaucratic structure of the university institution in high capacity network cases. In this way, the study could track influence networks (leadership networks) where the actors all shared a common issue. The researcher tracked how actors defined or understood the problem, finding that even an incomplete understanding of the issue did not preclude interest organization and actions. Capital and operating monies, amounting to tens of millions of dollars were being spent by these networks each year on the issue. The emergent network data was interpreted using case logic (Stake, 1995; Merriam, 1998). Because of the possible complexity and size of a network that spanned over 220,000 students and 4000 faculty and staff (and government agencies), a mixed method interpretive study design was used to avoid the reductionist requirements of a forbiddingly technical (qualitative) project, which is a feature limiting network studies today (Scharpf, 1997).

Primary (semi structured) interview data was collected by using a snowball sampling method, with that information triangulated by comparing interview data with secondary institutional policy documents and data, committee documents, and work samples offered as examples of work done in the policy making (problem solution) process. A data audit was also performed. All data from interviews was transcribed verbatim, verified by participants, and subsequently coded into a relational database using the policy network and autonomy criteria outlined by Coleman & Skogstad (1990). Subjective/objective ontologies for each actor on the subjects of organization and technology were also coded. For primary analysis using the structural models of policy networking (Atkinson & Coleman, 1996), categorical aggregation methods were used to extract themes from the coded data, for micro (actor), meso (network) and macro (environmental) analysis and representation (Miles & Huberman, 1994). Actors identified the top three influential others in the policy problem response... Actors also characterized their relations and transactions with others. Open-ended interview questions allowed the definition of sociological, philosophical and management ideologies held by each. Then, the relational networks showing the first referent were mapped so interviewing continued until the networks “closed” when no new names were adding to the list. Node to node or “tie” strength was determined by reciprocal actor nomination and cross-referenced descriptions of the tie (by each actor). Degrees of net centrality and density were calculated for each network to provide structural descriptions (Krackhardt, 1992). By asking actors questions about leadership, philosophy and organization (how people manage ideas and knowledge), it was possible to do a post hoc analysis to interpret organizational types as well using post structural organization theory to upgrade the descriptive capability of policy network models. Also, the network capacity and the network typology of each network were determined according to political/organizational typologies found in policy theory (Lindquist, 1996; Coleman & Skogstad, 1990). The interpretive framework of Ibarra (1993) was then applied as a method triangulation to validate each network’s work flow or non work flow status and change/action potential, a process that also validated descriptions of the “tightness” or “looseness” of ties found by classic reciprocal nomination and tie strength analysis. A post hoc analysis of the networks was performed using categorical aggregation of emergent themes, so a micro, macro, and meso level post hoc interpretation of the key philosophical, political, societal and organizational elements in each network was done to provide an analysis of network organization and leadership characteristics (including definition of bureaucratic vs. post bureaucratic interest organization (knowledge management) processes within the networks). The resulting report presented the network patterns, capacities and autonomies of each network, presenting network organization types with a concomitant identification of overarching ontologies held by actors toward organization (and technology), and a clear description of the rationale for each member’s commitment to knowledge exchange.

Next, for parsimony reasons, is a summary of the evidence from research about network capacities found from three case studies: (1) a low capacity case, (2) a no-capacity case - a case where (the largest) network turned out not to be a network at all, and (3) a High Capacity case. A detailed summary of the high capacity case capacity is presented as an example of the application of this combination of approaches and methods to determine knowledge network characteristics. This study shows, for the first time, why people come together across institutions (even with far from perfect knowledge of the issue), what they know, and it shows how well and why these leaders were able to engage themselves to get work done. Such a methodology should have potential for future knowledge network study in complex situations.

4. A Low Capacity Network Case Example– Too Many Interests, Some Codependence In A Workflow Network

The low capacity network case was a closed system of faculty and administrators who coalesced to respond to the challenge of organizing technology in university education. Two issues or raison d’etres pulled people together in this network, mapping as cleavages in the network diagram (Figure 1). One low capacity cleavage had weak ties to the other interest group because of their common but self-interest in distance education. The higher capacity cleavage (a larger group) came together because they believed the institution required a progressive image. Both groups were unsure of their role in the network (as policy makers), while all actors exhibited a supporting ethos to serve students and the institution. The two interests in the influence network decreased the cohesion and the organization capacity of the network to get policy done, or to prioritize what knowledge mattered most (actors readily admitted this problem but could not describe it the way the relation map does).



























Figure 1: Low Capacity Network Case


The “distance cleavage”, as a subsystem, had no connection to the government (Sri3) but through the other cleavage. All actors in the case were found to have a strongly bureaucratic or objectivist organization ontology, and most actors preferred to submit decisions to their respective committees (85% of the members sat on each other’s committees). Most actors knew that the super ordinate committees they chose to send policy creation (work) “up” to had no funds, and likely would not pass the recommendations. As such, this is a hierarchical work flow network that is tightly knit, impeding innovation and flexibility to respond to challenges (Ibarra, 1993). This network was classified as a pressure pluralist organization, where both actors and cleavages created a low capacity issue (knowledge) organizing network without organizational autonomy from its loosely connected primary stakeholder. Such a condition hobbles institution level leadership and innovation.

5. A No-Capacity Case Example – No Interest, No Codependence (No network)

The case with the largest number of people nominated as influential in the policy making (solution) processes turned out to be a collection of influential individuals with few reciprocal (influence) relationships (in Figures 1 and 2, double headed arrows indicate reciprocal nominations).

This network identified no less than seven issues that drew them to work on the educational technology issue (to seek solutions and make policy), and the findings are that the main interest (held by 6 of 15 actors) was to create education technology policy to improve the viability of the institution.

On all capacity criteria but the ethos and value criteria, actors scored low in this network. Applying the neo institutional test as in each case, it was discovered that there may not have been a recognizable pan-institutional issue (educational technology issue) in this case, as this university was in the very early stages of adoption (Rogers, 2005). In other words, these actors had such a varied understanding of the issue that they could not identify why they were responding to it, and subsequently had problems identifying who (else) mattered in this knowledge management process. Indeed, the network itself was almost completely horizontal, with a chain of one-way nominations identified by most actors. All but one actor held a predominant objectivist or hierarchical (committee decision making based) organization world view, and most exhibited determinist philosophies about both organizing and technology. As such, many documents collected from the various sub committees represented by these actors triangulated findings that these people spent a tremendous number hours in different groups managing the issue, but that as an influence network they could provide no policy or planning outputs of consequence to the institution. They also had and no connection to the government or to other institutions. Note that there were many ties, and much communication, so a network could (incorrectly) have been analyzed using (closed system) conventional relational social network theory.















Figure 2: No Capacity Network Case

(Circles and Squares Identify different interests)

6. A High Capacity Network Case – One Interest, High Codependence, Loose Ties In A Non-Workflow Network

Figure 3 depicts the structure of the “Calliope” University case, which among three cases showed the highest capacity to organize interests. In this macro environment, the government had strategic plans for the universities, and required the institutions to generate plans that aligned with government plans about education and technology – and the government used large sum policy instruments (grants) targeted micro level (faculty and groups) in the university system, so the policy (macro) environment was far more organized than in the other two cases, where no similar government plans, policy or alignments with the institutions was evident.

Actor Motivation to coalesce to do this work: Overall, the reason for people coming to this to work out the problem responses was found to be a desire to increased market share for the institution, and every member indicated this one issue or interest driving their (network) organization. Though holding a predominantly determinist view of education technology, these people came together to set policy (including very large capital and operational budget sums) because of their common idea that technology will give a market edge to the university. They came up with this understanding as a group, but held the ideal individually as well, and actors came to this network from across many faculties, government agencies and administrative departments. Common knowledge and a desire to come together to respond to a powerful issue was found both to create and to hold this network together - creating links and bridges between, across and “around” traditional organizational flow chart “boxes”. There are powerful applications for this kind of analysis in companies today where collective motivation design is critical.

Network Composition: These 9 actors include service group experts, physical plant people, executives from the academic and administration chambers, and people from the professorate as well as from related (higher education) government. The core or non-workflow (non hierarchical relation) core of this group emerged primarily from an expert service group within the university.




















Figure 3: Network: High Capacity Network Case

Network Capacity: In this network, people from across the institution share one interest with the government and they have the same joint concern, so both the government and the institutional network demonstrate a high degree of autonomy and capacity as they organize their interests and knowledge (they are both doing what they want to do, and they are codependent of each other to achieve their one main goal – to achieve market share gains for the institution). They collaboratively define and solve problems independently (they cooperate in that process as a network, without hierarchical structure). Of great interest is that no actor in this network (or any case) believed that they were providing management or leadership, yet these networks describe all the leaders directing educational technology funding in huge systems.


















Figure 4: Network Type: High Capacity Network Case


The following is a description of (9) policy (knowledge?) network capacity characterisitics:

1.      A clear concept of role in the process: (High). 8 of 9 actors knew what they were there to do, why they stayed in, and what they contributed in terms of expertise and knowledge. Each brought specific language, technical or leadership skills to the network, and they all knew who had what kind of expert knowledge (IT, pedagogy, market data, and plant and property information) so they could find new actors or leave if a project opportunity so required. The inner circle in the group formed a non-workflow type network, where they came together to get projects done to increase market share (not because it was their job duty per se). The second reason why this network worked on the problem was to attract funds. The membership in this network changed frequently, depending on the nature, skills and demands of the particular project. This network, over 3 years, attracted $28 million public and private funds by sharing expertise and finding new information in order to organize issues and interests (knowledge) to generate technology supported (market share improving) projects. The same knowledge network capacities are critical to the successes of private businesses and corporations today.

2.      A supporting value system: (High). 9 of 9 actors declared a teleological approach to selecting and setting policy to get projects done, to getting the projects funded, and to getting projects implemented for the benefit of clients (students) - and for the benefit of the institution.

3.      A unique, professional ethos: (High): All were members of various faculty and CUPE union groups, yet all but 2 actors worked on a "respect for persons" basis when making decisions within the group.

4.      Can generate information to answer unanswerable questions: (Relatively High): Because this network changes its members at the core (the non hierarchical part of the network), expertise and leadership changed as the group saw new project opportunities to improve market share. The outer ring (faculty actors) organized their interests hierarchically, using committees to channel their ideas, and as such, took much more time to generate information – but they flowed information to powerful faculty groups, resulting in a higher autonomy for the network to get things done. The central person, Ai2, was attributed by others to demonstrate, consistently, a respected and unique skill set combining leadership, educational technology and multi-disciplinary communication abilities. Ai2 is situated within a high tech expert service group, and is a recognized and trusted service gatekeeper (directional valve) for the network’s flow of information. Interestingly, this central joint IT/educational technology/leadership person position has changed twice since the study completed in 2000. All IT network member leaders have changed over at least once since 2003. Do knowledge networks put too much pressure on managers who lead across institutions but are accountable to one functional department? This is a condition worth more study.

5.      Can maintain cohesion: (High). This group is project driven, with a changing core of loosely connected individuals who work on a project basis primarily, and who organize their interests according to one principle – market share capture – they work in a concertation type network (Figure 4) where the competitive problems between members and government or other organization do not matter (in the latter case, because the network is relatively closed).

6.      Can organize and manage complex tasks, leading to a work output (result): (High): This group gets the consulting expertise they need by contract, taps the academics when they need to, and manages finances and large amounts of targeted funding and capital. The same is true of the government actors in the network. The capacity could be higher though, if the network were not so small and closed (Collins, 2001; Drucker, 1996; Atkinson & Coleman, 1996). So this network, while capable, hence might not be able to handle large macro environmental changes such as a government change or a shift in market types.

7.      Can rise above the (near term) self interest of the group (network): (High): 8/9 members of the group identified, independently, that if the market share (coalescing) issue were to change, they could change with it to get information technology to strengthen the institution’s market share. They made this claim and offered examples from major budget restriction changes n the past. This decision was offered by all actors, irrespective of (meso level) faculty collective agreements and (micro level) ideological or subject specific differences, which were significant given that people came from across a large university and government.

8.      Institutional contexts (programmatic or political, & managerial and financial management capacity): (High). This programmatic group sets agendas on a flexible basis, aligns its interests with institutional and government (partner) goals, and has a high capacity to manage finances to allocate project financing on the issue across the system (Garcea, 1997). By contrast, the lowest capacity network in the study was a political knowledge network.

9.      Organization Ideologies: Figure 4 shows the result that this network could be classified as a (concertation) type organization network (Lindquist, 1996). All actors had a technological determinist stance to educational technology (the issue). The centre ring of actors in the non-workflow core did not have a bureaucratic or functionalist view of organizing, as evidenced by their steady membership churn (reconfiguring the team) as new project ideas came forward from the rest of the institution. The inner core demonstrated a strong fiscal determinism, but did not evaluate the results of their funded projects or revisit them. A sense of “closed” institutional reality also existed in all members, who believe they could change to suit whatever challenge comes up, which also makes the group susceptible to macro policy, economic and institutional partnerships not of their choosing (Atkinson & Coleman, 1996). In business, network measures of high performance would include a valuation of the social capital, return on investment calculation and market studies.

This high capacity network was found to have loosely connected, non workflow (non hierarchical) organization preferences in the core, and this group also demonstrates high levels of bonding (between university staff), low levels of bridging with other institutions, and high levels of linking with its codependent partners (government, in this case). Adapting to institutional and network change is easy for this network (Barabasi, 2003; Collins, 2001; Ibarra, 1993). Using a social capital macro lens to interpret these findings, the researcher can conclude that network could further increase its positive social capital asset generation (for all institutions involved) by becoming a more open network to include its clients and even its competition (students) in the knowledge organization processes. A subsequent post hoc and policy community analysis (Kowch, 2003) demonstrates weak-link, flexible power relations between the network and its government, it’s attentive public and to other institutions mentioned by the network – indicating an overall positive contribution to social capital generating capability because the co dependent partner (government in this case) is an integral part of the knowledge network organization process.

7. Conclusions and Recommendations

This paper demonstrates one proven way of combining social network, policy network management and organization theory to define and interpret knowledge networks in large, complex contexts. Indeed, the author suggests that by including the study of how networks prefer to organize their knowledge within a neo-institutional framework, we can use this information as inputs to designing (high) capacity for knowledge flows by focusing on the parameters affecting those flows. But we must also realize that the earlier functionalist research stance that created these various model parts, must be carefully adapted for use in future knowledge network research. The results could inform new leadership and management tactics for codependent, competitive institutions existing in the knowledge era.

Clearly, more research must be done in large and small systems, with specific business problems or knowledge sets to test, develop, and validate key network capacity (knowledge flow) theory and practice. However, by combining this policy framework with updated organization and KM theory, the research presented here indicates the beginning of a powerful cross disciplinary descriptive methodology for describing and designing higher capacity knowledge networks at the institutional level. The solution may prevent KM from following the fate of policy network research and development, which was missing parts of social and organizational theory that are necessary for the concepts to make sense today, particularly in complex, changing collaborative business environments today.

8.0 References

Adler, P., & Kwon, S. (2002). Social capital: Prospects for a new concept. Academy of Management Review (27), 17. 17-40.

Allen, G. (1997). Managing the flow of technology. Cambridge, MA: MIT Press.

Alvesson, M., & Deetz, S. (1996). Critical theory and postmodernism approaches to organizational studies. In Cynthia Hardy (Ed.), The Handbook of Organization Studies (pp. 191-218). London, UK: Sage.

Alvi, M., & Tiwana, A. (2003). Knowledge Management: The information technology dimension. In Easterby-Smith, M. and Lyles, M. (Eds.) Blackwell Handbook of Organizational Learning and Knowledge Management (pp. 105-121). Oxford, UK: Blackwell Publishing.

Atkinson, M., & Coleman, W. D. (1996). Policy networks, policy communities and the problems of governance. In L. Dobuzinskis, M. Howlett & D. Laycock. (Eds.), Policy studies in Canada: The state of the art (pp. 193-213). Toronto, ON: University of Toronto Press.

Barabasi, A. (2003). Linked: How everything is connected to everything else and what it means. New York, NY: Plume Publishing.

Bourdieu, P. (1986). The forms of capital. In J. Richards (Ed.), Handbook of theory and research for the sociology of education (pp.241-258). New York: Greenwood Press.

Burrell, G., & Morgan, G. (1979). Sociological paradigms and organizational analysis: Elements of sociological and corporate life. London, UK: Henemann.

Burt, R. S. (1995). Structural holes: The social structure of competition. Cambridge, MA: Harvard University Press.

Choo, C. W. & N. Bontis (Eds.) (2002). The Strategic Management of Intellectual Capital and Organizational Knowledge. Oxford, UK: Oxford University Press.

Coleman, J. (1988). Social Capital in the creation of human capital. American Journal of Sociology (94), S95-S120.

Coleman, W., & Skogstad, G. (1990). Policy communities and public policy in Canada. Toronto, ON: Copp Clark.

Collins, J. (2001). Good to great. Why some companies make the leap.. and others don’t. New York, NY: Harper Business.

Cross, L., Parker, A., & Sasson, L. (2003). Networks in the knowledge economy. London: UK. Sage.

Davenport, T., & Prusak, L. (2000). Working knowledge: How organizations manage what they know. Boston, MA: Harvard Business School Press.

DiMaggio, P., & Powell, W. (1991). Introduction. In W. W. Powell and J. DiMaggio (Eds.), The new institutionalism in organizational analysis (pp. 1-38). Chicago, IL: University of Chicago Press.

Drucker, P. (1994). Post capitalist society. London, UK: Harper Business.

Drucker, P. (1996). How generational shifts will transform organizational life. In F. Hesselbein, M. Goldsmith & R. Bechard (Eds.), Organizations of the Future (pp. 71-86). San Francisco, CA: Jossey-Bass.

Easterby-Smith, M., & Lyles, M. (2003). Organizational learning and knowledge management: Agendas for future research. In Easterby-Smith, M. and Lyles, M. (Eds.) Blackwell handbook of organizational learning & knowledge management (pp. 639-666). Oxford, UK: Blackwell Publishing.

Easterby-Smith, M. (1998). Organizational Learning: Diverging communities of practice. Management Learning, (29), 3. 259-72.

Franklin, U. (1993). The real world of technology. Toronto, ON: House of Anansi Press.

Garcea, J. (1997). Saskatchewan’s aborted municipal service act (Bill 33): Pegasus or Trojan horse? Paper presented at the Canadian Political Science Association Annual Meeting, St. John’s, Newfoundland.

Ghoshal, S., & Bartlett, C. (1999). The Individualized Corporation: A fundamentally new approach to management. San Francisco, CA: Harper Business.

Granovetter, M. (1973). The strength of weak ties. American Journal of Sociology, (6). 1360-1380.

Howlett, M., & Ramesh, M. (1995). Studying public policy: policy cycles and policy subsystems. New York, NY: Oxford.

Ibarra, H. (1993). Structural alignments, individual strategies, and managerial action. In Nitin Nohria (Ed.), Networks and organizations: Structure, form and action. (pp. 165-188). Boston, MA: Harvard Business School Press.

Judge, R. (2004). Social Capital: Building a foundation for research and policy development. Horizons (6), 3. 9-12.

Kogut, B., & Zander, U. (2003). Knowledge of the firm and the evolutionary theory of the multinational corporation. Journal of International Business Studies, (34), 6. 516-540.

Kowch, E. (2003). Policy Networks and Policy Communities in Three Western Canada Universities: Neo Institutional Responses to a Pan Institutional Issue. University of Saskatchewan. Doctoral Dissertation.

Krackhardt, D. (1992). The strength of strong ties: The importance of philos in organizations. In Nitin Nohria (Ed.), Networks and organizations (pp. 216-261). Boston, MA: Harvard Business School Press.

Lane, J. E. (1999). New institutional politics: Performance and outcomes. London, UK: Routledge.

Lane, J. E. (2000). New public management. London, UK: Routledge.

Lave, J. and Wenger, E. (1991). Situated Learning. Cambridge, UK: Cambridge University Press.

Lindquist, E. A. (1996). New Agendas for research on policy communities. In L. Dobuzinskis, M. Howlett & D. Laycock (Eds.), Policy networks: Empirical evidence and theoretical considerations (pp. 11-23). Toronto: Sage.

Mawhinney, H. B. (1996). The new focus on institutions and the reinvention of schooling. In R. L. Crowson, W. L. Boyd and H. B. Mawhinney (Eds.), The politics of education and the new institutionalism (pp. 23-42). London, UK: Falmer Press.

Merriam, S. (1998). Qualitative research and case study applications in education (2nd Ed.). San Francisco, CA: Jossey-Bass.

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: A sourcebook of new methods (2nd Ed.). Thousand Oaks, CA: Sage.

Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, (23), 2. 242-266.

Nonaka, I., & Takeuchi, H. (1995). The knowledge-creating company: How Japanese companies create the dynamics of innovation. London, UK: Oxford University Press.

Pal, L. (1997). Beyond policy analysis. Ottawa: ON: ITP Press.

Polanyi, M. (1967). The tacit dimension. Garden City, NY: Anchor Books.

Putnam, R. D. (2001). Bowling alone: The collapse and revival of American community. New York, NY: Simon and Schuster.

Roberts, K. & Grabowski, M. (1996). Organizations, technology and structuring. In S. Clegg, C. Hardy & W. Nord (Eds.), Handbook of organization studies (pp. 409-458). London, UK: Sage.

Robey, D., & Boudreau, M. (1999). Accounting for the contradictory organizational consequences of information technology. Information Systems Research, (10), 2. 167-185.

Rogers, E. (2005). Diffusion of innovations (5th Ed.). New York, NY: Free Press.

Scarbrough, H., & Swan, J. (2003). Discourses of knowledge management. In Easterby-Smith, M. and Lyles, M. (Eds.) The Blackwell Handbook of Organizational Learning and Knowledge Management (pp. 495-512). Oxford, UK: Blackwell Publishing.

Scharpf, F. W. (1997). Games real actors play: Actor centered institutionalism in policy research. Boulder, CO: Westview Press.

Snowden, D. (2002). Complex Acts of Knowing: Paradox and Descriptive Self Awareness. Bulletin of the American Society for Information Science & Technology (28), 4. 23-29.

Stake, R.E. (1995). The art of case study research. Thousand Oaks, CA: Sage.

Thatcher, M. (1998). The development of policy network analysis. Journal of Theoretical Politics, 10, 389-417.

Tsoukas, H., & Mylonopoulos, N. (2004). Knowledge Construction and Creation in Organizations. British Journal of Management, (15), 1. Supplement 1, S1-S8.

Van Wijk, R., Van Den Bosch, F., & Volberda, H. (2003). Knowledge and networks. In Easterby-Smith, M. and Lyles, M. (Eds.) Blackwell Handbook of Organizational Learning and Knowledge Management (pp. 428 – 454). Oxford, UK: Blackwell Publishing.

Vince, R., K. Sutcliffe and Olivera, F. (2002). Organizational Learning: New Directions. British Journal of Management, (13) , 3. Supplement 2, S1.

Wasserman, S., Faust, K., Iacobucci, D., & Granovetter, M. (1994). Social network analysis: Methods and applications. Cambridge, UK: Cambridge University Press.

Wasserman, & Galaskiewicz, J. (Eds.) (1994). Advances in social network analysis: Research in the social and behavioral sciences. New York: Sage.

Wenger, E. (2002). Cultivating communities of practice: A guide to managing knowledge. Watertown, MA: Harvard Business School Press Books.

Wilks, S., & Wright, M. (1987). Conclusion: Comparing government-industry relations: States, sectors, and networks. In S. Wilks & M. Wright (Eds.), Comparative government industry relations (pp. 274-315). Oxford, England: Oxford University Press.

About the Author:

Dr. Eugene Kowch is an assistant Professor of Education Technology and Leadership in the Graduate Division of Education Research at the University of Calgary. His research involves the study of complex technology-supported organizations, with a focus on integrated knowledge management and planning methodologies. Studying the social economy and organizational development, Dr. Kowch is involved in policy and planning at the institutional, regional, state and federal levels in both public and private sectors. He can be reached at, or at