Challenging Methodological Traditions: Research by Email
Donna McAuliffe

The Qualitative Report, Volume 8, Number 1 March 2003


Engaging human service practitioners as partners in research about sensitive areas of front-line work can be difficult for a range of reasons. Time constraints, geographic limitations, trust in the research relationship, issues of privacy, and fear of professional judgment are only some of the barriers that researchers need to overcome in order to assist workers to become involved in a reflective process about areas of practice. This article outlines the development of a new method of qualitative data collection designed to aid the reflective process and assist practitioners to engage in an ongoing dialogue about complex ethical dilemmas they had experienced in relation to their work with clients, colleagues, managers and organisations. These ethical dilemmas occurred in the contexts of health, mental health, child protection, work with young people, community work, disability, family violence, aged care and research. This is the story of how the concept of Email-Facilitated Reflective Dialogue was born. It is the story of how Email-Facilitated Reflective Dialogue became a method of data generation and a tool for reflection on issues of ethics, how twenty social workers throughout Australia experienced it as a reflective medium, and how we, as partners in research, experienced and evaluated the process.

Key Words: Qualitative Research; Data Generation; Social Work Ethics; Ethical Dilemmas

The Seed of the Idea

I am one of those people who sleep with a notebook and pen beside my bed or under my pillow. I will be fast asleep only to find myself sitting bolt upright with thoughts and connections jangling around inside my head demanding release. I have learnt over the years that the clarity of reflection that comes at 2.00am deserves to be recorded, and so I dutifully awaken and put pen to paper. The reflective process takes many forms at this hour, ranging from a stream of consciousness flow to diagrammatic representations of concepts that may have eluded me during daylight hours. It is as though the cloak of darkness shields me from extraneous distractions and allows me a peaceful place in which to reflect on my actions, words, and ideas with honesty and lucidity. It was in this space one night some years ago, as I struggled with an array of methodological complexities, that I found myself grappling with the question of why it is often so difficult to engage social work practitioners in the research process. These thoughts started me on a journey that was to challenge the boundaries of qualitative data collection and generation.

My early reflections on the question of practitioner research took me back to my own experiences of having been a participant in qualitative research, as both an interviewee and a member of various focus groups. In these situations, the times and places were prearranged, the interview questions set, the tape-recorder in position, and the clock was always a centrepiece. It was only later when the interview was over and done, and the transcripts neatly collated, that I would think back on my contribution and wish I could be given a second opportunity to do it all again. I would think back and wish I had given the other answer to that particular question, been able to more fully articulate my feelings or explain my responses, or refused to answer a question altogether. I remembered later, as I spent time reflecting back over the interview, the small details that were an important part of the context of my story: Too late for those details to be included and perhaps they were irrelevant anyway. I considered, also, the many reasons that I had given in the past for not participating in a range of other potential research projects. As a busy practitioner, time was always an excuse. As much as I may have thought I had something to offer, when it came to nailing a time for an interview, it just never happened. Issues of access and geography were also raised in the past when I had read recruitment calls for participants in newsletters or journals and been disappointed that they were located in another state or another country. Economic realities restrict researchers from moving too far from home. In areas of sensitive research, there was the factor of physically coming face to face with a colleague or a peer, someone who might judge, condemn or disagree with decisions I had made or actions I had taken. How vulnerable is the professional reputation when practice, however well justified it might be, comes under scrutiny in the name of research.

As I swept my mind's eye across these questions and my own reluctance as a social worker to participate in research, I wondered whether there could be other ways of conducting research that would encourage engagement in the process by eliminating some of the barriers that I had identified. I recognised that one of the central issues was related to the reflective process that was needed to truly explore an issue in depth. I concluded that reflection on a critical incident from practice takes time; More time than that afforded in a one or two hour interview. I wondered whether data generated in a space and a timeframe that was governed by the research participant would be of better quality than that generated by a researcher-controlled agenda.

My musings around issues of methodology were synchronously matched to my emerging competency as a user of information technology. As a self-confessed luddite through my years in the world of practice, I had been reluctantly enticed out of my cocoon into the world of computers, and found that the opportunities for engagement with a virtual world were endless. In the process of moving from a community-based social worker to a full-time postgraduate student, I became infatuated with electronic mail, happily typing away to colleagues, peers, authors, and researchers - exploring electronic databases, on-line mediums and joining listservs and discussion groups. My experience as a list-lurker on the US 'socwork' listserv convinced me that social workers were more than prepared to engage in high levels of self-reflective flagellation in what was a relatively anonymous forum when given a topic to which they could relate and on which they had strong views. As I pondered the questions of the reluctant social work research participant, formulated my research questions, and reached a level of comfort and familiarity with my email communications, a methodological plan began to formulate. What if I could interview social workers by email? What if I could develop a process that enabled a focus on reflection of a critical incident, over time, at the other's own pace, giving them the power to edit, delete or write at 2.00am if they so chose. Would it be possible to sustain a reflective email dialogue over weeks or months? Would such a process be akin to blasphemy in traditional qualitative research circles? Or would it challenge the boundaries of traditional qualitative data collection, embrace a new medium, and enable social workers from around the country or overseas to engage in dialogue with me about sensitive issues of practice. I had no idea if anyone had done this before, and if so, how they dealt with lack of personal cues, voice tone and body language, development of rapport, trustworthiness of the data, and assurances of confidentiality and privacy. It was all unknown territory and guidance from others who might have blazed a trail was conspicuously absent.

My plans for exploring new ways of generating qualitative data were also influenced by the sensitivity of my research topic. I was interested in conducting exploratory research about how social workers experienced and managed ethical dilemmas in front-line practice. I expected to be discussing problems of breaches of confidentiality, violations of practitioner-client boundaries, difficulties with organisational practices, and conflicts between professional and personal values. These ethical issues have traditionally given social workers cause for much angst, and result in a sense of isolation for those who dare not seek support when struggling with moral quandaries (Banks, 1995; Holland & Kilpatrick, 1991; Loewenberg, Dolgoff, & Harrington, 2000). Given these particular sensitivities, and the potential for disclosure of unethical practice, I was alerted from an early stage to the possible barriers to conducting research in anything but the traditional way. I decided, however, to take the plunge into unknown waters and set about devising a strategic plan to develop a new method of computer-mediated data generation.

Testing the Waters

My first hurdle was to convince my supervisors that challenging methodological boundaries in the search for innovative ways of collecting and generating qualitative data would, within itself, be a valuable exercise. The development of new knowledge, finding the 'cutting edge', exploring unexplored territory - all of these are expected in some way in academic work. I was spurred on by the words of Shalamut Reinharz (1990, p. 239) who asserts that 'the feminist spirit is one of breaking free, including breaking free of methodological traditions'. Michael Quinn Patton (1990, p. 346) was also encouraging of creativity in research, advising that 'creative approaches are those that are situationally responsive, appropriate, credible and useful'. My supervisors encouraged me to search the literature, and put forward a proposal for consideration. Their support was greatly appreciated, as I knew that within academia, this was classified as 'risky business'.

I remember clearly around this time, a conversation that I had with a visiting lecturer who had written many books about qualitative research. As I enthusiastically set about explaining my ideas to conduct in-depth interviews by electronic mail, I was confronted with a stony silence. I was told, in no uncertain terms, that I was engaging in a process akin to sacrilege. Qualitative research, I was told, was about observation, non-verbal behaviours, cues, rapport, relationships - none of these could be achieved by connection between remote computers. This conversation could have put paid to any ideas to progress further. I decided instead to use these cornerstones of qualitative research as firm foundations and forge ahead to design a pilot study to test the parameters of my ideas. The idea of generating rich quality data through the process of reflective engagement in a dialogue, moved me a step beyond the development of a method of data collection, as it meant that the process itself became paramount, and charged me, as the researcher, with additional responsibility.

The next step was to search the research literature and ascertain whether there was any evidence of others having conducted in-depth interviews by electronic mail. This search threw me headfirst into a world dominated by quantitative methods where I discovered a wide range of computer-assisted interviewing methods. These methods were firmly located within positivist traditions and relied on surveys and questionnaires, where qualitative data was in short answer form only. I discovered new acronyms such as PAPI (pen and paper interview), CATI (computer-assisted telephone interviewing), CASI-V (question text on screen - visual) and EMS (electronic mail survey) (de Leeuw & Nicholls, 1996). As well as searching the print literature, I emailed prominent researchers around the world, both directly and through discussion lists, asking if anyone had any knowledge of qualitative data being collected by email. This was 1998 and no one had - but I found that many treated this as a novel idea and I received a great deal of encouragement. As my search continued, I became more convinced that no empirical literature was to be found in the social sciences to either support or discredit the use of electronic mail as a method of systematic data collection aimed at eliciting information-rich descriptive and reflective accounts of lived experiences. Selwyn and Robson (1998, p. 2) confirmed this lack of evidence, stating 'given its growing importance as a medium of communication, discussion of email as an academic research tool has, to date, been scarce'.

Although there was an absence of literature (at that time) to support the proposed method of data generation, I was able to locate a number of comparative works that explored the advantages and disadvantages of the use of computers in both social work and research. Galinsky, Schopler, and Abell (1997) had conducted research into the use of technology-based support groups in social work practice and cited convenience, anonymity and accessibility as benefits. Lack of interpersonal cues, technological glitches, and possibilities for deceit, were noted as potential problems. Bunting, Russell, and Gregory (1998) used electronic mail for international collaborative data analysis in health research and listed similar benefits and problems of that approach. Miller and Gergen (1998) researched the use of an electronic bulletin board devoted to the therapeutic discussion of suicide and concluded that the benefits outweighed the disadvantages, particularly in relation to cost, convenience, anonymity and lack of social markers that stimulate prejudice. These studies provided useful indicators for technological and interpersonal issues that I needed to address when designing the methodology. Had I conducted my research even three years later, I would have had the benefit of an excellent resource on online qualitative research developed by Mann and Stewart (2000) to assist me on my journey, and this would have answered many of the questions that I answered by trial and error. My experience, however, preceded the publication of this book. There has also, over the last few years, been a growth in the body of literature in education, highlighting the gains that have been made in the exploration of online learning modalities to assist students and those with English as a second language to improve communication skills (Dorman, 1998; Lapp, 2000). Email communication is at the forefront of many of these initiatives.

Taking the Plunge

I experienced a growing sense of excitement tinged with trepidation, as I prepared to conduct a pilot study to test the feasibility of what I had now termed 'Email-Facilitated Reflective Dialogue'. I needed to ascertain the scope of potential technological difficulties such as incompatible computer systems, and inevitable Internet service provider problems. I had no idea how long it would take to collect the data in this way - a week? A month? Six months? How would I store, retrieve and analyse the data? Was it possible to establish rapport with people I had never met or spoken to? Would I be able to respond to emotive content in a sensitive way in writing, without the immediate ability to reach out and offer support and consolation where needed? And what of the ethical issues relating to confidentiality and privacy? As I thought through these issues, I realised that I had a responsibility to rigorous testing before moving forward to conduct a larger study. I worried that my relative inexperience with computers would let me down, and I wondered in those 2.00am moments, whether it would be much safer to retreat to the safety of traditional qualitative data collection by face-to-face in-depth interviews.

I recruited four social workers for participation in the pilot study. I had had some previous discussions about the possibilities of conducting email by research with two of these people while attending a conference the year before, and it was their encouragement and enthusiasm that spurred me onwards. Both were from interstate and had asked me to contact them if I ever reached the stage where they could assist me in any way. The other two social workers were local, and had also indicated their willingness to engage in a research pilot. I sent all four carefully worded consent forms and detailed information about the study and the method. We established email connections. We ensured that technical support was at hand if necessary. We did not meet again, and had no telephone contact. Then, over the next month, we set about engaging in a reflective dialogue about ethical dilemmas in social work practice.

I learned a great deal from the experience of the pilot study. I learned that technological problems could be overcome, and that data could be organised in a systematic way in word processing files, and could be easily transferred across to the Ethnograph for data analysis. I also learned that email is far from a secure system and that confidentiality issues cannot be downplayed. One of the pilot participants was writing from interstate and was using her home computer to write at night. She had been writing about a complex issue that involved boundary issues in relation to a client. Her husband had no knowledge of certain aspects of the situation, which had some relevance for their relationship, until he inadvertently accessed one of our email messages. This situation, which I reacted to with horror (despite the fact that the drama was resolved at the other end), highlighted the need for participants to have password protection on email systems and to take responsibility for their writing in the knowledge that emails can potentially be accessed either at home or work by others.

As well as the ethical quandaries, my primary concern was whether or not I would be able to establish rapport with participants using an electronic medium. I found that in the absence of verbal and non-verbal cues, rapport needed to be established instead by the use of language. I needed to pay careful attention to how I worded opening and closing statements, and it was important to acknowledge feelings expressed by participants when they disclosed sensitive information or requested my opinions on actions that they had taken. It was important that my role as the researcher was clear, and I made statements to the effect that it was not my place or intention to pass judgement on actions or decisions, or to comment on whether the social worker had acted ethically or unethically. It was my role to engage in an in-depth exploratory dialogue in a supportive manner, assisting the practitioner to reflect back on an ethical dilemma and investigate its anatomy and impact. At the end of the pilot interviews, I found that I had rich quality data that gave excellent insight into the lived experiences of ethical conflict. Furthermore, I had developed on-line relationships with these people and felt that I knew them well within the context of the research. I also had clarification of a range of methodological concerns, and the positive evaluative comments convinced me that, with modification, Email-Facilitated Reflective Dialogue could prove a useful tool for generation of quality data. Those in 'higher places' assessed my research proposal, as did the University Ethics Committee, and encouraged me to continue.

Drowning in Data - Eventually

I learned my lesson in patience in the first few weeks of data generation. I had recruited 20 experienced social workers from around Australia, including one from overseas, for the study. These participants had responded to an article that I had written about the study for the National Newsletter of the Australian Association of Social Workers. All had access to email, either from work or home. All had complex ethical dilemmas they were prepared to write about. I had written to each of them about myself, telling them some details about my practice and reasons for my interest in ethics. This included a reassurance that I, too, had experienced many ethical dilemmas in my years of practice and understood the complexity and emotion that surrounds ethical conflict. Participants were encouraged to maintain regular contact and to inform me if they were experiencing any technical problems or expected a disruption to the flow of the dialogue. I made a commitment to responding to messages within 24 hours and to let them know of any delays experienced from my end. I sent out the initial message that signalled they could begin to write. And I waited…and waited.

I realised at the end of that first long week, where I sat anxiously glued to my computer screen waiting for the anticipated responses to come flooding in, that although I had provided a new opportunity for involvement in research, social workers were still clearing the depths of in-trays, and were still having to find valuable time to write. On that level, the convenience of email made little difference. But as the responses started to come in and I began to dialogue daily or weekly with these social workers, a relationship began to form between us. I had never met most of these people, and had spoken only briefly on the telephone to some of them. I continued to worry that I could not make eye contact, use words and gestures to establish rapport, put them at ease, gently elicit their stories, or offer non-verbal support. They could not gauge my response to their disclosures, show me the depth of their emotions, or gain my immediate reassurance that what they were telling me was valuable, and that I understood the decisions they had made. I was concerned that I was missing some of the basic tenets of good qualitative research. But despite this, relationships were formed. One participant sent me an electronic photograph of himself so that I could visualise him sitting at his desk as we wrote back and forth. Others began with apologies for taking time to get back to me, providing short insights into areas of their lives, telling me about the weather, their families, out of control workloads or federal election issues. And importantly, emotion began to enter the dialogues. Fear, sadness, confusion, frustration, anger, and helplessness permeated the dialogues as these social workers constructed and poured out their stories. As I read and re-read parts of the dialogues, I found myself more and more drawn into their worlds, supporting, gently probing, and waiting eagerly for the next message that would take us further.

The issue that presented me with the most difficulty was gauging when to 'prompt' for responses, particularly as there were at times lengthy delays between pieces of writing. I developed a range of 'prompts' that proved most successful in eliciting almost immediate response. I used, for example, the 'humour prompt' where I wrote 'just checking you haven't become lost in cyberspace…'. The 'something amiss at my end prompt' said 'it seems our system has been down - have I missed something you've sent?…'. The 'additional information prompt' asked " 'I've been reading back over your material and wonder if, when next you write, you could include some information on…'. The 'suspense prompt' simply stated 'the suspense is killing me…', while the 'general concern prompt' commented 'I haven't heard from you for a while so just wondering if everything is going OK at your end…?'.

In using these prompts, I found that I needed to maintain a fine balance between allowing participants to move at their own pace, and providing reminders that the research process was continuing. I worried that applying pressure would only yield poor data so tried to use prompts sparingly. In some cases, participants appreciated the prompts as these served as reminders for them to make some time in their schedules to continue their part of the dialogue. Time factors, workloads and priorities, study and family commitments posed the major impediments to regular writing. Apologies and lengthy explanations were often provided in response to a prompt message. As most people were writing from work computers, the research needed to be factored into daily routines. Interruptions were constant and participants often preferred to wait until they could find adequate space and time before settling into writing. This often meant that they would only have one opportunity a week to respond. As this email dialogue required a level of reflection, it was difficult to foresee how long it would take to conclude the data collection. I had certainly not factored these impediments into my data collection timeframe! Literature from the online learning field has now given me a fuller understanding of the issues of 'silence' that I encountered during this period, and supports my attempts to prompt responses. Experiences of online teaching in which students and educators have negotiated similar periods of 'silence' recommends that active communication and the avoidance of isolation is critical in sustaining online relationships (Benfield, 2000).

The extremes of time and style within the ethical dialogues were illustrated by two examples. At one end of the spectrum was a social worker who used the reflective dialogue process as a 'cathartic therapeutic experience' (her words) in which she documented an ethical dilemma that had caused her to leave the country and practice overseas. This intensely emotive process was concluded within a week and amounted to some 50 pages of transcript. We were both exhausted by the end of it. At the other end of the spectrum was a social worker who presented his ethical issue as a narrative-style story with players in the dilemma described as characters in a novel. The instalments, many quite lengthy, were written in creative literary style as the complex and difficult issues were explored in a way most comfortable for the writer. This dialogue was competed after 30 weeks. The other dialogues were completed between 5 to 24 weeks. To my surprise, and relief, there were no dropouts.

As each reflective dialogue drew to a close, I found myself not wanting to finalise the contact, and they wanting to keep telling me more. I was conscious of the dangers of personal experience methods, that 'one of the common experiences of those who focus on experience in all its messy complexity is that they lose track of the forest for the trees and find it hard to draw closure to a study' (Clandinin & Connelly 1998, p. 157). In the spirit of good research, we reached our conclusions and said our good-byes.

Sink or Swim

Due to the experimental nature of this method of data generation, I decided to conduct a further 10 in-depth interviews to allow for some comparative evaluation. The social workers that I engaged in this part of the study had expressed interest at the initial time of recruitment, but did not have secure email access in their workplaces and could not therefore be confident that their material would remain confidential. These social workers did not have access to computers at home, but were keen to share their experiences of ethical dilemmas. I conducted a two-hour, face-to-face semi-structured in-depth interview with this group, using the same research format and questions that I had presented in the email dialogues. In the comparison between those engaged in email dialogues, and those interviewed face-to-face, I looked for differences and similarities between my subjective relationship with participants, the structure, management and analysis of the data, and comparative time factors.

The relationship between researcher and participant and the establishment of rapport and trust are key issues in qualitative research. How that relationship is conducted, the boundaries constructed around it, and the ethical nature of agreements made are all critical factors in ensuring valid and reliable data (Minichiello, Aroni, Timewell & Alexander, 1990). How does this relationship develop if the researcher and participant never meet face-to-face, and if they do not even have telephone contact? I was preoccupied with this concern in the early stages of the development of the email dialogues and this influenced many of the steps that I took in the conduct of the study. I found that attention to the written word was critical in every response. When a participant emailed me an emotive message (and there were many of these given the topic), I agonised over my response. Without body language and non-verbal responses to accompany written sentiments, it was critically important that the words conveyed the intended meaning. I chose my words with great care. I used a combination of reflective comments and questioning about content, light humour where appropriate, a sharing at times of my understanding of a point from my own experiences, and acknowledgment of the reality of their experiences. This created a relationship that in most cases, flowed easily backwards and forwards and by the end of each dialogue, I felt that I had come to know each person well.

The major contrast in conducting the face-to-face interviews was not in the content of the data or structure of the discussion, but in the confrontation of raw emotion. Reading emotive language, as powerful as this can be, is quite different from coming face-to-face with anger, despair, helplessness, confusion, frustration, fear, and sadness. These were some of the emotions displayed openly by social workers during interviews. This served as a valuable reality check and reminder that similar emotions would have been likely had I interviewed those involved in the email dialogues. Ethical dilemmas by their very nature create emotion and this dimension may well have become lost in the analysis had only email dialogues been conducted. In terms of relationship, I did not feel that I had established better rapport with one group over the other. The email dialogues felt, in some cases, more honest due to their anonymity, and the relationship was entrenched over a much longer period of time. The interviews, however, had an impact that was sharp and deep.

In relation to participant recruitment and selection, one of the major advantages of email was the ability to broaden the sample population to a national level, and to include practitioners from rural and isolated areas. This also extended to one overseas participant and further illustrated the advantages of email research to overcome geographical limitations and encourage involvement by practitioners who may not normally be able to access interviews or focus groups. The selection of participants for face-to-face interview, on the other hand, was limited to local practitioners that I could access easily. They had the choice of whether the interviews were conducted in their home, workplace, or in my office.

The email dialogues were more structured in a chronological sense as writing the story enabled a sequence of events to unfold, and participants had the opportunity to re-read what they had written and edit it if necessary before sending. This was one of the primary advantages of data generated in this way - the asynchronistic nature of email communication allowed participants to consider their responses without the pressures of time. They could focus clearly on my written responses and questions, and could respond to these accordingly. It was much easier for those involved in interview to become sidetracked as we discussed the multitude of complexities surrounding the ethical dilemma. One issue that I quickly became aware of was that it was much easier for the email participants to ignore difficult questions. They simply did not respond. Those interviewed were not as able to sidestep similar difficult questions, as I was able to probe and refocus.

The major timesaving advantage of data generation by email was that transcripts were automatically created in a format that required little change. There were, however, other parts of managing the email process that did require quite significant amounts of time. These included the time spent in negotiating the beginning of the process; reading and re-reading of instalments as they arrived; the formulation of responses and comments; the management of data across computer systems and the attention to confidentiality requirements; the prompting and support of participants to continue writing; the monitoring of data quality and adherence to the research questions; feedback of completed transcripts; and some data cleansing/reduction in preparation for entry into a data analysis package. In-depth interviews, on the other hand, were completed within two hours and could be immediately transcribed and imported directly into a computer-based data analysis package such as the Ethnograph. Whether there are timesavings to be made in using email in preference to interview is largely dependent on the participants. If questions are short, focused, and concentrated on one area of inquiry, then email dialogues may provide a faster method of data collection, as they do not require intensive transcription time.

Reaching the Shore

From the outset, I made it clear to participants that this was an experimental method of data generation, and that I would be looking for some evaluative comments at the completion of each dialogue. From the number of prompts that I had developed to keep the process moving, I was expecting more negative than positive appraisals. I was interested to find an overwhelmingly positive response to the method characterised by the use of terms such as 'interesting and worthwhile', 'excellent method', 'exciting experience', 'easy, efficient and effective' and 'an innovative process'. The main advantages of email research were summarised as the ability to remain anonymous (honesty), flexibility and convenience, access for those not normally included in research, ability to focus on reflection, and a greater sense of control of the process. One participant commented that the method

…provided me with the circumstances I needed to be reflective…I found the act of composition from the keyboard taking me places I had no conscious intention of going and exploring depths and nuances I didn't anticipate... the freedom one has at the subject end to expatiate in solitude upon the questions put by the researcher certainly seems to be conducive to more searching exploration of the heart and mind.

Another wrote:

It offers an easy way for people to become involved in a research project. I mean we do not even have to leave our desk!.

These positive comments were balanced by some consideration of the difficulties and the use of terms such as 'disjointed', 'difficult', and 'limiting'. The primary disadvantages related to external factors such as access and time constraints, and discomfort with the medium. A participant who experienced frustration with writing commented:

I found the process somewhat difficult, in part because of the subject matter and the time elapsed, but also because of the process which was 'disjointed'. I suspect some folks may drop out because it is easier to do an interview in one hit rather than to have to come back and forward…the method will also be somewhat affected by people's ease/verbosity to type. If you are a slow typer or not one to put pen to paper, this method may be frustrating.

These comments, as well as many others contributed by the social workers in this research, indicate that practitioners are prepared to engage in new ways of exploring sensitive aspects of practice. From my perspective, generating qualitative data by email allowed high quality data to be gathered in a systematic and well-organised fashion, without the need for time-consuming transcription of interview tapes. I was able to spend time digesting and thinking about issues raised in the written material, and send back incisive questions and responses that kept the dialogue on the track of the research questions. There were surprisingly few technological hiccups and no data was lost during transit. I was able to keep each participant's dialogue in a separate computer file with all identifying details immediately deleted. The downside was in the delays that interrupted the flow of communication due to people going on holidays, being inundated with work or losing access to a computer. The success of the method appears proven by the fact that I obtained 20 comprehensive email dialogues, each running from between 3,000 and 30,000 words on which to base my exploration and analysis of ethical dilemmas experienced by front-line social workers.

A Word of Caution

As I reflect back on the experience of developing a new method of collecting and generating qualitative data that increased opportunities for practitioner involvement in research, I recognise the need to proceed with some caution in the future. Research needs to be rigorous in its attention to detail in relation to methodology. I am fearful that some may find the use of computer-mediated communications a quick, easy, cost-effective and resourceful way of collecting data. While this is true in many respects, convenience should not, to my mind, be the guiding force behind adoption of such a method. I developed Email Facilitated Reflective Dialogue as a 'reflective space' for practitioners to consider issues of ethical practice. Essentially, 'reflective dialogue' was the central concept, not the facilitation of research by email. The focus on generation of quality data, as I have said earlier, is a more active process that pays attention to language, relationships, and interpretation of silence. Many have asked me whether I would use this method again as a way of collecting qualitative data? My answer is 'with caution'. As computer technology becomes more a part of our working lives, and of the lives of our colleagues and clients, we should continue to look for ways of using such technology to our advantage in research. Choice of methodology that is appropriate to the research questions, and to the sample, is governed by a multitude of issues, and these need full consideration before making the decision to adopt computer-mediated methods. Thankfully, there is a newly emerging body of literature about online communications that will provide a guiding beacon for online researchers of the future.


      Banks, S. (1995). Ethics and values in social work. London: McMillan Press.

      Benfield, G. (2000, July). Teaching on the Web - Exploring the meanings of silence. UltiBASE e-Journal. Retrieved July 4, 2002, from

      Bunting, S. M., Russell, C. K., & Gregory, D. M. (1998). Use of electronic mail (Email) for concept synthesis: An international collaborative project. Qualitative Health Research, 8(1), 128-135.

      Clandinin, D. J., & Connelly, M. (1998). Personal experience methods. In N. K. Denzin & Y. S. Lincoln (Eds). Collecting and interpreting qualitative materials (pp. 15-178). Thousand Oaks: Sage Publications.

      de Leeuw, E., & Nicholls, W. (1996). Technological innovations in data collection: Acceptance, data quality and costs. Sociological Research Online, 1(4). Retrieved July 4, 2002, from

      Dorman, M. (1998). Using email to enhance instruction. Journal of School Health, 68(6), 260-267.

      Galinsky, M. J., Schopler, J. H., & Abell, M. D. (1997). Connecting group members through telephone and computer groups. Health and Social Work, 22(3), 181-188.

      Holland, T. P., & Kilpatrick, A. C. (1991). Ethical issues in social work: Towards a grounded theory of professional ethics. Social Work, 36(2), 138-144.

      Lapp, S. I. (2000). Using email dialogue to generate communication in an English as a second language classroom. Australian Journal of Language and Literacy, 23(1), 50-62.

      Loewenberg, F. M., Dolgoff, R., & Harrington, D. (2000). Ethical decisions for social work practice. Itasca, Illinois: F.E. Peacock Publishers.

      Mann, C., & Stewart, F. (2000). Internet communication and qualitative research: A handbook for researching online. London: Sage Publications.

      Miller, J. K., & Gergen, K. J. (1998). Life on the line: The therapeutic potentials of computer-mediated conversation. Journal of Marital and Family Therapy, 24(2), 189-202.

      Minichiello, V., Aroni, R., Timewell, E., & Alexander, L. (1990). In-depth interviewing. Melbourne: Longman Cheshire.

      Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage Publications.

      Reinharz, S. (1992). Feminist methods in social research. New York: Oxford University Press.

      Selwyn, N., & Robson, K. (1998). Using email as a research tool. Social Research Update, Issue Twenty-One. Retrieved July 4, 2002, from

Author Note

+Dr Donna McAuliffe is a lecturer in the School of Social Work and Social Policy, University of Queensland. She teaches courses in professional ethics, social work theory and direct practice at both undergraduate and postgraduate levels and completed her doctoral research into ethical dilemmas in social work. Donna has worked for many years as a social work practitioner in both the government and community sectors in Queensland, Australia. Her areas of research interest include ethical practice and qualitative practice-based research. She can be contacted School of Social Work and Social Policy; University of Queensland; St Lucia, Brisbane; Queensland; AUSTRALIA 4072; her email address is

Article Citation

      McAuliffe, D. (2003, March). Challenging methodological traditions: Research by email. The Qualitative Report, 8(1). Retrieved [Insert date here], from

Donna McAuliffe
2003 copyright

Return to the top of the paper.

Return to the Table of Contents.