As promised, below is an outline of the methodology I used to study indepth a distributed group as they worked on two collaborative writing projects.
The original data collection was based on the research questions: How do individual members interpret the experience of a collaborative writing process within distrubted workplace groups? What factors influence the interpretation of experience that individuals take with them outside of the group as a result of participating in the collaborative writing process? The purpose in answering these questions was to gain a deeper understanding of what the individual perspective is and the various factors, including group dynamics, context, the role of writing and technology, culture, organizational structure (including agency and power structures), and types of learning, that influence the individual in creating his or her own knowledge, contributing to the group’s shared knowledge, and contributing to organizational knowledge and learning.
Due to the complexity of each of these factors that might influence the individual, it was necessary to have a deeper understanding of the individual, group, and organizational processes and contexts in which collaborative writing takes place. Therefore, qualitative methods, specifically grounded theory and ethnomethodology, were used (Charmaz, 2006; Clancey, 2006; Garfinkel, 1967; Patten, 2002).
After the data was collected, it was apparent that there were many directions in which the data could be analyzed. There were three areas that could be analyzed based on the literature review: the effect of group communication, the impact of organizational structures, and/or the effects and affects of written formats on the collaborative writing process. As a result, as often happens in grounded theory research (Charmaz, 2006), the research questions were rewritten to address the emerging themes and concepts found in the data. I decided to focus on the impact of the collaborative writing process from an organizational learning perspective. The questions still were used to gain a deeper understanding of what the individual perspective is and the various factors, including group dynamics, context, the role of writing and technology, culture, organizational structure (including agency and power structures), and types of learning, that influence the individual in creating his or her own knowledge, contributing to the group’s shared knowledge, and contributing to organizational knowledge and learning. However, the questions were more focused on the methods of common understanding created through individual and group practices located within the organizational social and knowledge structures (Garfinkel, 1967).
The questions were then changed to: What knowledge do members of a distributed workplace group identify as being important when creating a group product? What factors influence the choice of what knowledge is important?
The subquestions included:
• How do individuals define “knowledge”?
• 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?
• 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?
While there are many qualitative methods from which to choose, ethnomethodology allows researchers to look at how the individual and social system (the collaborative writing group, organization, clients, and sponsers in this case) influence each other (Brandt, 2005; Schneider, 2002). Ethnomethodology, with its roots in sociology, looks at the effect that a social system has on individual behavior, which behavior in turn informs the development of the social system either by modifying the structure or reinforcing the social dynamics, assumptions, and power structures (Brandt; Schnieder). Researchers using ethnomethodology (as opposed to ethnography) look at everyday practices as coconstructed social activities based on the perspective of the individual (Clancey, 2006; Garfinkel, 1967). Thus, data is analyzed through the lens of both the individual and the organization through the interaction ritual chains (Hilbert, 1992).
According to Brandt, who has looked at workplace literacy practices using ethnomethodology, this approach not only looks at how an individual’s writing is influenced by the social context in which the writing takes place, but also how the individual then becomes the part of the social context by justifying his or her choices and helping to reinforce organizational writing formats. The context and the individual cannot be separated since the internal processes of the individual helps to create the social context.
Therefore, ethnomethodology will require that I look at the individual as a co-creator of the context for the group and the organization. However, rather than looking at it from an organizational or group perspective (thus separating the individual from the context), ethnomethodology allows me to look at it from the individual’s perspective through the interaction with the social context, however complex that might be (Clancey, 2006). In using this methodology, therefore, the influences on individual and distributed group outcomes in a collaborative writing project in the workplace may be both internal and external to the individual, but their location will not be as important as the impact on the individual’s perception of the collaborative writing process and outcomes.
It is important, therefore, that this study includes:
• a complete description of the perceived contexts by group members at the organizational, individual, and group level;
• individual member epistemologies;
• the perceived social structures and discourse communities in which individuals and the group as a whole work;
• a description of the process the group uses to achieve their task and understanding of each other’s position in order to describe the context of their work;
• individual perceptions of what they should and what they actually bring into the collaborative writing process;
• individual perceptions of the effect members have on the collaborative writing process and their group members, and how they themselves are affected by the collaborative writing process and the other group members;
• perceived learning and knowledge creation due to the collaborative writing process; and
• perceived value of the collaboration at the individual, group, and organizational level.
While ethnomethodology will help inform the type of data that is collected and impact how data is analyzed, grounded theory, especially constructed grounded theory, will guide the research process. Constructed grounded theory uses the identified themes emerging from the data to construct theory. It differs from classic grounded theory in that it interprets the data in developing theory rather than looking for explicit codes initially (Charmaz, 2006).
A major criticism of research on group dynamics and processes in the past is that studies created groups and group tasks in an artificial environment, therefore, minimizing the complexities of group work. More recently, research on groups have studied naturally occurring groups in their own environment, so as to look at the relationship between members in a more authentic environment and capture the dynamics that are the result of organizational structure, shared culture, organizational politics, and shared past experience (Gersick, 1988; Jehn & Mannix, 2001; Jehn et al., 1999; McGrath et al., 2000). As I am trying to understand the external, as well as the internal factors that influence an individual’s experience, it was important that the group I study is a naturally occurring group that would normally collaborate together on a written document.
The first step in recruiting a natural occurring distributed work group was to identify organizations that would allow the study to be conducted, as required by research protocols laid out by the University’s Institutional Review Board. Groups were chosen from organizations dealing in “knowledge” as part of their business. Brandt (2005) identifies this knowledge as the intangible tacit knowledge that adds to the value of a product. Because a large part of services are the intangible product, capturing knowledge and codifying it through writing to make it more tangible is very important to service organizations. There may also be strong external pressures to a group collaborative writing process such as government regulations, organizational quality control to standardize processes, and the demand for knowledge in a tangible format (such as a book, webpage, prototype, or software). As a result, sites were chosen from organizations whose products do not fall into the first three categories of the North American Industry Classification System (NAICS product codes for: 1. Agriculture, Forestry, Fishing and Hunting, 2. Mining, and 3. Manufacturing). Possible service industries included government organizations, economic development, advertising, education, consulting, financial services, transportation and travel, retail, healthcare services, legal services, or software development (United States Census Bureau, 2002).
Organizations were identified through trade organizations (i.e. Capital District Trade Organization), listservs (Communication Faculty listserv), and other professional groups and contacts. I made contact with management within the organization, gave them an overview of the project (based on the application made to the Institutional Review Board) and identified two organizations that worked with distributed work teams.
I then needed to work with an organization in identifying pre-existing distributed groups that would be interested in participating in this study. The group as a whole would need to agree to the study, meaning that if one member did not want to participate, the group would be excluded, rather than replacing that member with someone new that would agree to participate.
In identifying possible groups, potential teams needed to meet the following criteria:
• the group or team should be distributed during some part of the process,
• use technology to support their collaboration,
• and have a core group of between 3 and 7 members.
Groups that are “distributed” do not work within the same physical space on a daily basis. They may be distributed by department, by location (e.g. field office and home office, US based and foreign based) or by speciality (e.g. sales, support, management). The group or team may include face-to-face interaction, but part of the collaboration process should include interaction supported by technology (i.e. telecommunications, internet or e-mail, video conferencing, or shareware). This would help inform the study as to how enforced structure (including the limitations technology might place on interaction) and conflicting contexts (local and group due to distributed team) might affect the collaborative writing process. Finally, research has identified the optimal size of groups at 3-7 members (Moreland & Levine, 2001).
In order to limit the effect that group size might have on the collaborative process, optimal size of the study group would have been 3-5 members. However, in many naturally occurring groups, there are peripheral members that would also have an effect on the group process and at any given time more than 5 members may have been working on the project. It was important, therefore, to include these peripheral members in the study. As a result, the group studied was a core group of 4-5 members working on the collaborative writing project, with the input of 2-3 additional members at any given time during the study, all of which gave their consent for the study.
The first site organization was chosen because of its international distributed work teams. Two groups were identified as potential study participants and I contacted each group member directly with an overview of the project. It soon became clear that because of cultural differences and work pressures, that international members of the group were reluctant to participate. So I changed the site (and organization) of the study. The recruitment of the participants for the second site was modified somewhat. I gave a presentation that was an overview of the study to potential participants. They then gave feedback as to whether they would be willing to participate in the project. Based on this, groups were identified as potential study subjects.
I chose the group that met the criteria for the study. Many of the groups had been working together for a while. The final group chosen had never worked together, was at the beginning of their project, and were distributed in three different locations. In addition, the group used email and a project management software, Basecamp, to communicate.
Initially, I monitored the group’s work through Basecamp, after receiving their signed consent to participate forms. I identified two collaborative writing projects that the core group was working on simultaneously and decided to include both projects in the study. Because of the complexity of the phenomenon of collaborative writing, it is important that in-depth data be collected on both individual perceptions and group perceptions. Since one of the things that I would like to understand is how the dynamics of individuals and groups affect the creation of knowledge at the individual level, the question of agency and enforced structure (from the context) is very important. In order to understand the contexts, group dynamics, and individual perceptions, I studied the one group in-depth before, during, and after collaborating on these common documents. Looking at the two documents gave two contexts in which to study the same group.
This study was divided into three phases. The first phase, pre-task and collaboration consisted of a group of interviews looking at what an individual perceives he or she brings into the collaborative writing process and the perceived context in which he or she was collaborating. In addition to the interviews, data was collected on group interaction in Basecamp, meetings through meeting minutes, and additional documents such as email, document drafts, and planning documents.
The second set of individual interviews were conducted 3 months after the first set of interviews and the group interview. This had the unanticipated advantage of being able to collect examples of other documents, work, and group processes that were the result of those documents created during the collaborative process in the first set of interviews. In addition, it allowed me to explore group member perceptions of the organization, project, group, and knowledge creation over a longer time period.
I personally transcribed all interviews. The average length of these interviews was 45 minutes, the shortest being 38 minutes and the longest being 72 minutes long. In total, 15 interviews were conducted. With one exception, all transcriptions were verbatim from the audio taped recording. The one exception, due to malfunction of the recording equipment, had to be pieced together with partial recordings and notes from the interview. In addition, 40 documents were collected, including background information on the project, postings by group members on basecamp, drafts and final copies of documents studied, and additional documents that were the result of the documents studied.
Using Charmaz’s (2006) process for coding in constructed grounded theory, each transcribed interview was initially coded at the line level using action words. Starting with an interview from each department, initial codes interpreting the interviews were developed using action words. After the fourth interview, codes were reviewed and common terms were written on index cards. Using constant comparison within the interviews and between interviews some codes were combined, others were dropped, and new ones were created (Glasser & Strass as quoted in Charmaz, 2006, p. 54) based on the analysis of the remaining interview transcripts. Once it was obvious that there was a theoretical saturation in coding the data (Charmaz, Patten, 2002), these codes were then used to develop themes.
Looking at each question, codes were combined that developed themes/concepts or “relational statements” (Glasser & Strass as quoted in Patten, 2002, p. 490) that could be used to analyze the data and begin to develop theory.
Brandt, D. (2005). Writing for a living: Literacy and the knowledge economy. Written Communication, 22(2), 166-197.
Charmaz, K. (2006). Constructing Grounded Theory: A Practical Guide Through Qualitative Analysis. London: Sage Publications.
Clancy, W. (2006) Observation of Work Practices in Natural Settings. A. Ericsson, N. Charness, P. Feltovich & R. Hoffman (Eds.), Cambridge Handbook on Expertise and Expert Performance. New York: Cambridge University Press, pp. 127-145. Available at http://homepage.mac.com/wjclancey/~WJClancey/
Garfinkel, H. (1967). Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall, Inc.
Gersick, C. (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31(1), 9-31.
Hilbert, R. (1992). The classical roots of Ethnomethodology. Chapel Hill: The University of North Carolina Press.
Jehn, K., & Mannix, E. (2001). The dynamic nature of conflict: A longitudinal study of intragroup conflict and group performance. Academy of Management Journal, 44(2), 238-251.
Jehn, K., Northcraft, G., & Neale, M. (1999). Why differences make a difference: A field study of diversity, conflict, and performance in workgroups. Administrative Science Quarterly, 44, 741-763.
McGrath, J., Arrow, H., & Berdahl, J. (2000). The study of groups: Past, present, and future. Personality and Social Psychology Review, 4(1), 95-105.
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.
Patten, M. (2002). Qualitative Research & Evaluation Methods, 3rd Edition. Thousand Oaks: Sage Publications.
Schneider, B. (2002). Theorizing structure and agency in workplace writing: An ethnomethodological approach. Journal of Business and Technical Communication, 16, 196-214.
United States Census Bureau. (2002). North American industry classification system (NAICS): United States Government.
- V Yonkers
- Education, the knowledge society, the global market all connected through technology and cross-cultural communication skills are I am all about. I hope through this blog to both guide others and travel myself across disciplines, borders, theories, languages, and cultures in order to create connections to knowledge around the world. I teach at the University level in the areas of Business, Language, Communication, and Technology.