How to...
Write for a practitioner audience

The following is a summary of our main practitioner titles. For more information, click on the journal title to go direct to the journal's home page and find out more about the sort of articles it publishes (journal information), how to prepare your manuscript (author guidelines), its editorial team, and to read some sample articles.

Click on the journal title to go direct to the journal's home page:

A training manager at a large multinational company has developed a set of training materials using the Web to deliver all product-related training to the company's worldwide sites, after a strategic decision has been taken to use this method in preference to face-to-face training. She is now thinking of publishing a case study and is considering what it is about her experience that would benefit other people. In particular, she is asking herself:

• Was there anything new about the technology?
• Was there anything new about the teaching strategy (she used computer-assisted assessment (CAA) and some multimedia, both of which are tried and tested methods)?
• Did the experiment result in real learning benefits, demonstrably greater than those that would have been achieved in face-to-face training?
• Was the method of delivery new (in fact, several other organisations have adopted a similar method of delivery)?
• Was there anything new about the way in which the project was evaluated?

In other words, the training manager needed to consider how her experience would offer substantial business benefits to her counterparts in other organisations. If she had, say, set up webcams in training centres at the local sites to collect data about the use of the training material, then this would have been an unusual and innovative method of evaluating training at a distance.

Use of web-based materials in themselves, especially with CAA which is tried and tested, is not particularly innovative and, therefore, lessons will have little explanatory power.

(Referring to the example above) How can one most effectively evaluate training when this takes place on a worldwide basis?

What makes a community of practice operate successfully when it is enabled by online interactive technologies?

"The aim of the study was to explore why corporate social responsibility (CSR) is a concern for firms, what it means for them, and how it is conducted and communicated. The research questions were:

• Why, for what purposes is CSR necessary or useful for a company? What kinds of motives, purposes or expectations regarding CSR are there?
• What is CSR? What is its rationale and philosophy, what kinds of principles steer it, and what does it consist of?
• How are CSR principles led and managed in practice? How are CSR programmes carried out?
• How is CSR communicated? What kind of communication strategies do companies have? How do they mirror public relation's definitions or paradigms?"

(Elisa Juholin, "For business or the good of all", Corporate Governance, Vol. 4 No. 3)

The organisation

The reported research project involved an in-depth case study of virtual communities of practice in a large multinational corporation, Caterpillar Inc. Caterpillar is a Fortune 100 corporation, manufacturing heavy construction and mining equipment. It employs more than 100,000 employees and its products are distributed in nearly 200 countries. The company's competitive advantage depends heavily on the utilization of the professional knowledge of its employees, especially mechanical, electrical, and chemical engineers, designers, new product development personnel, equipment and material testers, and others. Therefore, it is not surprising that knowledge management (KM) has long been a central concern for the company. 

Initially, Caterpillar's earlier KM efforts concentrated mostly on knowledge capture and dissemination. However, in the mid-1990s the company's strategy in this area started to shift to incorporate the community of practice ideas. Caterpillar's first virtual communities of practice (or "communities of knowledge sharing", as the company employees call them) emerged in 1997. At the time of writing, there were more than 600 online communities with 15,000 members worldwide. Caterpillar's communities are supported by the Knowledge Network, an Intranet-based system designed to provide an infrastructure for community functioning and supported by a group of KM technology experts, and employees of the Caterpillar corporate university. The system allows users to find subject area experts, post questions to specific experts or to the community at large, post and find knowledge entries, conduct online chats and asynchronous threaded discussions of questions and problems, and connect to numerous other online communities. Most of the communities were formed at the initiative of employees, not as a result of interventions planned by the KM group or the top executive team. Communities tend to form around specific subject matter expertise or professional activity areas but are open to all interested employees.

A typical community includes a community manager, one or more "delegates", a number of "experts", and "subscribers". Managers are typically elected by the team, and are, usually, senior, experienced members who have earned the team's respect through a strong history of contributions to the company. "Delegates" are associate managers, who can run the community in the manager's absence, or take on certain parts of the community management duties. "Experts" are people recognized for their skills and knowledge in certain areas. They actively participate in the community by posting knowledge entries, assisting managers in reviewing new postings, and answering questions posted to the community in general or to individual experts personally. Finally, any member of the organisation, interested in a community of practice's (CoP) subject matter, can become a "subscriber" to that community.

The study design

This research project was based on a qualitative case study design, with the main unit of analysis being three communities of practice. These three communities were selected using the purposive sampling approach. One of the communities was among the most well-established and successful communities, with a large (more than 1,000 employees) membership, and high community "traffic" (measured by the number of postings, permanent knowledge entries, and various online activities). The other two communities were classified as less successful and struggling to establish themselves. Both had smaller memberships (several dozen people), and significantly lower levels of online "traffic".

The major method of data collection was based on semi-structured interviews. In addition, the researchers collected a variety of company documentation, visited a total of five different sites housing various work units, and familiarized themselves with the functioning of the knowledge-sharing network over the company Intranet. Interviews were conducted with a total of 30 members, including managers of three communities, community experts, community members, and managers in administrative units responsible for managing and supporting the Knowledge Network (the software system at the backbone of Caterpillar's Internet communities). The procedure for selecting participants was that of purposive sampling: the lists of community participants and documentation on frequency of their participation in the community (the number of postings and knowledge entries contributed by them; the number of times they have accessed the system, etc.), provided by the Corporate University, were used to identify groups of heavy, moderate, and light users. Subsequently, random samples were drawn from each of the groups. The selected community members were contacted by email to solicit their participation in the study. Interviews with community managers and delegates were conducted face-to-face. The remaining interviews were accomplished by phone. Interviews lasted from 45 minutes to 2 hours, and were audio recorded and transcribed verbatim.

Interviews were conducted using a semi-structured schedule, developed based on the review of the literature and industry reports on knowledge management, virtual teamwork, and CoP, and the information obtained from Caterpillar representatives during the initial project planning meetings. The schedule was pilot tested in an interview with one active community delegate. The pilot test allowed the researchers to adjust several of the interview questions and eliminate questions that seemed redundant.

In addition to the interview data, researchers collected and analysed company documentation, which included: conference presentations and papers, developed by the Caterpillar employees, and describing the CoPs and KN; internally circulated manuals for KN users; reports and statistics on KN use and CoP membership and participation levels.

Data analysis

The interview data were analysed using two methods. First, the data obtained by dichotomous questions and questions that required the respondents to provide specific numbers were analysed by calculating descriptive statistics for the sample. Second, the narratives obtained by way of open-ended questions were coded and analysed by two researchers independently using the qualitative data presentation and analysis methods (Miles and Huberman, 1994). This included coding of individual interview data to identify major themes and categories, development of summary sheets for each interview, and development of cross-case data tables.

The qualitative analysis reported here involved an iterative process. The researchers started with coding the answers to open-ended questions, which resulted in identifying categories and issues pertaining to each of the questions. For example, to answer one of the questions, "What are the barriers to employees' contributing their knowledge to virtual knowledge-sharing communities?" the researchers’ content-analysed not only those segments of the transcripts where a specific question about barriers was asked, but also the whole transcript, trying to find relevant discussions. Two researchers performed the coding independently. Categories generated by the individual coders were compared and discussed in research meetings between the two coders. These discussions resulted in re-coding of some data and re-analysis of relevant segments. Once a consensus was reached on categories, the two coders went back to the interview material to ascertain how many categories were present. 

The iterative analysis of the interview data involved was augmented by the documentation analysis. This was accomplished by constantly referring to the information provided in the company documentation for validation. For example, participants' comments on the uses of the system were compared with relevant segments of the KN manual, and their statements regarding the frequency of certain uses were verified by referring to the statistics provided by the company.

Participant checks and validation

In "For business or the good of all", the author uses a case studie approach to examine the importance of corporate social responsibility in Finland (see example below). The methodology is described in a brief and straightforward way. Note how the author presupposes no knowledge on the part of the reader about how interviews work as a research method, what snowball sampling is, how the data analysis is described in fairly simple terms, without long technical description of the method of analysis.

Subjects and methods

One of the most important sources of case study information is the interview. Most commonly, case study interviews are open-ended in nature, and the investigator asks the respondents for facts as well as opinions. A second type of interview is the focused interview, or theme interview, in which the subject is interviewed for a brief time, say, an hour in a conversational manner. The interviewer leaves the questions rather open, allowing the interviewee to present his or her own view. The aim of the case study was to examine spoken argumentation as it mirrors the interviewees' own deliberations on the issue. 

The units of analysis or cases were a sample of Finnish companies, which sought to prioritise CSR. The subjects of the study were senior executives responsible for CSR issues in their organisations. They represented elite sources (Tuomi, 2003), which is seen to enhance the likelihood of obtaining most recent and relevant information. The sources of information were internet sites, newspaper articles and information received during the Congress of Corporate Social Responsibility held in Helsinki on 4 June 2002 by the Finnish Association of Communicators. The method known as snowball sampling proved to be a practical and efficient way of recruiting the subjects. Accordingly, the author opened the research process by making contact with a representative of Kesko, a company that has been highly profiled as a forerunner of CSR in Finland and Europe. The first interviewee recommended certain colleagues of his, who in turn recommended some of their colleagues. The number of interviewees available who have knowledge on the research theme was limited: a prominent respondent argued that he has less than ten colleagues in the world. Each participant in the case study comprises a "whole" study of its own.

Theory-oriented themes were found to be a suitable method for this study. The interview themes were the background and history of CSR (Buchholz, 1985; Carroll, 1995; L'Etang, 1996; Pearson, 1989), motives (Aula, 2002; Grunig, 1984, 1989, 1996), contents (Panapanaan, 2001), and organisation and communication (Grunig, 1984, 1989, 1996; Panteleeva, 2002). An additional challenge was to create a core category based on the thinking of the grounded theory (Strauss, 1994).

The data were analysed case by case and theme by theme. The analysis was carried out by applying Dey's method (1993). Accordingly, the analyses are reduced in a process in which the interview data are first classified into categories for comparison with the theoretical themes. The categories are then assigned to themes. Finally, the categories are united to represent more general typologies.

In practice the procedure consisted of the following phases. Seven of the eight interviews were audio recorded and transcribed. The first and the second in-depth reading of each transcript simultaneously involved listening back to the recording. (O'Dwyer, 2002). Using the interview guide (themes) as a broad framework each transcript was subject an iterative reading. This resulted in the creation of a mind map for each case. The mind map was a significant help in identifying myriad themes and sub themes in the data.

The interviews initially yielded nine categories. A further four categories emerged during a close reading of the interview transcripts. 

In the following stage all 13 categories were entered into a summary table, where the core contents of every case and every category were recorded. This approached showed a need to combine some of the categories due to overlapping contents, which generated six categories and six typologies.

(Elisa Juholin, "For business or the good of all? A Finnish approach to corporate social responsibility", Corporate Governance, Vol. 4 No. 3).