Copyright Sociological Research Online, 2001

 

Nicola Illingworth (2001) 'The Internet Matters: Exploring the Use of the Internet as a Research Tool'
Sociological Research Online, vol. 6, no. 2, <http://www.socresonline.org.uk/6/2/illingworth.html>

To cite articles published in Sociological Research Online, please reference the above information and include paragraph numbers if necessary

Received: 2001/4/26      Accepted: 2001/8/16      Published: 2001/8/31

Abstract

The arrival of the virtual realm and computer mediated communication (CMC) continues to attract considerable interest from a wide range of disciplines. Hine (2000) has suggested that previously negative understandings of CMC have been transcended. The virtual realm is now welcomed as a site for richer and more sustained interaction than previously envisaged. For the research community, the rapid development of the World Wide Web has opened new horizons and provided access to a new frontier and tool for data collection. The researcher can now engage in research on a world-wide, low cost, almost instantaneous scale - and in ways which potentially overcome some of the barriers imposed by more conventional research approaches. However, this somewhat idealistic view obscures both methodological and ethical difficulties that have become apparent throughout this research. If these difficulties are left unchecked, they may serve to undermine the use of the Internet as a tool for social research. The primary aim of this paper is to expose these difficulties and thus broaden the scope of discourse surrounding the Internet. A secondary aim is to explore the implications of the use of the Internet for the feminist methodological and research project. My aim here is to problematise the transference of existing methodological frameworks to an online setting. In this respect, I have presented this paper in the form of a research trajectory, outlining the course of my research from its conception to latter stages. The intention here is to suggest an avoidance of the use of the Internet as an 'easy option' and encourage a more developed focus on the justification, applicability and benefits of Internet research to the particular project. What has become apparent is that the effectiveness of CMC is much dependent on who is being researched, what is being researched and why.

Keywords:
Computer Mediated Communication(CMC); E-mail Interviewing; Ethics; Feminist Methodology; Internet Methodology; Methodological Dangers

Introduction

1.1
The primary aim of this paper is to contribute to the debates concerning the value and potential of the Internet as a research tool. More specifically, the temptation to see the development of this collection of new research techniques as providing an easily accessible route to previously hidden populations must be countered. For some scholars, the Internet provides a potentially irresistible terrain, a 'seductive data set' (Jones, 1999: 12) accessed at the touch of a button. However, while many scholars are writing about the Internet, the significance of virtual reality and the development of the virtual community (Hine, 2000: Jones, 1995, 1997, 1998), few are writing about its application as a research tool, the problems encountered accessing respondents and the effectiveness of transferring existing methodological frameworks to the realm of the Internet (Jones, 1999). While this paper explores issues which underline the potential of the Internet as a research tool, a cautionary tale is also added concerning the need to ethically appraise and justify the use of computer mediated communication (CMC). Further, this paper intends to counter the idealistic image of CMC as an equalising phenomenon - a portrayal overly influenced by an individualistic Cartesian view of the subject (Spears & Lea, 1994).

1.2
Of primary importance to the research community must be the issue of justification - the extent to which this type of investigation benefits and is applicable to, the research project. While avoidance of this new arena is not the resolution to these difficulties, the use of the 'Internet for Internet's sake' must be resisted. In this respect, this paper offers a 'research trajectory' and follows the general course of this research project from its inception to completion. Reliance on the Internet as a research tool was as much a product of the research process itself than a pre-planned strategy to access a somewhat hidden and sensitive population.

1.3
A secondary aim of this paper, however, is to explore the implications of the use of the Internet for the feminist research project. Feminist technoscience studies have grappled with the problem of the distinction between the social and the material in sociological understandings (Haraway, 1991; Wacjman, 1991) and feminist scholars are exploring women's participation in and use of the Internet (Spender, 1995; Harcourt, 1999). Recent studies have shown firstly, that the number of women using the Internet has increased dramatically and, secondly, that women especially like the 'chatting' capabilities of the Internet (Spender, 1995; Stewart et al, 1999). Turkle 1995, (as cited in Stewart et al, 1999) also suggests women are more likely to express themselves when their gender identity is hidden. Yet, little has been discussed regarding the more specific implications of the Internet as a tool within the feminist methodological and research project. My purpose here is to outline these deficiencies, alongside both the benefits and problems encountered accessing female respondents in an online setting.

Background, theoretical approach and arguments

2.1
My research focus emerged from an exploration of the diverse feminist debates concerning the development of assisted reproductive technologies (Corea, 1985; Stanworth, 1987; Smart, 1992; Oakley, 1993). Critique within this field has seldom sought the perspectives of the involuntarily childless and can arguably be termed conceptually inadequate - an ironic position which laid the foundations for my research (Illingworth, 2000). I argued that firstly, any assessment or critique of the nature of assisted reproductive technologies is limited in the absence of these perspectives. Secondly, this pointed to the need to explore the experience of infertility and involuntary childlessness and accept that such subjectivity is valid. Finally, my aim was to contribute to a less restricted appraisal of the development of assisted reproductive technologies by incorporating more effective theories of experience, thus lifting the self-imposed restrictions of feminist thought within this field. To ignore such subjectivity is to ignore experiences which bear little relationship to detractors in the field and which may have significant implications for the direction and focus of future critique (Gimenez, 1991). As Rose (1987) has argued, the rise of the feminist movement has been characterised, in part, by a concern with a woman's right not to have children - an issue which has assumed far greater importance than the fight to ensure women can.

2.2
Methodologically, this pointed to the need for a research method which adequately captured and reflected women's experience. If feminist critique was to move past its limitations, the experiences of the involuntary childless needed to be uncovered and captured, underlining the importance of experience in producing more representative accounts of women's lives (Stanley and Wise, 1990; Harding, 1986; Smith, 1988).

Sensitivity

3.1
I was aware of the potentially sensitive nature of this project - from both a personal and respondent point of view. From the outset, this demanded a methodology that recognised and explored my role within the research process and the need for the employment of conscious subjectivity, as opposed to the hierarchic and detached objectivism of conventional textbook methodology. Particularly when researching sensitive topics, the need for the respondent to be able to feel comfortable and in control becomes crucial (Renzetti and Lee, 1993). The dominant positivist requirement for the compartmentalisation of the researcher/interviewer role - researcher first and person second - is out of place in a sensitive situation and particularly in this study where women would necessarily share personal, emotive and intimate life experiences. By embracing the feminist methodological approach, dismantling the traditional hierarchical nature of the interview and replacing it with a model both empowering for the researcher and the researched, my aim was to enable both trust and a dialectical relationship to develop within the research process.

Methodological Dangers

4.1
While this research has been influenced by the feminist methodological challenge, I also acknowledged the dangers of methodological separatism. There has been much discussion over whether there is or should be a specifically 'feminist' approach to doing social research (Ramazanoglu, 1989; Hammersley, 1994) and the exclusionary potential of using differing 'male' and 'feminist' methods. What is clear is that sole reliance on a feminist methodological approach and concurrent emphasis on qualitative methodology, may, albeit unwittingly, succeed in buying into the very paradox it is protesting about. Rather than contributing to a more 'human' understanding of the social world, the feminist challenge may reinforce existing dichotomous choices and fuel existing ideological combat between quantitative and qualitative methods (Oakley, 1998). In agreement with Maynard (1998: 129), 'the negative characteristics attributed to quantitative research are not necessarily intrinsic to them'. In consideration of these criticisms, I chose to be more eclectic in my approach. The research methods used in this study utilised a multi-method approach, embodying both empirical and interpretative elements.

Initial Forays into the Field: Lessons Learned

5.1
Initial intentions had been firstly, to access an Assisted Conception Unit both known to me, and which knew me. Secondly, I aimed to explore the potential of the Internet in conjunction with research in a more conventional setting. Reasons for this are as follows: -

5.2
Firstly, experience and knowledge of the unit, staff, procedures and surroundings would prove beneficial, particularly when negotiating access. Secondly, familiarity within the setting would provide reassurance to respondents and add some degree of commonality. In addition, self-disclosure, or reciprocity, would not only help overcome respondents' potential inhibitions but would also place the interactions on a more equal footing, promoting a 'true dialogue' rather than the interrogatory style of the traditional interview setting (Cook and Fonow, 1984). In researching such a sensitive field, I was, as Smith (1988) advocates, starting from subjective experience and therefore shared common ground.

5.3
Finally, post-graduate coursework had generated my interest in the potential of computer-mediated research. Technological advance in this field offered exciting and challenging new research opportunities while impacting on the more conventional research paradigms. Despite some limitations and initial reservations, the Internet appeared to provide the opportunity to forge closer links between the researcher and the respondent. Further, this offered the possibility for the respondent to verify and become a part of the research process itself, points in keeping with the feminist methodological stance informing this study. For many vulnerable groups, such as the infertile, the Internet has become a primary source of information and social support (Sharf, 1999). I was keen to explore this field, its potential and the extent to which online interaction further problematised or enhanced research interaction. As Holmes (1997) suggests, the perceived advantage of virtual reality and CMC lies in its ability to simulate the 'real' world without any of its limitations, thus, potentially, leaving the individual 'free' to create.

The Visual Terrain: First Contacts and Field Dangers

6.1
Preparatory contact was arranged with the Senior Consultant and head of the Assisted Conception Unit, with a view to explaining the nature of my research in this field and negotiating access. After a brief resume of my research proposal and several probing questions from the consultant concerning my educational background, academic status (including degree classifications) and 'suitability' for research in this field, I pointed our conversation to the effects of the experience of infertility, particularly in relation to peer and social pressures, familial expectation and the exclusionary potential of involuntary childlessness. In short, this was met with derision, particularly the suggestion that those experiencing involuntary childlessness may also experience a degree of stigma in a predominately family-orientated society. According to this consultant, the fact that those undergoing treatment were willing to allow up to six of his medical students to view acutely personal procedures, suggested that they were in no way embarrassed or stigmatised. I felt that there may be different explanations for this but, wary of my delicate situation in terms of access, took a mental note to explore this further once in the field.

6.2
Significantly - and applicable to the development of the feminist research project in this field - access difficulties highlighted the significance of power relations in the conventional setting and the effects of these on both researcher and researched. Firstly, it became clear the consultant involved exercised a significant and intractable degree of control over the research site, making it clear that this degree of control would necessarily extend to cover research design, interaction, process and findings. I was requested to submit a lengthy research proposal including a detailed account of expected research findings - a problematic issue given the nature of my research and methodology which informed it. The consultant would then 'vet' my research design. Subject to approval, I would be conditionally accepted for access and passed to the hospital Ethics Committee for official approval, a process that would take four months. While hastily planning my research schedule and ways to bypass these obstacles, the four-month delay was effectively to end chances of access, breaching already tight deadlines. Secondly, the boundary between my professional and medical relationship with this consultant had been (deliberately) blurred. Much of this interview was taken up with an (involuntary) analysis of my own medical situation thereby problematising my own research relationship with both the consultant and the unit. Finally, and as this paper will explore, the effects of power relations in the setting also problematised the participation of respondents.

6.3
I have discussed this meeting at length to highlight important issues that can be easily overlooked when planning research projects - particularly the assumption that a privileged position in the field equates with unproblematic access. The full dimensions of gaining access to the research field may only become apparent after the event. If the infertile, for example, are perceived as deviant, then research by them and on them may be considered dangerous and in need of management. In this study, while appearing to welcome research in this field, the preparatory meeting and obstacles placed in my path effectively ended all hopes of access. More significantly, the degree of control exercised over the setting may have challenged the validity of the data itself and subsequent knowledge generated.

Computer Mediated Communication: The Internet as a Research Tool

7.1
Initial research design, aims and objectives thus required reformulation. My subsequently imposed reliance on the Internet as the primary tool for this research highlighted important issues. Firstly, online participation offered personal anonymity in a very emotive field. Secondly, because of the sensitive nature of this research, a number of respondents emphasised their reluctance to participate had this research been conducted in a more conventional, face-to-face setting. Finally, respondents also highlighted the detrimental effects of power relations in this field through an unwillingness to participate in a face-to-face setting but also through a reluctance to participate in research conducted in a clinical setting, particularly within a designated assisted conception unit. Many voiced fears (albeit unsubstantiated) of a reduction in funding and a reluctance to promote images of dissatisfaction during treatment. In this respect, online research afforded greater insight and offered respondents greater freedom - a factor that may have remained hidden and thus potentially damaging to the validity of my original research plans, as I have suggested. Further, had these plans progressed unhindered, I may, unwittingly, have contributed to an enhancement of power relations within the research setting. This is a point at odds with the feminist methodological approach and underlines the potential of the virtual realm. If one of the primary aims of the feminist research project is to uncover the invisible voice of women and reflect women's lived experience, the possibilities of CMC must be embraced more fully.

7.2
The following section provides an overview of the development of Internet research and an appraisal of some of the methodological difficulties and limitations encountered. While the methodological difficulties are similar to those experienced in the more conventional research domain, they are more acute within this new field. An in-depth exploration and understanding of these issues is a vital precautionary measure before undertaking this type of research.

The Virtual Realm

8.1
The rapid development of the World Wide Web has lifted the restrictions of geographical boundaries and opened new research horizons. The research community has access to a new frontier and tool for data collection (Cornick, 1995; Sproull, 1996). In effect, the Internet provides the research community with the chance to interface with respondents in ways which may overcome some of the barriers imposed by conventional research approaches (Coomber, 1997; Chen and Hinton, 1999; O'Connor and Madge, 2001). The researcher can now engage in research on a world-wide, low cost, almost instantaneous scale and, according to more favourable reports, reduce the impact of instrumental biases such as the race or gender of the investigator. The Internet provides a medium whereby the researcher has access to a world of behaviour and ideas - in the form of emails, bulletin boards, discussion forums and chat rooms - through which individuals may share their innermost thoughts and feelings. The role of the Internet in the formation of the virtual community has already been well documented and explored (Jones, 1995; 1997; 1998; Hine, 2000) and, as Hine (2000) suggests, earlier negative understanding of CMC as an impoverished medium has been transcended. The 'impoverished' viewpoint has now been challenged alongside the notion that the Internet could provide a far richer form of interaction than previously envisaged. The more optimistic view of this new research field suggests that the Internet appears to provide the opportunity to concentrate on specific areas of interest, ask the right questions, talk to enough people, read the readily transcribed answers and develop and identify enough patterns until the researcher 'knows'. The researcher no longer has to make empirical assumptions and infer from samples to the wider population. Research thus arguably deals less with inference and more with actuality and reality (Cornick, 1995).

The 'Captive Audience' Effect

9.1
The Internet provides a highly efficient means of identifying potential respondents. One of the most effective means of recruitment is the use of discussion groups. Effectively, these are forums where groups of individuals discuss particular topics of common interest or common experience. The key to these groups is that they are there to exchange information and ideas about well-defined topics and, as such, provide a rich vein of information (Taylor, 1998).

9.2
I located and subscribed to two such discussion groups dealing specifically with infertility, carefully wording my recruitment message to avoid giving the impression of a nave and overly inquisitive 'outsider' and highlighting both my interest in and experience of the field. Initial responses were poor. A subsequent email received from a user advised that the research sites chosen were relatively under-used. My recruitment message had been passed on to a more active discussion group and bulletin board for a limited period.

9.3
The response to this second-phase recruitment was encouragingly high. At the end of the limited period (seven days), I had received emails from 65 potential respondents. In general, responses to online recruitment requests are higher than that expected using more traditional procedures (Taylor: 1998) and may be due to the fact that respondents participating in discussion groups are already interested and committed to discussion in the topic area. However, concerns that the sensitive nature of this study would deter potential respondents were soon dispelled. Initial distribution of questionnaires generated a 70% response rate. Of these, a further 71% agreed to participate in semi-structured, email interviews. While the high response rate may be a product of the research topic itself, it is also indicative of the 'captive audience' effect.

9.4
Far from being problematic, the physical distance between researcher and respondent appeared liberating and an attraction to respondents. Again, while this may be an issue indicative of the research focus (illustrated by respondent's comments highlighting a reluctance to participate within a conventional research setting), a point of generalisation may still be made. The increased physical distance between researcher and respondent potentially further dilutes the effects of power relations within the setting - a point of central concern to the feminist methodological critique.

9.5
Yet, while this new paradigm undoubtedly offers exiting research opportunities, the somewhat idealistic view of its potential presented above hides methodological and ethical difficulties that have become apparent throughout this research. Further, if these difficulties are left unchecked, they may serve to undermine the use of the Internet as a tool for social research. There is a temptation to go headfirst into this new data terrain with little forethought. Online research may offer new possibilities - and, in this case, opened the door to a previously hidden world of experience unrestricted by geographical location - but it also presented many limitations and disadvantages. This point is particularly emphasised to challenge the misconception that online research is an 'easy option'.

Cyberspace or Connected Space?

10.1
There is an overwhelming body of literature exploring aspects of qualitative interviewing within the conventional research spectrum (Oakley, 1981; Burgess, 1984; Renzetti and Lee, 1993; Opie, 1981; Kvale, 1996). Yet the extent to which, if at all, existing methodologies (and their implicit ontological and epistemological assumptions) transfer to an online setting is one that warrants further exploration and has much to do with how the online setting is both perceived and understood.

10.2
The term 'cyberspace' is most commonly used to denote a 'field of interaction existing in an independent reality, separate from offline environments, bodies and concerns' reflecting the 'commonly held view that correspondents have no bodies, faces or histories' (Kendall, 1999: 60). Turkle (1995) also suggests this necessitates understanding the body as multiple - as multiple personas and as varying presentations of the self. On the surface, this, perhaps, would appear to support the need for a developed understanding of a 'new' virtual ontology and reinforce claims for a specific Internet methodology (Sudweeks & Simoff, 1999). However, while I agree communication and interaction via the Internet problematises existing methodological assumptions and the tools of the research trade, the extent to which this also warrants a specific methodology is open to doubt. However, I am not suggesting virtual reality can be viewed as simply another form of communication or interaction. It is more powerful than this and may, at time, assist users to transfer or alter their sense of self (Bromberg, 1996). Yet, as Kendall (1999) argues, identity may be fluid and potentially multiple on the Internet but people similarly engage in these practices - the different presentations of the self to different audiences - in other arenas of everyday life and did so prior to the existence of the online forum. There is evidence, as in Kendall's study, that people tend to perceive their identities and selves as integral and continuous. Contrary to earlier thought, there is now a developing body of opinion that suggests that the differences between self-presentation in real life and online is far less divergent than might be expected (Mann and Stewart, 2000: 210). This arguably misconceived presumption may also be the result of the continuous use of the term 'cyberspace' itself - a conceptualisation which presupposes an 'independent reality' (Kendall, 1999). If this were replaced by the idea of a 'connected space', this would allow recognition of the fact that individuals exist online and offline simultaneously (Kendall, 1999: 60). Further, this allows for the exploration of how participants may blend the two and the possible effects of offline context on participants' online experience. Sveilich's 1995 study on inflammatory bowel disease provides a useful example of this. Sveilich argued that sufferers experience feelings of isolation and confusion upon diagnosis, compounded by lack of energy - a symptom which prohibited participation in a more conventional and face-to-face support group and encouraged sufferers to seek support online - underlining the need to acknowledge the offline context of online interaction.

Reflections: Democratisation of Exchange

11.1
That virtual (disembodied) participation is an ideal medium for conducting unbiased research, fostering a 'democratisation of exchange' (Selwyn & Robson, 1999) and non-coercive, anti- hierarchical dialogue - is one which is promoted by many 'cues filtered out' theorists and particularly Spender (1995). This perception is particularly relevant to the feminist research project. Yet the potential for diffused power relations emphasising the apparently egalitarian nature of CMC cannot be presumed (Kendall, 1999; Mann & Stewart, 2000). While CMC may eliminate the visual social cues which can inform power relations in the face-to-face setting (Mann and Stewart, 2000: 162), anonymity and the lack of visual cues online does not necessarily foster an equalisation of power relations in the setting. The use of email interviewing in this study alleviated some of the interpersonal problems commonly associated with conventional interviewing techniques, yet this technique was far from unproblematic. As will be discussed, the primary advantages this type of interaction promoted obscured both weaknesses in validity and interaction problems arising from the lack of conventional face to face dynamics and observational and non-verbal cues.

Reflections: Demographic Problematics and Technical Restrictions

12.1
On a practical level, this new medium brings with it more acute methodological challenges in terms of its demographic composition and technical restrictions (Taylor, 1998). Respondents will also be users of advanced information technology - a factor which obviously precludes representation of those who do not have access to the Internet and which makes generalising about research findings from Internet users to the general population highly problematic. This also problematises Cornick's (1995) suggestion that research will deal less with inference and more with actuality and reality. Further, these restrictions may also serve to reinforce existing social divisions between the 'information rich' - those who have access to both computers and online facilities - and the 'information poor' - those who have no, or limited access (Turkle, 1985). This may be more or less problematic dependent on the nature of the research itself. For example, sample bias may be of less concern to a researcher when they are interested in particular types of behaviour, as distinct from research which looks for more representative or generaliseable behaviour (Coomber, 1997).

12.2
To illustrate this, the dominant group of respondents in this research were those employed full-time - 69.1%. This may be a reflection of, firstly, the representational problems mentioned and, secondly, the cost of treatment. 65.2% self financed treatment by necessity. Further exploration of more quantitative data revealed 35.0% of respondents envisaged their final treatment costs to range between 2000- 10000. This may suggest only those financially well equipped can afford the cost of treatment, which is an issue in itself but this also further excluded those less financially able and problematised the demographics of those taking part in the study.

12.3
Further technical and logistical problems were encountered when respondents had out of date, incompatible or more up to date software. For example, the questionnaires were forwarded, in Word format, as an email file attachment. A number of respondents either did not have the required software to open the attachment or did not possess the required technical competence to open an attachment, save, complete and return it. The presumption here is that many people using the Internet are skilled or able users. In this respect, continued research participation relied heavily on the goodwill of respondents as questionnaires were reformatted and forwarded with detailed user-instructions. Again, the superficial advantages of speed and instantaneous communication must be weighed up against these more practical considerations.

Reflections: Validity, Interaction, Discourse and Rapport

13.1
The primary advantages this type of interaction promoted obscured weaknesses in validity and interaction problems. The advantages of speed, 'friendliness' and instantaneous dialogue combined with research unrestricted by geographical location and transcription costs, masks a number of issues which affect the research relationship. This research utilised email communication as it placed less reliance on the good will of the respondent in terms of supporting and installing conferencing packages and increased privacy within a sensitive field. A small-scale pilot study prior to this research had already underlined the importance of both privacy and anonymity to respondents.

13.2
While an important advantage of the virtual interview is a tendency to be less self-conscious and embarrassed (Nguyen and Alexander, 1996), I was also concerned that a conferencing or chat-room format would allow little room to negotiate the sensitive nature of this research topic and establish rapport on an individual basis. This could further jeopardise attempts to moderate discussion within the setting itself and could act as a potential deterrent for some respondents. Returned questionnaires had already indicated that many respondents, although supportive of chat and discussion forums in general, felt reluctant to participate in research which used such an 'open' format. Again, this was possibly due to the nature of the research itself and the need to discuss sometimes personal and intimate experiences.

Reflections: The Virtual Interview as a Disembodied Experience

14.1
Email interaction is not comparable to personal interaction. The disadvantages outlined below in transferring researcher- respondent relations to the computer-mediated setting must be explored further and remain to the forefront when decisions are made to utilise the Internet as a resource. Difficulties emerged which influenced both the researcher-researched relationship and the emotional labour of the researcher conducting the research.

14.2
The nuance involved in face to face communication is lost - the researcher cannot 'see' a smile or a sigh. To illustrate this, one interviewee used what, at the first point of contact, appeared to be particularly reactionary and hostile language:

'People shouldn't get funding if they don't do something for it ... if they didn't drink, smoke, go on holidays they could pay for their own treatment. They don't deserve it. We've had to do without - no holidays and no extras. I work at nights!' (Extract: 1)

14.3
The lack of physicality and reliance on written rather than verbal communication potentially leads to the loss of important observational elements and cues vital to the validation of researcher-respondent exchange. Crucially, the concept of validity goes further than communication or dialogue alone and must necessarily include observation and interpretation. Yet it is precisely this aspect which on-line interaction excludes and, further, underlines one of the primary benefits of the more conventional, face-to-face interview.

14.4
To overcome these difficulties second and sometimes third stage interviews were conducted to help re-affirm the content of the original interviews and avoid the potentially detrimental effects, mentioned above. Referring to Extract (1), subsequent re-interviewing combined with a more overt concern with textual analysis and offline context, revealed this language to be one not of hostility but one borne out of frustration, emotion and desire. The respondent concerned had been refused funding for treatment and, after primary interviews were completed, apologised for her communication, explaining she had been extremely upset at the time.

14.5
The tendency to become more open, less self-conscious and less embarrassed (Nguyen and Alexander, 1996) may attract some respondents to a research project when they would previously have refused to consider participation. This may produce richer and more detailed data but without the physical setting, the content and context of speech becomes problematic, increasing the workload of the researcher and highlighting the tendency of CMC to change discursive rules and procedures. Further, this questions the reliability and validity of data obtained. The advantage of readily transcribed transcripts, of speed and almost instantaneous access does not outweigh the effects of context and setting. Alongside this, external distractions may interrupt the flow of the interview, of which the interviewer may be unaware. This may result in temporary or prolonged disengagement with the interview itself, as O'Connor and Madge (2001) have highlighted.

14.6
The virtual interview, in effect, is a disembodied experience, lacking in visual and verbal clues and increasing the need to establish rapport. This problematises Spender's (1995) suggestion that lack of physicality in the setting reduces potential instrumental bias such as age, race, gender, ethnicity and sexuality. During my research, I offered personal socio- demographic information and used self-disclosure - techniques I suggest are necessary pre-requisites to establishing rapport in this setting. This was a method reciprocated by respondents. The technique was effective but, again, was time-consuming and demanding - and reflected an increased tendency for both researcher and respondent to 'stamp' their identity on communication. As my research was informed by a feminist methodological stance, which aimed to further dismantle the barriers between the researcher and the researched, this increased tendency to 'stamp identity' on communications potentially provided another route for instrumental bias to enter the setting and is a point which warrants further exploration.

Reflections: Skills, Power Relations and Cultural Competence

15.1
The level of skill required by the 'electronic' researcher arguably exceeds that required within the conventional research field. The skills range from the practical abilities concerning typing efficacy, speed and the ability to moderate and re-direct conversations and interviews with a keyboard, to the ability to establish rapport and trust between individuals linked only by a computer interface. The researcher cannot simply 'step' into a virtual community and must establish commonality and trust through the computer interface alone. Fundamental to this is the need to be culturally competent both within the research setting itself and with the technology being employed. In this respect, it is crucial to avoid fostering the impression of the 'nave outsider'. This research necessitated establishing contact with a support network with which I shared common experience - both in terms of a developed interest in the use of the Internet as a support site and through my own experience of the field topic.

15.2
However, while this prepared the ground for trust and rapport to develop, issues arose which challenged the researcher-researched relationship within the setting. During interviews, several respondents questioned my own situation and sought opinions on certain issues, effectively changing the power relations of the interview. While I welcomed this in terms of my methodological approach emphasising the advantages of shared experience, moderating the interview and maintaining focus, while simultaneously using the computer as an interface, became problematic and required a developed level of interaction management and manoeuvrability. Linked to this was respondent perception of my role as a medical expert in the field. While, on the one hand, email interviewing served to equalise the potentially hierarchical nature of the interview situation, the setting itself contributed to the development of alternative power relations. A number of respondents asked for specific medical information and advice - effectively adding a 'white-coat' to my research role. On hindsight, I begin to wonder how much their perception of me as a 'white-coated professional' affected their online interaction and subsequent data obtained.

Reflections: Leaving the Setting and Gaining Closure

16.1
Leaving the setting ceases to become an act of physical separation within this field. While on the one hand, the respondent need only press a button to leave the setting, on the other hand, the familiarity developed and, in particular, the ease of contact which email communication affords is, at times, difficult to relinquish. This quickly became apparent during this research - albeit a situation possibly enhanced by its sensitive nature.

16.2
Many of the respondents welcomed and were encouraged by the anonymity and privacy of email communication. While this proved to be a comforting and empowering dimension for respondents, it proved problematic for my role as researcher - the corollary to the greater trust and intimacy which was aimed for using this methodological approach. Many informants wished to prolong the research encounter, finding it difficult to relinquish their participatory role. Again, this may be an issue commonly experienced within the more conventional research field. However, I would suggest, again, the problem, in an online setting, is more acute. Many respondents maintained email contact out with the research schedule. More than a year on from this research, I still receive emails from respondents. Effectively, this results in the failure to achieve closure for both researcher and respondent and is a difficulty to which there is no easy solution. As a suggestion to other researchers in the field, I gradually increased the length of time taken to respond to emails. As time moves on, email interaction is now only periodic. In this respect, this highlighted the benefits of the conventional face-to-face domain and the concurrent greater ease with which 'closure' can be attained. Linked to this are the difficulties experienced leaving discussion groups. This is an issue further problematised by my unwillingness to foster the image of the 'exploitative researcher' by quickly leaving the group once suitable data had been gathered - potentially 'muddying the field' for future research. Likewise, this issue must be carefully considered. As research using the Internet increases, suspicion towards those utilising internet sites as fields of rich data, is increasing - a point underlined by my recent perusal of several well-known sites. While the Internet appears to offer new horizons and open new frontiers for the research community (Cornick, 1995), the nature of the medium itself suggests that these 'frontiers' may just as quickly begin to deny access.

Reflections: Hidden Populations

17.1
The initial potential of the Internet and the online community must be measured against its limitations, a point particularly applicable when targeting sensitive research areas. As Coomber (1997) argues, the use and effectiveness of online research is dependent on who and what is being researched. The lack of face to face dynamics and the lack of observational and non-verbal cues alone would suggest the demands placed on the researcher - although similar to those experienced in the conventional field - require developed familiarity with the terrain and an accomplished level of skill.

17.2
As I have outlined, the Internet presents itself as a field of rich pickings. In particular, there may be significant research benefits to be gleaned where the group being researched is normally difficult to reach and/or the issues being researched are of a sensitive nature (Coomber, 1997), as illustrated by the following quote:

'I wouldn't normally take part in something like this ... it's really personal ... I don't like doing things like this face to face. But there are things I'd like to talk about but I wouldn't feel comfortable at the clinic ... I'd feel like I was complaining.' (Extract: 2)

17.3
Yet, the researcher should, as I have emphasised throughout this paper, consider the wider ramifications of conducting research using the Internet. The objects of research within the conventional research paradigm are predominately members of socially disadvantaged or marginalised groups (Hood et al, 1999). The Internet, by its very nature, provides the medium for increasing research on these groups and promotes an easily accessible image. The tendency for members of vulnerable, disadvantaged or marginalised populations to participate online is high (Coomber, 1997; Jones, 1997) and, in some respects, the Internet can be regarded as a primary source of information and social support. The researcher must, therefore, consider the implications and justify accessing respondents online. Behind what may, at first, appear to be a field of rich pickings lies a minefield of complexity and difficulty.

Conclusion: A Cautionary perspective

18.1
While the use of the Internet as a research tool posed significant methodological dilemmas, its use, in this study, can be justified. The difficulties encountered gaining access, and the implications for the reliability of data gained from the more conventional setting, reinforced my decision to utilise CMC as an alternative form of communication. For many respondents, interchange via the computer has become a primary source of information and social support. The high rate of response to my research call and subsequent interest in more prolonged interviews and continuation of this research is testament to this. This is an area where anonymity may feel comforting and where feelings and experiences may be discussed which would be withheld in other interactive contexts - a point confirmed by several respondents in this study. Many respondents welcomed the empowering dimension and the recognition that this research provided. The 'invisibility' of women's experience in this field was lifted, if briefly, as they became the central concern of this research process - a point that both reinforces and justifies my decision to continue research in this field. The electronic environment is an effective tool which, in this context, presented a more neutral and egalitarian space in which participants could communicate. As I have suggested, if one of the primary aims of the feminist research project is to uncover the invisible voice of women and reflect women's lived experience, the possibilities afforded by CMC must be embraced more fully. Yet, communication via the virtual realm is not necessarily applicable to all research projects - an idealised view which, as I have suggested, must be countered. The use of the technology itself does not necessarily entail self-disclosure. The effectiveness of CMC, particularly when applied to the feminist project is much dependent on the context, content and purpose of the research itself. A point for the feminist research project to explore further is the suggestion that taking the body out of the setting results in an equalising corrective to interaction in the field - a view which presumes an individualistic and Cartesian view of the subject. What seems clear is that the effectiveness of CMC is much dependent on who is being researched, what is being researched and why.

18.2
This research has shown that the transference of conventional research procedures and devices to the online setting is problematic and requires careful consideration. I have centred my reservations here on issues ranging from the more practical considerations concerning technical access and restrictions and the skills required to interface online, to sensitivity, validity, interaction, discourse and power relations within the setting.

18.3
As Jones (1999) has suggested, at present we only have the methodological tools available to us to which we are accustomed. The feminist methodological approach is, arguably, more effective when researching in an online setting, allowing for a developed degree of interaction, reflexivity, disclosure and trust to develop within the research relationship. As this research has indicated, researcher and respondent self-disclosure was crucial to the formation and maintenance of meaningful relationships with respondents. While a degree of this may be the product of the research focus, the benefits of the anonymity of the online setting must also be highlighted. Yet the dangers inherent in simply transferring existing methodological tools online cannot be understated. Effective computer mediated research may be that which recognises the potential effects of the offline context upon online interaction. The online setting thus becomes one site among many others in everyday life. This is a potential resolution to some of these difficulties and which is part of my ongoing research in this field.

18.4
However, it is to one of these issues that I now turn for final comment. Debate concerning the value of computer mediated research, its methods, technical restrictions and concurrent problematics, is proving a popular terrain within the research community (Coomber, 1997; Selwyn and Robson, 1999; Chen and Hinton, 1999; Jones, 1995; 1997; 1998; 1999; O'Connor and Madge, 2001). Nevertheless, many of the issues raised may be more of a problem of the present than of the future. While this type of commentary must retain a prominent presence in the field, methodological critique must also move on to examine the more sinister potential and ethical implications of this new terrain. What has been neglected so far is a need for a developed awareness of ethical implications of online access to hidden, excluded, marginalised and disadvantaged groups. I have presented this paper in the form of a research trajectory, outlining the course of my research from its conception to its completion. The intention behind this is to problematise the idealised view of the value of the Internet as a research tool and encourage a more developed focus on the justification, applicability and benefits of Internet research for the particular project. For the purposes of this research, the virtual realm afforded greater insight and offered respondents greater freedom, thus enhancing the reliability and validity of this research.

18.5
As Hine (2000) suggests the previously negative understanding of CMC has been transcended. The Internet is an umbrella for many sites reflecting rich and sustained interaction and the possibilities for social research must not be understated. However, this research has highlighted that using the Internet and CMC as a tool for research raises issues that are in need of further consideration and exploration. It is in this direction that critique in this field should be guided and underlines the dictum that, as researchers, we are, as ever, in positions of responsibility towards both the research community and the respondents themselves.

References

BROMBERG, H (1996) 'Are MUDs Communities? Identity, Belonging and Consciousness in Virtual Worlds' in Shields R (editor) (1996) Cultures of Internet. Virtual Spaces, Real Histories, Living Bodies. London, Sage.

BURGESS, R (1984) In the Field: An Introduction to Field Research London, Allen and Unwin.

CHEN, P and HINTON, S.M. (1999) 'Realtime Interviewing Using the World Wide Web' Sociological Research Online Vol. 4, No. 3 < http://www.socresonline.org.uk/4/3/chen.html>

COOK, J and FONOW, M (1984) 'Knowledge and women's interests: Issues of epistemology and methodology in feminist sociological research' Sociological Inquiry, 56, 2-29.

COOMBER, R (1997) 'Using the Internet for Survey Research' in Sociological Research Online Vol. 2, No2 <http ://www.socresonline.org.uk/socresonline/2/2/2.html>

COREA, G (1985) Man-Made Women: How new reproductive technologies affect women London, Hutchison.

CORNICK, D. L. (1995) 'Cyberspace: Its impact on the Conventional Way of Doing and Thinking about Research' <http://www.csaf.org/cyber.ht m>

GIMENEZ, M. E. (1991) 'The Mode of Reproduction in Transition - A Marxist-Feminist Analysis of the Effects of Reproductive Technologies' <& lt;http://csf.colorado.edu/authors/Gimenez.Martha/nrts.html>>

HAMMERSLEY, M (1994) 'On feminist methodology: a response' in Sociology 28(1): 293- 300.

HARAWAY, D (1991) Symians, Cyborgs and Women New York, Routledge.

HARCOURT, W. (editor) (1999) Women @ internet. Creating new cultures in cyberspace London, Zed Books.

HARDING, S (1986) The Science Question in Feminism Milton Keynes, Open University Press.

HINE, C (2000) Virtual Ethnography London, Sage.

HOLMES, D. (editor) (1997) Virtual Politics. Identity and Community in Cyberspace London, Sage.

HOOD, S, MAYALL, B & OLIVER, S. (editors) (1999) Critical Issues in Social Research: Power and Prejudice Buckingham, OU Press.

ILLINGWORTH, N.A. (2000) 'Motherhood Meets Science: Salvation or the Enemy? An Analysis of the feminist critique of Assisted Reproductive Technologies'. Unpublished MSc Thesis, University of Stirling, Stirling.

JONES, S. G. (1995) Cybersociety: Computer-Mediated Communication and Community London, Sage.

JONES, S. G. (editor) (1997) Virtual Culture: identity and communication in cybersociety London, Sage.

JONES, S. G. (editor) (1998) Cybersociety 2.0: Revisiting Computer-Mediated Communication and Community Thousand Oakes, CA: Sage.

JONES, S. G. (editor) (1999) Doing Internet Research. Critical Issues and Methods for Examining the Net London, Sage.

KENDALL, L (1999) 'Recontextualising "Cyberspace": Methodological Considerations for On-Line Research' in Jones S (editor) (1999) Doing Internet Research. Critical Issues and Methods for Examining the Net London, Sage.

KVALE, S (1996) Inter-Views: An Introduction to Qualitative Research Interviewing London, Sage.

MANN, C and STEWART, F (2000) Internet Communication and Qualitative Research: A Handbook for Researching Online London, Sage.

MAYNARD, M (1998) 'Feminists' knowledge and the knowledge of feminisms: epistemology, theory, methodology and method' in May, T & Williams, M. (editors) (1998) Knowing the Social World Buckingham, Open University Press.

NGUYEN, D. T. & ALEXANDER, J (1996) 'The coming of Cyberspacetime and the End of Polity' in Shields, R. (editor) Cultures of Internet: Virtual Spaces, Real Histories, Living Bodies London, Sage.

OAKLEY, A (1981) 'Interviewing Women: a contradiction in terms' in Roberts, H. (editor) 1981 Doing Feminist Research London, Routledge.

OAKLEY, A (1993) Essays on women, medicine and health Edinburgh, Edinburgh University Press.

OAKLEY, A (1998) 'Gender, Methodology and People's Way of Knowing: Some Problems with Feminism and the Paradigm Debate in Social Science' in Sociology Volume 32, No. 4.

O'CONNOR, H & MADGE, C, (2001) 'Cyber-Mothers: Online Synchronous Interviewing Using Conference Software' Sociological Research Online, Vol.5, No.4 <http://ww w.socresonline.org.uk/5/4/o'connor.html>

OPIE, A (1981) 'Qualitative research, appropriation of the 'other' and empowerment' in Feminist Review 40: 52-69.

RAMAZANOGLU, C (1989) 'Improving on sociology: the problems of taking a feminist standpoint' in Sociology 23(3): 427-42.

RENZETTI, C.M. & LEE, R. M. (editors) (1993) Researching Sensitive Topics Newbury Park, Sage.

ROSE, H (1987) 'Victorian Values in the Test-Tube: The Politics of Reproductive Science and Technology' in Stanworth, M. (editor) (1987) Reproductive Technologies: Gender, Motherhood and Medicine Cambridge, Polity Press.

SELWYN, N and ROBSON, K (1998) 'Using email as a research tool' Social Research Update Issue 21, University of Surrey <http://www.soc .surrey.ac.uk/sru/SRU21.html>

SHARF, B. F. (1999) 'Beyond Netiquette. The Ethics of Doing Naturalistic Discourse Research on the Internet' in Jones, S.G. (editor) (1999) Doing Internet Research. Critical Issues and Methods for Examining the Net London, Sage pp243-256.

SMART, C (1992) Regulating Womanhood London, Routledge.

SMITH, D (1988) The Everyday World as Problematic: A Feminist Sociology Milton Keynes, OU Press.

SPEARS, R. & LEA, M. (1994) 'Panacea or Panopticon? The Hidden Power in Computer- Mediated Communication' in Communication Research, Vol.21 No.4, August 1994, pp427-459.

SPENDER, D (1995) Nattering on the Net. Women, Power and Cyberspace. Melbourne, Spinifex Press.

SPROULL, L. S. (1996) 'Using Electronic mail for data collection in organisational research' in Academy of Management Journal Vol. 29, No1, March 1996.

STANLEY, L & WISE, S (1990) 'Methods, methodology and epistemology in feminist research process' in Stanley, L (editor) (1990) Feminist Praxis London, Routledge. STANWORTH, M. (editor) (1987) Reproductive Technologies: Gender, Motherhood and Medicine Cambridge, Polity Press.

STEWART, C. M., SHIELDS, S. F., MONOLESCU, D, TAYLOR, J. C. (1999) 'Gender and Participation in Synchronous CMC: An IRC Case Study' in Interpersonal Computing and Technology: An Electronic Journal for the 21st Century October 1999, Volume 7, Number 1-2 <http://www.emoderators.com/ipct-j/1999/n1- 2/stewart.html>

SUDWEEKS, F & SIMOFF, S (1999) 'Complementary explorative data analysis: The reconciliation of quantitative and qualitative principles' in Jones, S.G. (1999) Doing Internet Research. Critical Issues and Methods for Examining the Net London, Sage pp29-55.

SVEILICH, C (1995) as cited in Bloor, M, Frankland, J, Thomas, M, Robson, K (2001) Focus Groups in Social Research London, Sage p78.

TAYLOR, S (1998) 'On-line Research. What can and will be done' Paper presented at Social Research Association Research on the Internet Study Day. October 1998, London.

TURKLE, S (1985) as cited in Kenway, J (1996) 'The Information Superhighway and Post- modernity: the social promise and the social price' in Comparative Education Vol. 32, No2 1999 pp217-231.

TURKLE, S (1995) Life on the Screen: Identity in the Age of the Internet London, Weidenfeld and Nicolson.

WACJMAN, J (1991) Feminism Confronts Technology Pennsylvania: The Pennsylvania University Press.

Copyright Sociological Research Online, 2001