The traditional model of collaborative knowledge is based on the type of knowledge (moving from tangible to intangible, explicit to tacit) and the place of collaboration (moving from the individual to the group to the organization). The assumption is that knowledge within an organization is housed and can be captured for use. However, with the rise in the use of distributed groups in the workplace, there is the recognition that knowledge that is distributed in the form of distributed cognition, may not be accessed in the same way that traditional organizational structures allowed.
For service industries, whose business is knowledge, it is very important that knowledge is accessible at the individual, group and organizational level, especially during the collaborative process. The collaborative writing process has the potential to make the knowledge creation and transfer process more transparent to group members. Therefore, this study looked at the collaborative writing process within a naturally occurring distributed group to determine what factors affected creation of knowledge within work collaboration for team members that may have had no previous interaction or limited interaction in completing their project. In order to answer the research question of what knowledge members of a distributed workplace group identify as being important when creating a group product and what factors influence the choice of what knowledge is important, I answered the subquestions:
1. How do individuals define “knowledge”?
2. What process or processes does a distributed group in the workplace use to create shared meaning and understanding during collaborative writing projects? What factors do they identify as shaping that process or processes?
3. What patterns of work activity are maintained and changed at the individual, group, and organizational level within a distributed group? Who do workers identify with in maintaining or changing work patterns in different contexts?
Individual definitions of knowledge
In answering the question of how individuals define knowledge, I first looked at the definitions the study participants gave for knowledge, know-how, content, and design. During the interviews, participants were asked to define the terms knowledge, know-how, content, and design. Knowledge and know-how definitions were elicited in the first set of interviews. In the second set of interviews, I asked participants to define design as this was a term used differently throughout the first interview when participants discussed topics such as information, knowledge, meaning making, and processes. Participants were also asked to define content as this was a term often used interchangeably with knowledge in the first set of interviews.
When asked to define the term knowledge, the study participants defined the term knowledge differently. There appeared to be three different groups among the participants: those that defined knowledge as something measurable or tangible (“information”, “content”, “things that you know”), those that had difficulty defining knowledge, and those that defined knowledge as something more intangible (“with knowledge, I know how to behave in any particular situation”, “shared meaning”). Surprisingly, those that identified knowledge as something tangible felt that they were within the organization and group’s power structure. They also were part of the traditional training group, which had a greater influence on the project according to the second round of interviews, 3 months later.
On the other hand, those that defined knowledge as something more intangible, were the most disgruntled with the project 3 months later. One participant had difficulty defining knowledge, also felt a disconnect from the group and organization throughout the project. On the other hand, another participant had difficulty initially in defining knowledge. However, eventually the definition he provided was fairly closely aligned with those within the power structure. Interestingly, both participants had difficulty in placing themselves within the organization and they also tended to be much more careful with the answers they gave during their interviews. This might be an indication that their understanding of knowledge would depend on the definition used by those they perceived as having power over their position within the organization, their departments, the group, and their profession.
These differing definitions of knowledge would indicate that knowledge workers need to be aligned with the epistemologies of their organizations to be satisfied. When there are different epistemologies, belief is in what knowledge is, within an organization, for example between departments, than there might be tension within distributed groups or between group members and those outside of the group. This tension was not obvious in the study group when they executed the routine collaborative written project (quarterly report) in which the format and even content was dictated from outside of the organization. Participants often noted that the quarterly report was not important to their work, but was a means to document the work group members had completed. As a result, it was not necessary that personal epistemologies aligned with the organization or the funder in writing the quarterly report. In other words, the quarterly report from the group needed to align with the funding organization’s epistemology. Additionally, the quarterly report was not perceived as being owned by either group members or the group itself. Rather, the organization (in the form of the Project Manager and Project Director) were perceived as owners of the report by the participants in the study.
However, in the creation of the document that required more group interaction and group ownership, differences in epistemologies were evident and became sources of tension. For example, there were tensions between members of the stand-up training, elearning, and management groups over what exactly the Project Map represented. There were differences in interpretation, often based on the different understanding of how end-users/trainees would learn, need to know, and use what they learned. Since the Project Map would result in the actual final product each group member would contribute to the project, participants felt a much stronger sense of ownership to the Project Map. In other words, knowledge workers might be able to accept different epistemologies in their work when they do not perceive the work as their own, but they would try to exert their own epistemologies when they felt the work was perceived as theirs (each individual taking ownership for the work).
So the greater level of ownership that an individual feels towards a project or piece of work, the more likely they will align that work to personal values and knowledge. Relating this to Skitka’s (2003) AIM model of Social Identify Theory (as discussed in Chapter 2), a greater level of ownership may create higher stakes for the individual within a group. A group member would want to be tied closer to the other group members’ norms if the individual member perceives he or she have a higher level of ownership in the work or work processes. Therefore, if his or her work is not accepted by the group, more than his or her work is questioned; his or her social identity and group acceptance is at risk. On the other hand, if the work is perceived as conforming to the norms and values outside of the group, as the quarterly report did, the group member would perceive the ownership of the work as someone else’s. As a result, their social identity within the group is not at risk. Therefore, he or she might be more open to changes in the document, especially if those changes are perceived as coming from those outside of the group and the organization’s power structure.
There was the perception that knowledge can be located, owned, and/ or accessed either by an individual or the group when needed. Externally owned knowledge (e.g. the funding agency or organization) did not always need to align with personal epistemologies for collaboration to take place within the distributed group. However, with internally owned knowledge, the more work was perceived as being owned by an individual, the greater the necessity that the epistemologies were aligned with the organization, group, and group members in order for collaboration to take place.
Unlike knowledge, the definition of content was consistent throughout the group. In most of the definitions, the term “information” was included. In many cases, some description of a tangible product was included. The participants included descriptors such as: “a variety of media”, “source material”, “images”, “what goes in a page”, “curriculum”, and “whatever the written material is.” Additionally, most participants included skills and knowledge or the word know in their definitions. In many cases, the definition also included how content was applied.
Interestingly, however, is that while content was consistently defined, and often used interchangeably with knowledge, knowledge was not consistently defined. In other words, they defined content as a tangible subset of knowledge. While there might be differences as to what content is valuable or needed, the idea of what content is within a group did not seem to differ.
Many of the definitions were similar for know-how. Often participants identified skill and process, “how to use” and “experience.” Those who defined knowledge as something that was possessed also implied that know-how was possessed.
While participants defined know-how differently, they all seemed to share a common understanding that know-how was not easily measured, had to do with a process or skill which helped to create know-how, and was informal learning. It also seemed that many of the definitions included how know-how was applied. Some of these included:
Application of both what to do and knowing how to use it, mainly in an efficient way.”
“I guess I know…I’m going to use as…I know the know-hows of web development, know-hows of graphic design.”
“…, know how get things done even if it’s delegating.”
“I guess that’s knowledge that translates….into an ability to do something.”
“…intuitive gut navigation.”
While content seemed static, know-how seemed more action oriented. Know-how also appears to be an individual concept. When defining know-how, the participants tended to use the first person. No one used a collective pronoun in their definitions.
After the first series of interviews, I noticed that many of the participants used the term design, but each in a slightly different way. As a result, a definition of design was included in the second interview. In fact, the definition of design had very little in common from speaker to speaker. In addition, it seemed the most difficult term for the participants to define, most having long pauses before they answered.
Those from the stand-up training department tended to perceive design as a definitive construct using terms such as “strategies”, “content”, “framework”, and “curriculum.” For those in the elearning department, design was a more situated term to define, grounded in the creative and meaning making process. For example, “Design is the…planful…elegance and pattern…which gives…definition and meaning.” Not only is there some situated aspect to design, but those in the elearning department identified a sense of agency in their definitions.
Such a divergence in the understanding of what design is could lead to differences in understanding during the creation of a collaborative document, especially when there is no structure to the document, such as the Project Map. The first writing project, using a clearly defined structure was, “Well, the quarterly report’s always a by-product of individual contributions.” On the other hand, the second collaborative project studied was a document created by the group to help identify the various aspects of the elearning project. In discussing this document in a group interview, the difficulty in creating an agreed upon product was obvious:
Well, what formally…I think the … for the classroom trainers, they have a document. Module 1 contains X number of elements. Module 2 contains X number of elements. Eh…for me, there’s a sort of isomorphic mapping of those content elements onto a schema that reflects those same strains and tho…and that same order.
P: There’s a certain way to do that in the classroom. So, um… I think that’s…I don’t know if you can…say right now what your product is going to be. It’s going to be some type of elearning product.Those in the stand-up training department seemed to perceive design in what Buchanan (1992) refers to as the categories of design: “Categories have fixed meanings that are accepted within the framework of a theory or a philosophy, and serve as the basis for analyzing what already exist (p. 12).” The elearning department looked at the possibilities of design, however. This is what Buchanan refers to as placement of design. “Placements have boundaries to shape and constrain meaning, but are not rigidly fixed and determinate. The boundary of a placement gives a context or orientation to thinking, but the application to a specific situation can generate a new perception of that situation and, hence, a new possibility to be tested. Therefore, placements are sources of new ideas and possibilities when applied to problems in concrete circumstances (p. 13).”
R: Right. I mean…
P: But what it’s going to look like and how it’s…how it’s going to work is not…really isn’t there yet.
R: It isn’t really there... I mean, we have an idea…
It appears that coming to the collaborative writing process from these two different approaches affects the knowledge creation and collaborative process. In the one instance, categories, the design is part of tangible knowledge: artifacts, clearly defined processes and skills. In the second instance of placement, design requires spatial knowledge, or the ability to link ideas and construct or create knowledge by building new ideas and theories. The categories of design, then, use knowledge that can be identified or represented tangibly (i.e. diagrams, processes, symbols, formats) whereas the placement of design requires environmental, social, and cognitive interaction to create knowledge networks where externally held knowledge can be accessed when needed. During the collaborative process within this distributed group, the different approaches in design resulted in tensions due to different expectations.
Three types of knowledge
In analyzing the data collected through interviews, documents, and online interaction, knowledge could be placed into three categories: tangible representation of knowledge which is represented by policies, forms, formats, curriculum, degrees or credentials, records, and other artifacts at the individual, group, departmental, organizational, and/or professional level; procedural and tacit knowledge, which would include an understanding of work processes and the knowledge created as a result of those processes; and partaged knowledge, which was created through the linking of ideas, social relationships, cognitive interaction, and/or cultural interaction.
These categories vary from the traditional model, namely content (or explicit knowledge), competency (or tacit knowledge), and expertise (which is performance based). The tangible representation of knowledge is close to Kolb’s (1984) apprehensive knowledge. This is knowledge that can be articulated, represented in various forms (such as visuals, documents, presentations, interviews), and stored for future use. Unlike the more traditional content or explicit knowledge, tangible representation of knowledge may include tacit knowledge. For example, educational credentials (e.g. licensing, degrees) tangibly represents certain knowledge that may include implicit and explicit knowledge. These credentials, as we will see later on, can be used as currency within a group. Within a traditional model of knowledge management within a collaborative group, this would be considered a competency or implicit knowledge that would be expected to be applied in any given situation. This is often used in Best Practices, for example, that can then be replicated and used throughout an organization. In comparing credentials to competency using the traditional model, the assumption would be that they are one in the same.
Using this new definition of tangible representation of knowledge though, a credentialed employee or group member may not know how to access the knowledge or apply the knowledge needed to complete work or contribute to the group because of the lack of understanding or experience with the complexities within the environment or the situation. He or she, however, does have some knowledge that may need to be reformulated in order to be relevant for a given situation. One participant, for example, had credentials in healthcare, but the group did not perceive him as having procedural or tacit knowledge for the project. In order for him to use the tangible representation of knowledge, he would need to learn how to apply it (creating know-how for the project).
Reconceptualizing competency to procedural and tacit knowledge, moves knowledge and know-how from the individually held ability to a socially constructed understanding of how things work within a given situation. It expands the term from cognitive individually possessed knowledge about procedures and processes (competency) to an understanding of the situation (requiring analytical ability), the intangible variables that affect the situation, and the interpersonal relationships and meaning negotiation that create social cognition. When participants discussed competencies in the interviews, they were often referring to tangible representation of knowledge rather than the intangible tacit knowledge. Their discussion of procedures and processes, however, often included such factors as work quality expectations, reconciling processes and procedures between group members and departments, and understanding the environment in which their work and the work of their group members were situated.
The term partaged knowledge is derived from the French word partager which means both to share and to divide. Partaged knowledge is knowledge that one would need to be able to access and link to other knowledge (i.e. linking ideas, putting into context). This might be internal such as what happens during an individual’s writing process. Initially there may be many ideas, seemingly without any correlation (divided). Through the writing process, an author must link together those ideas into one cohesive whole (thus the sharing or putting together through interaction of ideas).
The same can happen with group processes in which members come into the group (especially a distributed group) with different expertise, access to resources, cultural influences, and experience/mental models of the work (divided resources and expertise). Through their work processes, their knowledge is partaged through knowledge networks throughout the group and beyond. This knowledge is then accessed when needed and modified or translated for use within a given situation. Partaged knowledge therefore includes the ability to co-create knowledge, divide the knowledge for later use, access the knowledge when needed, and translate or interpret the knowledge for a given situation. Partaged knowledge is the most valuable for knowledge based organizations as it allows for knowledge to be evenly distributed throughout the organization, thus making them less vulnerable should an employee leave. It also allows for others who are not directly exposed to content, work processes, experience, and/or environments to be able to access knowledge outside of an individual’s knowledge base. Knowledge can be part of the network internal to the group, external to the group, within the profession, internal to the organization and external to the organization. However, partaged knowledge is difficult to quantify, control, and capture.
Partaged knowledge is created through creative practices (writing, design, problem solving) rather than through the imposition of formats or processes. While the traditional knowledge model identified expertise as an intangible form of knowledge held at varying degrees by individual group members demonstrated through performance, partaged knowledge recognizes that expertise, tacit knowledge, and content can be held by individuals, the group, the organization, or even stakeholders. Partaged knowledge is the ability to access expertise and tacit knowledge, when needed in the form that will fit the situation. Once accessed, the knowledge will then need to be translated, renegotiated, and appropriately applied to the situation. Since partaged knowledge is knowledge that one would need to be able to access and link to other knowledge, there must be a sense of internalization that allows for the linking and understanding of the knowledge.
Unlike the traditional collaborative knowledge model, the model identified through this study resulted in continuum of tangibility where each of the three types of knowledge (tangible representation of knowledge, processes and tacit knowledge, and pariaged knowledge) are actually on a continuum of tangibility, as shown in Exhibit 1 (Unfortunately, I can't seem to upload the images in this format).
As knowledge becomes less tangible, groups are able to create knowledge through interaction and negotiation. In other words, partaged knowledge is creative or knowledge that is created through the group process. Exhibit 1 also is a visual representation as to how knowledge becomes more spatial and less linear when it is partaged.
Each type of knowledge was manifested, accessed, created, and valued differently at the individual, group, and organizational level. By redefining the traditional model of collaborative knowledge within a distributed group, organizations will have a better understanding of how knowledge is used, created and recreated, transferred, withheld, managed, and lost at all levels (individual, group, departmental, organizational, and professional).
Level of Identity
The other set of factors that affect knowledge within a distributed group has to do with the level of identity to the knowledge. Using social identity theory (McGrath et al., Skitka, 2003; Van Knippenberg, 2000), the closer the knowledge is related to a person’s social identity, the more likely that person will want the group’s norms, values, and knowledge to be aligned to the individual. When faced with information or an event on the group level that contradicts an individual’s personal identity construct (value, knowledge, epistemology, personal schema), that person has three choices: modify his or her own personal identity, modify the group’s beliefs, values, or understanding, or leave the group in order to maintain the individual’s personal identity (Levesque et al., 2001; McGrath et al., 2000; Moreland & Levine, 2001; Skitka, 2003; Whitworth et al., 2000). However, there is a fourth option that the group in this study used: distance ownership of the work, contribution to the work, and/or expertise needed to complete the work. In other words, the product becomes the group’s, department’s, organization’s, or an external entity’s rather than the individual’s. Related to this is the perceived agency an individual may have over their own work. The greater the perceived agency for a task, process, or final product, the greater the level the individual perceived of ownership over their work or work products (i.e. writing, designs, etc…).
In the traditional model, ownership was based on the location of the group work (individual, intragroup, intergroup, organization). This did not allow for influence outside of the organization on knowledge creation and access. Most knowledge management models (Conceicao, Heitor, Gibson, & Shariq,1998; Cook & Brown, 1999; Foss, & Pedersen, 2002; Nonaka, 1994), for example, assumed that knowledge was held by individuals within the organization. Information could then be transferred from individuals to others within the organization, thus creating knowledge at different levels. Occasionally, there was discussion of transferring the knowledge to external stakeholders (Mason & Lefrere, 2003; Yakhlef, 2002). However, the ultimate owners of the knowledge, to be kept or given away, were the individuals where the knowledge was housed (Cook & Brown).
If knowledge ownership, however, is perceived as being something that is created through agency granted at different levels, the management of knowledge requires the granting of agency at different levels. The closer to the individual that agency is granted, therefore, the higher the level of ownership (and the closer to personal identity) the individual will feel for that knowledge. The new model for distributed collaboration, therefore, would use a continuum of the location for agency and ownership of the work.
Exhibit 2: Location of agency and ownership (not able to upload it to my blog)
In looking at the Quarterly Report, for example, the old model would place the writing as an individual product, with high individual agency and ownership. This is because each individual wrote their own section which often was edited by Robert; but much of the original writing was in the words of the individual contributor. The location of the work was somewhere between the individual and group levels. However, participants repeatedly distanced themselves from the quarterly report. In fact, the purpose and format of the quarterly report was perceived as being imposed on the individual and group by the funding agency. Therefore there was very little perceived ownership and agency in writing the quarterly report at the individual and even group level.
This is an important distinction to make as a traditional model would look to capture the individual contributions, interpreting it as expertise at the individual level. However, much of the individual knowledge was withheld or not captured as the individual knowledge the group possessed was not perceived as having value for the organization and funding agency.
As discussed in the group interview:
R: You know one thing the quarterly report doesn’t do is it doesn’t capture…it…it fails to capture a lot of work that is, eh, either a false start or kind of cons…concept building or teaching one another. And that’s…that’s been an incredibly important subtext of this all of this interdependence has been teaching one another about our work. And the quarterly report doesn’t ever…it’s only interested in what you did. Meaning, like what have you got evidence for. Um…what’s yo…what kind of paper can you point to that would, you know, document and, you know, documentation only goes so far.
In addition, the quality of the work, since it was perceived as being owned by the organization and funding agency, did not affect the social identity of the individual, therefore, there was little time and effort put into writing the quarterly report.
Distributed Group Processes
In looking at a distributed group collaborative writing processes, there were four levels in which meaning making took place: the individual, the project group and departmental (intra-group), the organizational (intergroup, including inter-departmental), and exterior to the organization (including the profession). Each level had different discourse communities with established norms, communication structures, valued knowledge, means of making meaning, and culture.
Placing the findings of this research within the organizational learning, knowledge management, and small group literature, the idea of group knowledge needs to be extended from the traditional model. The traditional model assumed that knowledge was ultimately located in the individual. Organizations only needed to capture and manipulate the knowledge through knowledge management, organizational training, and human capital development. However, in studying the collaborative writing practices of this distributed group, the construct of knowledge is more complex. It includes the social, environmental, political, formulaic, functional, locational, and cognitive factors that affect the perception and use of knowledge in knowledge intensive organizations.
Exhibit 3 is a summary of the types of knowledge accessed, used, or created by distributed groups during their work. This knowledge grid identifies twelve actions that happen during a distributed groups’ collaborative work process. These are dependent upon the location of agency and ownership of the work and the type of knowledge accessed, used, or created during collaboration. While the focus of the study was on the collaborative writing process, the summary could be applied to any distributed group processes, products, work tasks, projects, or communication.
Knowledge definitions need to be reconceptualized using the location of agency and ownership, on the one hand, and the type of knowledge based on tangibility/complexity on the other hand. The knowledge grid outline in Exhibit 3 is a starting point to understating the internal and external environment that affects knowledge use and creation by a distributed group. Each of the twelve types of knowledge represent the political, environmental, cognitive, and social factors that result in different types of perceived knowledge within a distributed group. The next section will look at each type in the context of group collaborative work.
Tangible Representation of Knowledge
Individual-Credentials: Credentials are the establishment of tangible representation of knowledge possessed by the individual. They can take the form of reports and other documentation of individual work; degrees, drafts, postings, or notes that contribute to the group process; and transactional knowledge such as degrees, awards, and job titles. Unlike the traditional knowledge model, credentials within a distributed group is a socio-cognitive construct. Credentials represent valued knowledge that can be used or transformed into an identifiable form to be used as currency for the individual (eg.. future jobs), within the group, between groups, and externally (e.g. product or service sales). The value of the knowledge is situated within the power structure and environment in which the individual works. As a result, perceived credentials will need to be identified and negotiated with each work task. In other words, standard credentials for distributed group work could result in misalignment between the knowledge needed to complete a work task and the actual individual knowledge needed for the work. Therefore, credentials are situated within the knowledge needed for a particular task.
Credentials, unlike documentation, deliverables, and certification, are perceived by the individual as being owned by the individual to dispense whenever the individual believes it is to their advantage. As a result, a group member might have hidden credentials that they feel are undervalued or not needed by the group. Credentials also may be tied to the individual’s social identity, so the undervaluing of the individual credentials may result in a group member disengaging from the group, withholding knowledge, or presenting knowledge in a form that is inaccessible to group members (e.g. unfamiliar formats, technical jargon, limited access documents). In the last case, the individual then becomes invaluable to the group as they are the only ones able to translate knowledge into a form that is identifiable and useful for the group.
Intra-Group-Documentation: Documentation is the transcription of group knowledge into formats that can be stored, accessed at a later date, and used as a collective memory by the group. The group documentation might take the form of meeting minutes, web-based postings, drafts of work (with feedback notes), work plans and checklists, and emails. Documentation of the group processes may be used to create shared mental models; communicate assumptions, interpretations, expectations, resources, and understanding of individual situated knowledge; or create a shared vision, cognitive dissonance, or a collaborative product in order to make the knowledge creation process transparent for those within and outside of the distributed group.
The main difference between credentials and group documents is that group documentation has input from multiple members. As the knowledge grid indicates, ownership and agency is on a continuum, from perceived ownership of a document being owned by a group member with input from other group members to group members claiming equal ownership of the document. The type of ownership and level of agency in producing a group product (such as a written report) depends on the perceived power structure, both internally and externally, a group member’s social identity within the group, the perceived value of individual group member input and contribution of knowledge to the group as a whole, and alignment between the individual and group in the vision of the knowledge work.
Interestingly, documents that are aligned with the group vision and accurately represent group vision become valuable because they make the group knowledge creation process more transparent. This can then be used to help improve a distributed group’s image, output, work environment, and standing within the organization.
Intergroup-Deliverables and Organizational Genres: Most organizations have standard formats, forms, and genres which are used to create documents that will be accessible across units, departments, and other working groups. These organizational genres usually translate knowledge from individuals and groups into standard representations of knowledge that can be understood between subgroups within the organization. The standards and structures that genres provide may also be referred to as style, standard forms (e.g. reports, printouts, templates), jargon, and communication protocols. These organizational genres in the form of standard formats then are used in delivering service products (also known as deliverables) to stakeholders external to the organization. In other words, the standard representations of knowledge is the product that is the core of an organization’s work, be it for profit (i.e. selling knowledge) or non-for-profit (i.e. providing the use of knowledge to benefit an external stakeholder).
Genres are used to create knowledge boundaries within the organization. As a result, ownership of the knowledge becomes collective within the group rather than the possession of an individual. While an individual may perceive that knowledge is aligned with the organization’s, they perceive that the organizational knowledge is owned at the organizational level. In other words, an individual must conform to organizational genres or they will need to either change genres at the organizational level (e.g. through approvals or training) or leave the organization. Often, organizational genres have been developed over time without any individual member realizing the change. Factors such as the environment, changes in vision, a dynamic power structure, and fluidity in interaction between groups (teams and departments) will affect organizational genres and expectations. At the same time, the organization itself often remains a stable structure within which the boundaries of the knowledge intensive work are established for individuals and groups. The organizational structures take the form of organizational genres. As a result, the organization establishes knowledge boundaries through organizational genres, which in turn are constantly being redefined as a result of the intergroup interaction within those organizational boundaries. So organizational genres are dynamic, especially when distributed groups are being used.
External-Professional Standards and Certification: Licensing, training using standard professional curriculums, and accredited degree programs are all ways in which professions provide tangible representation of knowledge the profession perceives as necessary to be part of their discourse community and community of practice. These representations of education and training, also referred to as content or subject matter, outline what a professional needs to know to delineate him or her from one profession to another.
Unlike the organization, an individual can self-select to be part of a profession, but those within the profession might not recognize that individual as a member of the profession. The profession, therefore, has ownership of the standards (professional genres) and knowledge which they may use as currency to control membership within the profession. An individual has the agency to align themselves with a profession, but they have little agency in the development of standards and certification unless they part of the professional power structure. As a result, professional genres are not impacted by distributed work the way that organizations are. Professional genres will be more static.
Procedural and Tacit Knowledge
Individual-Know-how: As discussed previously in this chapter, know-how is the application of knowledge, especially tacit knowledge. According to the definitions given by participants, know-how is an individual understanding of the situation and environment to which tacit knowledge is then applied. Because know-how is based on tacit knowledge, the boundaries of understanding are not well defined. As a result, it is important for the individual to interact with the environment, the task, and tacit knowledge to create a greater understanding of the environment and factors that will affect the performance of the individual. In the study, members that were exposed to a wider divergence of experiences were perceived as having a greater amount of know-how. Individuals with diverse experience were able to transcend the knowledge boundaries within their departments, moving between departments, organizations, professions, and external environments.
For an individual to be able to apply and develop know-how, they need to feel a high level of self-efficacy and agency within their environment. They need to be able to identify procedural knowledge and apply it to any given situation. In other words, the person with the ability to develop and use know-how will be able to identify patterns (cognitive, social, structural, and power) within the environment and develop procedures based on tacit knowledge appropriate to the environment. Those with little know-how will need to rely on others at the intra-group, inter-group, or external levels to establish procedures and translate implicit (tacit) knowledge about the environment.
To develop know-how, individuals need multiple experiences in multiple work contexts, time to understand the context of their work (including planning, feedback, and reflection), a safe environment in which the ambiguity of the work allows for trial and error problem solving, and dialog with others both internally and externally. The development of self-efficacy is also an important component to developing know-how because without self-efficacy, an individual will hand over the responsibility of knowledge development and procedures to the group.
Intra-group-Collaboration: Collaborative knowledge includes both cognition and social understanding of the group environment, including the external and internal factors that affect group procedures and tacit knowledge. Collaboration components include creating a shared mental model, developing group norms, perspective taking, and the ability to align multiple processes, procedures, and standards. This description of collaboration aligns closely with small group research and literature (Jarvela & Hakkinen, 2002; Levesques, Wilson, & Wholey, 2001; Lowery, Nunamaker, Curtis, & Lowery, 2005).
However, collaboration within the knowledge grid can be limited to knowledge within the group (intragroup knowledge) that might be difficult to access by non-members, those not perceived as having value to the group, or those who have limited access to the group’s processes. For example, the video group had limited access to the study group’s processes and therefore did little collaboration within the study group. The video group had limited procedural understanding of the study group’s processes and did not appear to share tacit project knowledge with the study group.
Collaboration needs time and communication space to develop group procedural understanding, trust, a shared vision, vision and task alignment with other groups that members are a part of, an understanding of the perspectives of group’s membership, and an understanding of the political and power structures that will affect the group’s work environment.
Intergroup-Knowledge Management: Traditionally, knowledge management was the culling and organization of information collected from individuals within the organization (Foss & Pederson, 2002). However, knowledge management within the knowledge grid situates information, routines, and understanding within the broader vision of the organization. Organizational routines, the context in which knowledge is managed, are “sequenced patterns of behavior and communication by multiple agents” within an entity influenced by power structure and organizational relationships (Cohen & Bacdayan, 1994, p. 555). According to Cohen & Bacdayan, knowledge is stored at the organizational level within the organization’s routines. The organization then will try to harvest knowledge by monitoring routines.
While a piece of information or routine might be relevant within a group or department, its relevance within the organization will change based on perceived importance by the organizational power structure, the relationship between groups and departments (including the level of communication), perceived situated usefulness by those within the organization, and alignment of organizational, departmental, group, and individual goals and vision.
For example, the video department had its own procedures to ensure access within the department of its knowledge resources on creating and producing quality videos (as defined by the department). There was a shared understanding of the video creation processes, based on the shared cognition of the department about video production. However, this cognition was not shared by those outside of the department. By simply collecting information about the video department’s production process, the organization would not be able to capture the tacit knowledge held by department members. In order for the tacit knowledge of the video department to be useful to the organization, the shared mental model of what good video production was would need to be developed within the video department to align with the organization’s mental model.
Therefore, it is not enough for knowledge management to be effective to simply collect and organize information and data. Information and data needs to be contextualized or situated. There also needs to be interaction between the power structure, departments, work groups, and individuals to negotiate and create a shared mental model. The collaborative writing process required at the organizational level, for example, is an opportunity for employees to create a shared mental model. The feedback mechanisms (i.e. editing, authorization, informal feedback) provide members of the organization to interact and create shared understanding (mental models) that manage cumulative knowledge. These all fit into the four modalities of knowledge creation: socialization, externalization, combination, and internalizations (Castelfranchi, 2004).
Another way to create shared knowledge at the organizational level is to provide interdepartmental cross training in order to develop a cumulative understanding. The result of collaborative writing and cross training can be that the knowledge is owned at the organizational level in which one person leaving the organization will not impact the organizational knowledge. On the other hand, individuals will only need to provide the organization with knowledge that the organization perceives as valuable using knowledge creation processes developed by the organization (Foss & Pederson).
External-Community of Practice/Best Practices. Within the knowledge grid, the professions create shared culture and values, goals, processes, understanding, and experience external to any one organization (Barab& Duffy, 2000). The profession accomplishes this through actions and interaction situated within the professional practice. Cook & Brown define practice as “action informed by meaning drawn from a particular group context (1999, p. 387).” In following a profession’s best practices as defined by those within the community of practice as being experts, members of the community create meaning through their actions (or practice).
Unlike an organization, the development of understanding and knowledge within a profession is more organic and less structured. As a result, tacit knowledge is more visceral and more difficult to transpose into explicit or identifiable form. Because the context of work, routines, and knowledge is situated in a broader environment, the knowledge that is known within the profession may be limited to a few universals. These universals or best practices owned by the profession become a way to identify membership within the community. As a result, the building blocks of the profession (best practices) have the potential to be greatly interrelated with a member’s social identity.
For example, the tone and jargon used in the healthcare counseling profession was an indicator of professional membership. When one of the study participants, not a member of the healthcare counseling profession, used a tone and jargon that study participants who were members of the profession did not perceive as being appropriate to the profession, professional members were almost offended. They felt a strong social identity with the healthcare counseling profession, so they would feel any breech of the profession’s shared understanding as unacceptable. It was difficult for them to convey to the member outside of the profession why her choice of tone and jargon was offensive because it was based on tacit professional knowledge that was difficult to identify.
Individual-Expertise: The definition of expertise within a distributed group needs to be expanded based on the idea that experts are able to access partaged knowledge. This means that an individual has the ability to access knowledge outside of their own mind. This external knowledge is part of an individual’s knowledge network. Individuals who access others knowledge networks then create what Boland and Tenkasi (1995) refer to as a knowledge net, from which to cull, filter, and highlight knowledge that can be used for a specific purpose (Goodwin, 1994). Expertise, then, is the creation and weaving of knowledge nets for specific purposes. The creation and weaving of the knowledge net is situated in an individual’s environment, task understanding, meaning-making, perspective, and social and cognitive abilities. Experts need to be able to identify the boundaries of their own knowledge; make connections with others though interaction, meaning-making dialogs, and perspective taking; create processes to access their own and others relevant knowledge in the time frame that it is necessary, and redefine knowledge boundaries that are situated in a specific context.
Expertise is also dependent upon an individual’s ability to recognize and react to fluctuations in the environment (Nonaka, 1994). To do this, an individual would need to continuously identify knowledge gaps (the difference between what is known and what needs to be known), understand relationships within the social environment (including perceived power, power structure, and perspectives), and identify relevant social and knowledge networks.
In order for expertise to be developed, individuals need to be given opportunities to create complex knowledge networks. These opportunities might include membership to a variety of distributed groups, cross-functional training, access to professional social networks and communities, access to a variety of communication tools and channels, and time, opportunity, and space for individuals to interact. Expertise is a continual process as the work environment is dynamic, rather than a static state.
Intra-group-Collaborative Design: Fundamentally, design is always done in the context of a group. Collaborative design gives ownership of the process and product to the group. However, design does more than develop a shared process and product. Collaborative design creates a shared mental model and understanding of the work task, identifies group member strengths, weaknesses, resources, and knowledge networks, and develops social relationships within the group. It also gives access to group member knowledge networks and group members that can translate knowledge from outside of the group so that it has meaning for the group. Design becomes both the categories (schemata, paradigms, values and perspectives) and the placement (possibilities) in the application of group knowledge (Goodwin, 1994, Nonaka, 1994).
A good example of collaborative design was the Project Plan which later became the project map. Since collaborative design is not static, this document provided both a framework which defined knowledge boundaries and a roadmap for knowledge creation within that framework. The framework became the basis for the shared mental model, which changed as the environment and external factors within which the group was located changed. The group accessed resources from the group’s knowledge network, translated knowledge and created new meaning from the knowledge net created by the group. They then developed new group perspectives by continually updating the Project Plan/Project Map. Nonaka (1994) refers to this group process as semantic knowledge. The insight created through the group design process “provides a new point of view for interpreting events that make previously invisible connections or ideas obvious or shed light on unexpected connections (Miyazaki and Ueno as cited by Nonaka, p. 16).”
To maximize partaged knowledge within the group, distributed groups need to have time to design collaboratively on a continual basis. It is important that group members have a mechanism to continually align perspectives and understand group members' knowledge networks as these networks redevelop. Interdepartmental training and cross functional training will also allow group members to update their knowledge networks, creating social and cognitive relationships outside their established networks. Training will also help identify group members that can act as an interpreter between and within groups and networks.
Intergroup-Organizational Culture: Each organization has its own patterns of work and knowledge that ebs and flows through it. These patterns and the continual creation of knowledge needed to maintain the patterns develop a shared system of beliefs, values, meanings, and symbols (artifacts, metaphors, and organizational rituals) that make up organizational culture (Cook & Yanow). Organizational culture not only informs work patterns, but the dynamic nature of organizational work patterns inform and mold organizational culture.
While knowledge management tries to control the work patterns, organizing knowledge to maximize efficiency, organizational culture gives context to work patterns, prioritizing competing demands and aligning knowledge within the organization whenever dissonance or changes create tension between parts of the organization. Organizational culture also gives meaning to work patterns, acculturating members and stackholders of the organization. These members and stackholders, in turn, create, pass on, maintain, and modify the organization’s cultural identity through their collective practices (Cook & Yanow, Nonaka). Knowledge is embedded within the culture. It is difficult to access this organizational knowledge by those unfamiliar with the organization, nor is this organizational knowledge held by individuals. Rather, the meaning of organizational rituals, artifacts, and symbols is created collectively by the use of rituals, creation of artifacts, and use of symbols (Cook and Yanow, Nonaka).
The project map is a good example of how knowledge is situated within the organizational culture and individuals align meaning through work patterns. This document was used a roadmap for the healthcare group’s work. However, individual interpretations of the project map did not align with the organization’s cultural value of the use of traditional stand-up training curriculum. The map was reinterpreted so that the work patterns changed. The project map document itself was not changed, but the work patterns that were based on the organizational culture did change.
External-Professionalism: The main difference between community of practice and professionalism is that while community of practice is a shared understanding (professional mental model) of the profession, professionalism knowledge exists outside of the individual(s) through formation of professional alliances and networks (Nonaka, 1994). Once a person is identified as a member of the profession, they will need to understand where and how to access community resources.
Professionalism is trans-organizational which means that there needs to be interorganizational interaction for professionalism to exist. This interaction can come in the form of interaction and training with stakeholders, professional organizations, and professional institutions (e.g. professional training programs, higher education programs). The interaction creates both a shared mental model for the profession and an understanding of where and how to access resources within the profession so that it is unnecessary for an individual to know all aspects of the profession, but will have access to all professional knowledge when needed. Conferences, professional training (and retraining), and active membership in professional organizations, therefore, will take on greater meaning as a means for an individual to participate in the knowledge networks where the profession’s knowledge is embedded. Without these activities, individual members will be excluded from the ownership of professional knowledge.
Organizations that are interested in capturing the knowledge created and used by distributed groups need to recognize the importance of each type of knowledge: knowledge that can be tangibly represented, tacit knowledge and know-how, and partaged knowledge. Partaged knowledge is often overlooked in traditional models and is especially important in distributed groups. As these groups come together, they create a shared, collective knowledge that is more than one individual can possess. Among the synergy created in distributed groups is the understanding of other members’ knowledge network in which knowledge is held outside of the group. However, by nature, the knowledge of distributed groups is dynamic, as members return to their own environments to create new meaning, understanding, culture, artifacts, and knowledge networks. To maximize knowledge creation and sharing within the workplace, organizations need to optimize their understanding of discourse communities, the power structures (formal and informal), and work patterns, that affect distributed group processes.
In the study, there were four distinctive discourse communities: 1) the profession(s) to which the study participants belonged; 2) the department(s) to which the study participants belonged; 3) the organization and the power structure within which the group operated; and 4) the group itself.
Not surprisingly, the most successful participants within the group were those that could move easily between and work within multiple discourse communities. Group members who were identified as being the ones that could contribute the most to the group and project, were those that had an ability to learn and move within new discourse communities. Other participants, along with supporting personnel external to the project were identified by numerous study participants as appearing to be unable to participate in the discourse community outside of their own profession or department.
Participants with power within the organization or group would be able to contribute to the project, as their discourse community could be made the standard for discussions. Those outside of the power structure, however, became disenfranchised, frustrated, and felt that their voice was not heard (when in fact their voice was not being interpreted correctly). One option to avoid this problem was for a new working discourse community to be developed for group processes and products. However, group members needed time to create the new heuristics for the created discourse community. As a result, the better option was to create distributed groups whose members could move between the two communities, thus becoming invaluable to the project. Because of group members’ ability to interpret messages from one community to the other, work could be completed simultaneously in each community. However, when those that were acting as interpreters either left the project or pulled back into one or the other community, the success of the group working within both communities was a burden on those that could move between both groups. This may have been why one of the group members complained that she felt she was carrying the burden of the project.
This finding about the discourse communities has a number of communication, training, and management implications. First, training should recognize the role of discourse communities in any new intervention. They should be prepared for miscommunication between discourse communities, parallel discussions, and the burden of change on those who do not necessarily have the expertise, but rather can move between discourse communities.
Second, there needs to be different channels of communication available for different discourse communities. These should not be exclusive of one discourse community or another, rather they should be available to open up dialogue between discourse communities. Formal channels might need to be set up to encourage and allow interaction between the different discourse communities.
Finally, there needs to be a way to ensure that group members are all adept at moving between the discourse communities so that the burden of work does not fall to one or a small group of people. This also ensures that when one person leaves the group, the group will not be paralyzed due to lack of interpretation between discourse communities.
Power structures and ownership
In comparing the two documents, the quarterly report and the project map, there was a difference in the perceived level of ownership. The quarterly report was perceived as being owned by the organization and/or the funding agency whereas the project map was perceived as being owned by the group and/or individual members. The amount of effort and the level of knowledge sharing decreased the further ownership was perceived by the individual. Related to this was an individual’s perception of their place within a task’s power structure. The more an individual perceived they had a higher level of agency within a work task , that their work was important to those within the power structure, and/or that they held a vital spot within the power structure, the greater level of ownership they claimed over a work product or process.
Study participants that indicated a high level of efficacy, ownership, and/or position of power over a specific work task or document tended to use more partaged knowledge to accomplish that task. Participants that perceived their knowledge as less valued, an outcome as owned by others, and/or having limited power over a specific work task or document tended to rely on knowledge that could be represented tangibly.
In order to access a deeper level of worker knowledge, therefore, group members need to feel as if their knowledge is valued. Previous research on information sharing in group decision making processes concur that there needs to be an environment that creates trust within a group, allows for disagreement or contrary information and opinion, rewards (intrinsic and extrinsic) for contribution to the group, and opportunity (through management and communication style) to create knowledge through environment, dialog, communication tools, and time to negotiate meaning (Galinsky & Kray, 2004; Larson, Foster-Fishman, & Franz, 1998; Jeffrey, Maes, & Bratton-Jeffrey, 2005; Miller, Jackson, Mueller, & Schersching,1987).
It is important that employees understand that there will be some work tasks and artifacts for which they will not be able to claim ownership. Government documents, for example, are owned by the government in that they must conform to government expectations. It is important in this case that knowledge that can be tangibly represented be used in creating the document. On the other hand, planning memos or documents to which group members can claim ownership can create tacit knowledge, or access partaged knowledge. Partaged knowledge will not be evident in the document, but will be evident through the effective use of group and group member knowledge in the creation of the document. Both types of documents and collaborative writing processes are important for the organization.
Those within the power structure need to hand over ownership at appropriate times in order to maximize knowledge creation within the organization.
One way to hand the ownership over to groups or individuals is through different types of feedback. During this study, managers and group members used different types of feedback which either provided ownership or withheld ownership at different levels of the process. For example, the IT Director gave specific stylistic changes to one group member needed to make for the quarterly report which took ownership away from from the group member and gave it to the IT Director. On the other hand, the IT Director signed off on the Project Map, giving approval for the document (a form of feedback) but maintaining the group’s ownership of the document. In the first case, ownership of the quarterly report was perceived as being owned by the funding organization and the organization itself whereas the second document was perceived as being owned by the group. Study participants, with the exception of the project manager, perceived very little knowledge creation from the quarterly report, but a high level of knowledge came out of the discussions and creation of the second document. The project manager, a member of the power structure (management) was a co-owner/creator of the quarterly report, so it is understandable why he perceived a greater level of knowledge creation from the quarterly report.
The study group, like many distributed groups, had competing goals and priorities within which they were working. As reported in previous research on workplace writing (Cohen & Bacdayan, 1994; Martin et al., 2003; Mason & Lefrere, 2003; Yakhlef, 2002), the collaborative writing process can create organizational memory through the conversion of tacit knowledge into knowledge that can be tangibly represented. The tangibly represented knowledge is the result of filtering tacit knowledge and converting it into a form that can be saved and accessed by the organization. Not all tacit knowledge will be recorded, and the process used to convert tacit knowledge into recorded organizational knowledge may create new tacit knowledge. For example, one participant created new processes in order to capture the explicit knowledge required by the funding organization. The quarterly report itself became a record of organizational memory (knowledge that was represented tangibly). However, the new processes created to capture this information generated new implicit knowledge that the participant then used to identify resources needed by those providing the information (stakeholders of the project).
More complex written documents such as analyses, planning documents, visual or creative products and representations of works (e.g. blueprints, advertisements, project management, PERT or Gantt charts), however, are tied into the complex environment within which a distributed group works. These writing processes of distributed groups fall into Lowery’s (2004) expanded definition of writing outlined in Chapter 2:
The potential scope of [Collaborative Writing] goes beyond the more basic act of joint composition to include the likihood of pre- and posttask activities, team formation, and planning. Furthermore, based on the desired writing task, [Collaborative Writing] includes the possibility of many different control approaches, team roles, and work modes (p. 72-74).
In the case of a distributed team, there may be competing control approaches, team member roles and goals, and work mode pressures throughout the duration of a collaborative writing task. The complex collaborative writing task generates more than a final product: it also helps to create knowledge networks and an understanding of multiple environments that impact the distributed group. The written product within these complex environments often becomes visceral as the competing work patterns require changes in the final written product. Thus, a distributed group’s writing process becomes one of drafts, feedback within the group, changes in work patterns, feedback from individual members’ environments, changes in work patterns, changes in the interpretation of the written draft, and changes (if necessary) to the final draft of the written product to align work pattern expectations within the group and their knowledge networks.
As demonstrated in the differences between the professions, group, and departments, distributed groups need both time and opportunity to interact as a means of developing shared cognition and mental models. This, in turn, will provide group members the opportunity to access/develop tacit and partaged knowledge. In addition, the choice of group members within a distributed group may impact the amount of knowledge created by a group. Members of a group with social capital outside of their departments will be able to use expertise and knowledge that is perceived as having value outside of the group. For example, some of the study participants were able to work with group members outside of their own departments because they had expertise and the ability to communicate with those outside of the group. When one member's social capital outside of the group lost value, she was no longer as effective in her work and eventually left the organization.
Akgun, A., Lynn, G., & Byrne, J. (2003). Organizational learning: A socio-cognitive framework. Human Relations, 56 (7), 839-868.
Barab, S. & Duffy, T. (2000). From practice fields to communities of practice. In Jonassen, D. & Land, S. (eds.) Theoretical foundations of learning environments. pp. 25-55. Mahwah, NJ: Lawrence Erlbaum Associates.
Boland, R. & Tenkasi, R. (1995). Perspective making and perspective taking in communities of knowing. Organization Science, 6 (4), 350-372.
Buchanan, R. (1992). Wicked problems in design thinking. Design Issues, 8(2), 5-21.
Cohen, M, & Bacdayan, P. (1994). Organizational routines are stored as a procedural memory: Evidence froma laboratory study. Organization Science, 5(4), p. 554-568.
Conceicao, P., Heitor, M., Gibson, D., & Shariq, S. (1998). The emerging importance of knowledge for development: Implications for technology policy and innovation. Technological Forecasting and Social Change, 58, 181-202.
Contu, A. & Willmott, H. (2003). Re-embedding situatedness: The importance of power relations in learning theory. Organization Science, 14 (3), 283-296.
Cook, S. & Brown, J. (1999). Bridging epistemologies: The generative dance between organizational knowledge and organizational knowing. Organization Science, 10 (4), 381-400.
Foss, N., & Pedersen, T. (2002). Transferring knowledge in MNCs: The role of sources of subsidiary knowledge and organizational context. Journal of International Management, 8, 49-67.
Engleberg, I. & Wynn, D. (2007). Working in Groups. Boston: Houghton Mifflin Co.
Galanes, G. & Adams, K. (2007). Effective Group Discussion: Theory and Practice. Boston: McGraw Hill.
Galinsky, A. & Kray, L. (2004). From thinking about what might have been to sharing what we know: The effects of counterfactual mind-sets on information sharing in groups, Journal of Experimental Social Psychology 40 (2004) 606–618.
Gersick, Connie (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31 (1), 9-31.
Goodwin, C. (1994). Professional vision. American Anthropologist, 96 (3) p. 606-633.
Jarvela, S., & Hakkinen, P. (2002). Web-based cases in teaching and learning: The quality of discussions and a stage of perspective taking in asynchronous communication. Interactive Learning Environments, 10(1), 1-22.
Jeffrey, A., Maes, J., and Bratton-Jeffrey, M. (2005) Improving team decision-making performance with collaborative modeling. Team Performance Management 11 (1/2), p. 40-50.
Kolb, D. (1984). Experiential learning: Experience as a the source of learning and development. Englewood-Cliffs, NJ: Prentice-Hall, Inc.
Larson Jr., J., Foster-Fishman, P. & Franz, T. (1998). Leadership style and the discussion of shared and unshared information in decision-making groups. Personality & Social Psychology Bulletin , 24 (5), 482.
Levesque, L., Wilson, J., & Wholey, D. (2001). Cognitive divergence and shared mental models in software development project teams. Journal of Organizational Behavior, 22, 135-144.
Lowry, P., Nunamaker Jr., J., Curtis, A., & Lowry, M. (2005). The impact of process structure on novice, internet based, asynchronous-distributed collaborative writing teams. IEEE Transactions on Professional Communication.
Martin, G., Massy, J., & Clarke, T. (2003). When absorptive capacity meets institutions and (e)learners: Adopting, diffusing and exploiting e-learning in organizations. International Journal of Training and Development, 7(4), 228-244.
Mason, J., & Lefrere, P. (2003). Trust, collaboration, e-learning, and organisational transformation. International Journal of Training and Development, 7(4), 259-270.
McGrath, J., Arrow, H., & Berdahl, J. (2000). The study of groups: Past, present, and future. Personality and Social Psychology Review, 4(1), 95-105.
Miller, C., Jackson, P., Mueller, J., & Schersching, C. (1987). Some social psychological effects of group decision rules. Journal of Personality and Social Psychology, 52 92), 325-332.
Moreland, R., & Levine, J. (2001). Socialization in organizations and work groups. In M. E. Turner (Ed.), Groups at work: Theory and research (pp. 69-112). Mahwah, NJ: Lawrence Erbaum Associates.
Nonaka, I. (1994). A dynamic theory of organizational knowledge creation. Organization Science, 5(1), 14-37.
Skitka, L. (2003). Of different minds: An accessible identity model of justice reasoning. Personality and Social Psychology Review, 7(4), 286-297.
Tuckman, Bruce (1965) Developmental sequence in small groups. In Classics for Group Facilitators.
Van Knippenberg, D. (2000). Work motivation and performance: A social identity perspective. Applied Psychology: An International Review, 49(3), 357-371.
Yakhlef, A. (2002). Towards a discursive approach to organisational knowledge formation. Scandinavian Journal of Management, 18, 319-339.