Methodology

Theoretical framework

This project used grounded theory, a research methodology originally conceived of by Glaser and Strauss (1967) as a means of generating theory in qualitative research.

It emphasizes the discovery and development of theory through an overlapping process of data collection, note taking, coding, sampling and comparison.

It differs from most other approaches in that it aims to help researchers grasp theory as it emerges from research, instead of using theory to test data once the collection stage is complete.

Researchers using grounded theory don't execute their research processes in a linear form. Instead, the processes of data collection, analysis and theory generation are fluid, often occurring simultaneously. As well, it conceives of researchers as social beings -- not impartial observers -- complete with valid previous experiences that help them better understand the subject being observed (Baker, Wuest, & Stern, 1992, p. 1357).

The methodology seemed well suited to the project because the scarcity of data on online journalism education and the newness of the field meant there were few theories to verify and little external data for comparison. It was also important to have leeway within the research method to revise and sharpen lines of questioning as sources illuminated the issues involved. Finally, it seemed most useful to the participants who requested a copy of this research that they glean straightforward insights into what was happening in their field without the top-down application of theory.

Two key components of the Glaser and Strauss (1967) model were the importance of theory emerging from data and the role of the researcher in enabling this. Grounded theory, the authors argued, requires hypotheses to be "systematically worked out in relation to the data during the course of the research" (p. 6). Further, they asserted that reading too much of other people's thoughts on a subject before acquiring data is an obstacle to this process of inquiry. In fact, they stated, reviewing literature prior to a study threatens the quality of results, arguing:

An effective strategy is, at first, literally to ignore the literature of theory and fact on the area under study, in order to assure that the emergence of categories will not be contaminated by concepts more suited to different areas (p. 37).

At first glance, these theoretical underpinnings seem a comfortable fit with this study's subject matter and the audience for the results -- journalism instructors. The methodology of journalism is, by its nature emergent, as reporters hone story ideas and interpret complex issues by researching, interviewing and corroborating. Further, the notion of the researcher aiming to begin a study with a "blank slate" of pre-conceived theories and categorizations compares favourably with the idealized aims of objectivity in reporting -- "the effort to understand the reality outside our minds" (Cumming & McKercher, 1994, p. 14) and journalistic synthesis -- making "it your own, rather than writing in bits and pieces of your sources' language" (Cumming & McKercher, 1994, p. 12).

Cumming & McKercher (1994), in fact, made powerful (if unintentional) allusions to key elements of grounded theory when they compared journalists to historians. The authors cited a quote by historian Barbara Tuchman as closely representing the work of journalists: "If the historian will submit himself to the material instead of trying to impose himself on the material (italics are the author's), then the material will ultimately speak to him" (p. 22).

However, practitioners of grounded theory have pointed out shortcomings in this original model. Some have suggested the researcher requires, in fact, extensive pre-understanding of the subject matter in order to focus research effectively (Allan, 2003; Selden, 2004). Others have called the coding process -- specifically the model of word-by-word analysis advanced in later years by Strauss and Corbin, confusing, difficult to administer and overly mechanistic (Sarker, Lau & Sahay, 2000; Allan, 2003). Another has called the initial grounded theory approach rigidly positivist and unrealistic in its conception of human nature.

Charmaz (2005) argued for a constructivist approach to grounded theory that accepts the researcher as fully human with a unique view of reality. She believed researchers must firmly locate themselves in the real world, not try to divorce themselves from it. She saw data analysis as an interpretation of reality, not an objective reporting of it. In short, she argued:

A constructivist grounded theory adopts grounded theory guidelines as tools but does not subscribe to the objectivist, positivist assumptions in its earlier foundations ... It does not assume that data simply await discovery in an external world or that methodological procedures will correct limited views of the studied world. Nor does it assume that impartial observers enter the research scene without an interpretive frame of reference (p. 509).

Glaser (2002) took issue with Charmaz (and her earlier writings on the subject), contending that his initial conception of grounded theory fully incorporated constructivist approaches. "Let us be clear," he argued. "Researchers are human beings." However, he immediately maintained that personal bias and interpretations needed to be "rendered objective to a high degree by most research methods and grounded theory in particular by looking at many cases of the same phenomenon ...to correct for bias" (para. 44).

But it's Glaser's treatment of "data as something separate from the researcher" and his implied aspiration to entirely neutral analysis that Charmaz (2005) saw as problematic (p. 510). "What we know shapes, but does not necessarily determine, what we 'find,'" she argues (p. 510).

Indeed, Charmaz's view of grounded theory seems a better theoretical fit for this study. It parallels Cumming & McKercher's (1994) conception of a more appropriate and nuanced ideal of journalistic impartiality -- that reporters simply try to be objective about their subjectivity. As well, it accommodates a key factor of this project -- that I am already immersed in the field of online journalism, by nature of my full-time job. I have read extensively on the subject. I have well-formed opinions on some of the questions I pose in the study. I've had previous professional dealings with one of the participants. In short, while I aim for a neutral analysis, my biases may not be correctable; my past experiences may shape the study beyond my ability to understand their full implications.

Charmaz further argued strongly for a critical approach to grounded theory -- one that would apply lenses of power and privilege to the subject matter. The study's findings certainly opened the door for exploration using this approach. For example, the demands and challenges cited by participants prompted inquiry into:

  • the reasons why many courses are taught by part-timers
  • whether institutions' requirements for tenure (extensive professional experience and academic certification) worked against the hiring of instructors who could teach effectively in a still new and rapidly-evolving field driven by young audiences
  • whether the structure of the courses (duration, scope, etc.) set instructors up for failure in teaching multi-media production -- especially when many institutions dedicate whole courses to the production of content in a single medium.

However, a critical approach didn't seem to address the fundamental topics I was interested in exploring -- the nature of teaching methods and their success with students. That is, I was less interested in the plight of the instructors themselves, than in how educators conducted their instruction. An interpretative approach seemed likely to yield the most useful data.

Finally, some other research methods seemed initially attractive but had shortcomings. Investigation of a single issue inherent in a phenomenological approach was simply too restrictive. An ethnographic study would have yielded rich data, but on-site interviewing and observation were impractical within the circumstances available for this study. An action research approach would have been useful too in exploring this community of practice, but, again, the available circumstances made involving participants in the research process difficult. And lastly, longer case studies might have been ideal in providing richer description of teaching practices. But ultimately, the circumstances of the participants were quite different and sampling a broad range of perspectives seemed likely to yield optimal results. Using fewer, shorter case studies seemed a solid alternative.

Research procedure

This project consisted of phone interviews with instructors at English-language Canadian colleges and universities who taught a practical course in online journalism, new media writing or reporting. The Association of Universities and Colleges of Canada and the Association of Canadian Community Colleges list 16 such institutions in the country. I sought to find faculty members who required their students to produce a form of journalism or narrative opinion online for their course in 2005-06. I did not include instructors who taught solely Internet research or production skills, or instructors who taught a lecture or seminar devoted exclusively to the discussion of new media issues. This study aimed to capture the experiences of instructors teaching students how to produce journalism -- not discuss it or master the mechanics of publishing it.

I found the instructors by browsing each institution's website to find suitable courses and contacts. I contacted the schools by phone or e-mail if they didn't appear to offer a suitable course or didn't provide current contact details for their instructors. The aim was to interview 10 to 15 instructors for the project.

I introduced my project in a personal e-mail message to prospective participants and directed them specifically to a website about the project. If they indicated an interest in participating, I asked them to reply to an e-mail consent statement that outlined the terms of the research (see Appendix B) and suggested times for a phone interview.

The interview consisted of 16 questions (see Appendix A) -- some of them multi-part. Four questions were factual and aimed to extract information about the instructor's circumstances. Four dealt with the instructor's goals for students. The final eight concerned the content and methods of teaching. I also asked participants to send me a copy of their course outline.

The interviews lasted between 40 minutes to an hour. I contacted a few participants a second time via either e-mail or phone to prompt them to clarify or expand upon their comments.

I recorded the interviews and used grounded theory to extract themes and sub-themes from our conversations.

Finally, I sought approval from three to use their comments and identities as case studies of approaches to the study of online journalism. I assured all of the other participants they would remain anonymous.

I selected participants for the case studies primarily according to the depth and range of the subjects' responses. As well, I attempted to choose interview subjects from different environments (large/small institution, college/university, type of course taught) who addressed different issues in online journalism education.

I chose grounded theory because the scarcity of data on online journalism education and the newness of the field highlighted the need to examine it without the constraints of pre-determined analysis. This field is still so new that prospect of discovering emerging themes seemed likely to yield clearer results than attempting to place it in a greater theoretical context. It was also important to have the leeway to pursue divergent avenues of questioning with the participants, because their circumstances were dissimilar in important ways. This is, in part, because academic institutions are still experimenting with how to teach courses that are, in many cases, only a few years old. Further, it seemed most useful to the participants who requested a copy of this research that they glean straightforward insights into what was happening in the field without the top-down application of theory.

I applied grounded theory by taking notes during the interviews. My comments noted insightful responses from the participants, posed further questions to them and referenced comments from other instructors. Immediately following the interview, I began the process of coding by briefly summarizing the participant's situation, their aims and approach at the top of the collection of notes for each participant.

I then transcribed the interviews into an electronic document, adding memos in coloured text beside the participant's comments that highlighted relationships to issues addressed by other participants.

I continued the coding process by re-reading the interview transcripts and categorizing responses in a separate document according to emerging themes. I then used these themes as a basis for re-reading the interviews a third time and examining participants' responses in the new context. I returned to reading the interviews a fourth time as I began the process of writing this report and describing the themes in greater detail.

Next: Findings