Undertaking Sensitive Research: Issues and Strategies for Meeting the Safety Needs of All Participants

Abstract: There are many phenomena that within specific cultural and social context are "sensitive". They may be defined as "sensitive" if they are private, stressful or sacred, and discussion tends to generate an emotional response, for example death and sex. Phenomena that deal with potential fear of stigmatisation, such as the study of sub-cultures, and studies that may reveal information of a politically sensitive nature may also be considered "sensitive". In response to the "sensitive" nature of such phenomena Ethics Committees act as gatekeepers during the research process to protect individuals and/or groups who form the sample from harm. Experience and a review of current literature clearly indicates that these are not the only participants affected by the research. The researchers, transcribers, supervisors and readers of publications may also be placed at risk. This risk may be physical and/or psychological. In order to protect all participants' physical and psychological safety protocols or guidelines need to be developed at the beginning of the research process to identify and minimise risk, or respond to risk as they arise during the research process.

Key words: sensitive research, risk, transcriber, researcher, supervision, safety protocols

Social research investigations often involve a consideration of issues, data and perspectives that may impact on the feelings, views, attitudes and values held by people involved in the research process. Such considerations may be investigated from a methodological perspective or they may be pragmatically, i.e., how they affect people involved in the research. Methodological considerations have b/een regarded widely in writings on quality, validity and reliability in qualitative research. However, much less has been mentioned about the pragmatic issues and concerns that qualitative researchers experience doing their research. Qualitative researchers in social fields, health areas and the humanities, at one time or another, have raised questions about the sensitivity of their research. This article starts to address the concept of "sensitivity" in qualitative research by addressing a range of aspects in doing qualitative research, discussing the nature of sensitivity in each of them, and offering some suggestions for strategies to address these sensitivities. [1]

Over the past few decades there has been increasing research interest in "sensitive" social issues such as sexuality, child abuse and death, that are emotion laden or inspire feelings of dread or awe (FARBEROW 1963). In addition, as the community has become aware of the impacts of domestic violence, partner abuse and societal/family breakdown attempts have been made to investigate the experiences of people in these situations. The aim has been to gain increased understanding and awareness of the impact of the experience of "sensitive" issues on people's lives. [2]

The definition of a "sensitive" research topic is dependent on both context and cultural norms and values. Raymond LEE (1993) suggests there are three issues that create a concern about sensitivity. The first of these issues are those considered private, stressful, or sacred, such as sexuality or death. The second are those issues that if revealed might cause stigmatisation or fear, such as youth studies that reveal illegal behaviour. The final issues are related to the presence of a political threat where researchers may study areas subject to controversy or social conflict. The sensitive nature of the research may not be apparent at the beginning of the research project, alternatively a subject that was presumed to be of a sensitive nature may not be. For example, although abuse is considered by many in Western societies to be a "private" and "sensitive" issue, women report being relieved to be able to talk freely about their experiences (WALKER 1979, 1984; McCOSKER 1995). Further to these considerations Joan SIEBER and Barbara STANLEY (1988, p.49) suggest that sensitive studies are those "... in which there are potential consequences or implications, either directly for the participants in the research or for the class of individuals represented by the research." In addition, Raymond LEE (1993, p.4) suggests that sensitive research "... poses a substantial threat to those who are or have been involved." Neither of these definitions restricts the focus to only the interviewee, nor do they define the nature of risk to all the participants in the research process. [3]

Studying "sensitive" topics creates both methodological and technical issues for the researcher. The issues may include (1) conceptualisation of the topic, (2) defining and accessing the sample, (3) mistrust, concealment and dissimulation between the researcher and participants, and (4) safety (LEE 1993). A review of the literature reveals that little has been written about safety issues surrounding participation in research related to "sensitive" experiences such as domestic violence, child abuse or death. Kathleen COWLES (1988) and DUNN (1991) provided a generic approach discussing issues that affect the psychological well being of the researcher and participants. Some authors focus on the psychological effects on the researcher (BURR 1996; DRISCOLL, HULL, MANDRYK, MITCHELL & HOWLAND 1997; YOUNG 1997; ROWLING 1999) In contrast, Barbara PATERSON, David GREGORY and Sally THORNE (1999) outlined a protocol for the researcher's physical safety. This discussion, albeit brief, is not reflected in current research processes, especially the protection of all those involved in research from harm. [4]

This paper discusses a number of "risks" that were reported or observed by a researcher and others (e.g. transcriber), during a study of women's understanding of the experience of domestic violence during the childbearing years (McCOSKER 1995). The study by Heather McCOSKER (1995) is a case study that raises issues relevant to considering methodological aspects related to physical, psychological and emotional safety for participants in a research program. The research focused on domestic violence and abuse as a phenomenon that is considered by many to be "sensitive". The research highlighted potential psychological and emotional risks that were associated potentially with all participants that might include the interviewer, interviewee, transcriber, supervisor, examiner, and reader (hereafter referred to as all participants). Few "risks" were identified during the development of the research proposal, except those pertaining to the interviewee, and no other risks to participants were recognised by the Ethics Committees, who reviewed the research proposal. [5]

The purpose of an Ethics Committee is to review proposed research in order to minimise potential harm, anticipate potential risk and, if necessary, request a researcher to pose solutions or alterations to their research. The process is designed to protect the well being of all participants. In general Ethics Committees have tended to focus on the interviewee's physical and psychological safety during the research process, and less often, the physical and psychological safety of others involved in the research. Debate continues on the adequacy of ethical guidelines to protect participants, especially those in powerless positions (LOFF & BLACK 2000) let alone consider the protection of others including the researcher. However, without review the human rights of potential subjects, especially those who are "vulnerable", may be compromised. An Ethics Committee, which should have no vested interest in the project, theoretically provides an impartial review with a view to protecting the participants from exploitation or harm. [6]

For the study of women and domestic violence ethical approval was sought from both a University and the relevant agencies to interview six women. During the research process the Ethics Committees considered the psychological safety of interviewees to be of paramount importance. Before the research could proceed a clear protocol was requested to respond to the potential psychological distress that the interview might cause the women. The protocol required first, that a counselling service assesses and determine that the women would not be placed at psychological risk by participating in the study, and second, that the women had completed a counselling course to assist them to examine their experience's of domestic violence. Access to counselling was to be established if the woman requested or appeared to need it, following an interview. The concerns of the Ethics Committees did not acknowledge how the associated sensitive material could be psychologically disturbing for the other participants. [7]

During the research process it became clear that there were potentially significant emotional, physical and spiritual consequences for all individuals involved in the research. It has been reported that researchers and transcribers may experience physical and emotional symptoms that include headaches, gastrointestinal disturbances, sleep disturbances and nightmares (COWLES 1988; BURR 1996; GREGORY, RUSSELL & PHILLIPS 1997). In fact, participation has the potential to alter relationships with others including partners and immediate family. Whilst submission of a research proposal for ethical approval to an Ethics Committee does encourage physical, psychological and emotional safety of the immediate focus of the research it does not guarantee safety for all participants involved in the research project. [8]

Both within the literature, and from field experience, there are two major issues of safety (physical and psychological) that need to be addressed from the perspective of the many people involved in a research study. Physical safety may relate to the interviewee, interviewer, and potentially their immediate families. For example, in studies of domestic violence and child abuse safety may include entering the abused or abuser's space (e.g. home, work). The researcher may be placing themselves in the power of someone else, with potentially harmful consequences (PATERSON, GREGORY & THORNE 1999). For example, interviewing a woman in an abusive relationship may place the researcher at risk of physical and/or verbal abuse by the woman's partner. [9]

The second issue is the psychological impact of researching or being a participant in "sensitive" research. There may be an immediate and/or a delayed impact on the mental health of any one involved that may include a physical response to the psychological impact. Kathleen COWLES (1988) reported episodes of insomnia and nightmares when she conducted interviews with survivors of murdered adults late in the day or evening. The issues being studied are usually those that precipitate a crisis or create major issues for those involved in the life experience. Interestingly, there is literature recognising the adverse psychological and emotional effects from exposure to traumatic situations, especially for emergency workers (DRISCOLL et al. 1997). [10]

DRISCOLL and colleagues reported strategies that were used to minimise the potential impact of research data on the data collectors in a study of coronial reports of traumatic death. Prior to the commencement of data collection the data collectors attended a three and half day workshop which included discussion of the potential psychological impact of the material which they would be coding. The data collectors were encouraged to socialise together during the workshop to facilitate recognition of potential support networks. The data collectors worked in pairs during the coding process in order to increase the validity and reliability of coding, but also to provide one another with psychological support. Whilst we are beginning to recognise the impact of those who experience phenomena such as abuse, and death, we are also beginning to examine the impact on those witnessing or surviving these events. Not withstanding, current research literature carries few articles (COWLES 1988; PARKER & ULRICH 1990) outlining strategies to be used to minimise psychological impact on participants, especially when a research focuses on a "sensitive" phenomenon. [11]

Research on "sensitive" phenomena, especially that in which data are collected through an interview, may create issues for all participants in the research process. Where the research is part of a Masters/Doctorate the supervisor may also be amongst those affected. [12]

Individuals and groups whose life experience is being sought through research are most considered in the research process, especially by Ethics Committees. There is not a doubt that these participants may experience threats to their physical and psychological safety as a consequence of their involvement in research. For most interviewees it is their exposure to life threatening or crisis precipitating phenomena that generates interest. Kathleen COWLES (1988) suggests that interviewees can respond both positively and negatively to the request to discuss life-events and experiences. An interviewee during research into the experience of domestic violence (McCOSKER 1995) made the following comment, "Being interviewed by you was more useful than the counsellors at X". Kathleen COWLES (1988) reported subjects who slept better and had less nightmares, which had been their response to experiencing the phenomenon following interviews. However, she also reported subjects who required interventions such as counselling and support immediately following an interview. This may lead some to state that they found the interview as useful as counselling, while others say it brings back things that they thought were dealt with and finished. [13]

Clear protocols need to be included in research proposals and submissions to Ethics Committees that outline how interviewee's risk is minimised when participating in a research project. As discussed further in this paper a protocol could contain guidelines about contacting interviewees, arranging interviews, protecting the physical safety of the interviewee and interviewer, terminating the interview, arranging referrals and debriefing. [14]

The impact of the researcher on the quality of the data and the potential biases that may be introduced are the subject of constant scrutiny and debate (KVALE 1996; ROWLING 1999). Methodological issues related to objectivity, contamination of research data and validity are emphasised in association with qualitative research (GERBER 1994; KVALE 1996). However, there are issues not considered often in current literature. For example, when planning research, what resources are available to guide the researcher in determining the potential physical and psychological impact on the interviewer of the interviewee's description of life experience? If researchers plan the nature and size of their sample should they not also consider how many interviews the researcher should undertake in a week? Kathleen COWLES (1988) reported that she could undertake no more than one or two a week. Whilst other researchers may be able to undertake additional interviews, there are no research findings or recommendations to guide researchers. For the researcher the sense of emotional exhaustion and being overwhelmed by the nature of the interviewees experience can be extreme (McCOSKER 1995). [15]

Confidentiality is an important ethical issue, however, so is the minimisation of harm to the researcher. Does the research proposal address issues such as with whom can the researcher discuss the contents of the interview without breaching confidentiality? Do plain language statements explaining the research to the interviewees include an explanation of who might read the transcripts or hear the taped interviews. For example, if the transcriber is not the researcher does this present a challenge to the issues of confidentiality? It is important to be clear that it may be the researcher's response to the content as well as the details of the interviewee's experience that will be the subject of the discussion. [16]

The essence of transcription is the transformation of data from an oral to a written form (KVALE 1996). The aim of transcription is to capture the intent of the participant, which requires elements of not only the words but also how they were said. It is an endeavour to capture the essence as well as the technical quality of the interview. Thus the perfect transcription is more than perfect technical verbatim copying, it also requires some interpretation on the part of the transcriber. The transcriber then becomes a human participant in the research process and not merely an extension of the tape recorder. However, they remain muted and invisible in most research programs (GREGORY, RUSSELL & PHILLIPS 1997) even though the transcriber is central to the quality of the research. They are often silent, and seemingly both unaffected and neutral in the research process, despite the fact that Vivien LANE (1996) emphasises the influential role the transcriber can have on the data that is transcribed for the researcher. [17]

However, there is limited knowledge of the impact of the research data on transcribers. Those people who have not been involved actively in the transcription process or have transcribed research about non-sensitive issues tend to view the transcription in a mechanistic and objective manner. Few novice researchers are advised about the psychological and emotional impact of interview content requiring transcription nor are transcribers briefed and debriefed properly during the process of transcribing data. The transcriber is exposed to the experiences of an interviewee's life. Discussion may include graphic description of disturbing, heart rending and frightening events. Interviewees can describe their consequent suffering, misery and heartache. These experiences may be novel and shocking to people and/or they can find parallels in a transcriber's life. A potential' connection' between the transcriber and the interviewee can be created through the process of undertaking interview transcription. The process can provide insight into the incredible power of the human spirit to conquer, overcome and learn from life and the sadness of human frailty. [18]

In research on women's experience of abuse (McCOSKER 1995) only one transcriber was used. She described feeling nauseated at times during the transcription and had to limit how much time she spent in any one day working on transcription. She also stated that the transcription process altered her understanding of the nature of abuse. Regular debriefing between the researcher and the transcriber during the transcription process and the documentation of both the impact and potential impact on the quality of the data became part of the research process. What impact this had on the transcripts was difficult to determine in this study. Vivien LANE (1996) wrote of the changes, which transcribers made through the use of punctuation including capitalisation of words such as doctor (Doctor), and tidying the language and grammar of the participants, highlighting the need for verbatim transcription. [19]

The transcriber should be considered in any research proposal, with a clear indication of how this person will be provided with a "safe" working environment while also maintaining the "quality" of the research. David GREGORY, CYNTHIA RUSSELL and LINDA PHILLIPS (1997) and other researchers suggested the following strategies. The strategies include that the transcriber:

The potential confronting nature of research data associated with "sensitive" phenomenon has not been highlighted in the literature. However, informal discussions with researchers in the area of women and violence revealed that they themselves often feel the need for debriefing. Theoretically, the research student was the appropriate individual with whom to discuss sensitive issues associated with the research. However, when issues have an association with personal experiences and are novel to the supervisor this may not be reasonable or desirable. Supervision has many dimensions, and many supervisors are involved in research supervision of projects due to their methodological expertise, rather than background knowledge of the phenomenon under examination. This creates risks potentially for a supervisory team when dealing with "sensitive" research content. All people involved in the research bring their own positive and negative life experiences, with the potential for significant responses to the data. It is fair to suggest that just as a researcher and transcriber may require support during a research project involving "sensitive" material, so too may the supervisor of a research project. In addition, the supervisor needs good listening and basic counselling skills in order to assist the research student to engage with "sensitive" data. The supervisor may be called upon often during meetings and at other times, to support the person who is learning to be a researcher. Adequate assessment of the potential effect of research of the proposed given the students prior life experience and support networks is an essential part of the supervision process. In addition, counselling services need to be available for both the student and the supervisor, to assist them to resolve any psychological and emotional distress. [21]

On completion of a research project Heather McCOSKER (1995), was asked to place a copy of her thesis in the library of a Domestic Violence Resource Centre. Questioning of the librarian revealed that the thesis was borrowed more by women seeking information about domestic violence than by students or academics. Women who read the thesis reported being able to see echoes of themselves in the study. This lead to questions about the relationship between "sensitive" research reports and the potential response of readers to the results of a study. How and where the data are reported may affect the approach taken, but the nature of the response should never be assumed. Issues for further consideration are the language and conclusions drawn from the study that should be assessed for their potential positive and negative impact on the reader, not merely for their scholarly importance. This may create conflicts in terms of expectations for those scholars who are writing a thesis, however, acknowledging the impact of the content and including the issues of support is important to the safety of all participants. [22]

How can researchers take the above advice and use it effectively in their investigations? When designing or planning a research project focussing on a "sensitive" phenomenon prior to submission for ethical approval it is useful to consider developing a protocol or guidelines to address both the physical and psychological/emotional impact of the research process for all the participants. Strategies for improving these impacts are presented in the following sections. [23]

When undertaking research in areas where participants, setting or others pose a physical safety risk to the researcher/data collector there should be a clear protocol or safety plan. The plan should include assessment of the following:

On completion of the assessment a protocol should be developed and include guidelines for the following: [25]

Researchers may be guided by an Ethics Committee to recruit and contact participants through "safe" mechanisms, for example, when examining the phenomenon of women and violence recruiting through domestic violence shelters or agencies that provide support and counselling. Barbara PARKER and Yvonne ULRICH (1990) suggest the use of an advertisement with a phone number connected to an answering machine or post office box. Whilst public recruitment may increase the number of interviewees, there are associated risks with this strategy that should be evaluated carefully in relation to the phenomenon under investigation. [26]

The time and place for an interview should be in environments where the researcher and the interviewee feel safe. For example, a woman in an abusive relationship may prefer to choose an interview time when the abuser is unlikely to return to the place of the interview. [27]

Interviews may be undertaken in a public place, or where a second person is easily accessible to the researcher. The location should be known to the researcher and interviewee prior to the interview and where possible the exits clearly accessible. [28]

A phone call where possible, or a check of the physical location by the researcher prior to commencing the interview is recommended in order to ensure safety. [29]

Always have a list of the name of the interviewee, and the place and time of the interview. Copies of the list should be given to at least one other person. This may include police, psychologist, social workers or caseworkers. Carrying a mobile phone with pre-programmed telephone numbers is a useful tool if the researcher believes there is the slightest risk to their physical safety. In cases where there is sudden or immediate risk to the physical safety there needs to be strategies available to resolve the situation. For example, in the study of abused women the researcher relied on the knowledge of the woman to decide when the interview was no longer safe for either the interviewee and/or the researcher (PARKER & ULRICH 1990; PATERSON, GREGORY & THORNE1999). [30]

There should be a process for amending the protocol if threats are identified that had not been recognised prior to an interview. Threats and a response to them should be documented clearly in the research process to protect interviewees and assist other researchers. [31]

Notwithstanding the debate within differing methodologies about the nature of objectivity, and the relationship between the researcher, the interviewee and the data, there is a need to be clear about the psychological safety of people involved. From the perspective of the interviewee this may require the researcher/interviewer to assume a "counselling" role within an interview. In the context of women and violence, Barbara PARKER and Yvonne ULRICH (1990) refer to this issue as the relationship between research and intervention. Strategies may include giving the interviewee time to cry or express significant emotion and acknowledging the importance of this to the well being of the participant. The relationship requires the researcher to develop and use his or her awareness of cues and/or signals by which the interviewee is indicating distress. While some may assert that this detracts from the quality of the data (MORSE & FIELD 1985), others argue that is contributes to the depth of the data (COWLES 1988; WILDE 1992). If the researcher indicates acceptance of the interviewee's emotional response the interviewee may feel that it is safe to reveal further information, which he or she may have felt was an "unacceptable" response or feeling (COWLES, 1988). [32]

It is important for the researcher to comment on the strengths of the interviewee, whilst at the same time allowing the interviewee to terminate the interview if too distressed. Part of the response may require the researcher to respond to, rather than ignore, an interviewee's question. Failure to respond to the interviewee may lead to a loss of trust, and alter the quality and nature of the data. A principal concern for any data collection process, irrespective of the methodology, is that asking people questions about their views, knowledge, attitudes or life experiences, particularly related to "sensitive" phenomena, may generate emotional responses that should be acknowledged by the researcher and the research process. [33]

In a study of women and abuse (McCOSKER 1995) strategies used by the researcher for coping with emotional and psychological distress created by the content of the interviews included: 1) limiting the interviews to one per week, 2) conducting the interviews in the morning, 3) arranging for someone else to undertake the transcription of data, 4) listening to no more than one hour of interview tapes without a break and change of activities, 5) not reading for the literature review about abuse while also undertaking the data collection, and 6) debriefing with a trusted colleague and the transcriber. The participants were aware that the debriefing would take place and with whom, and it was made clear that the discussion would include reactions to the material and not the content itself. The strategies were identified by trial and error rather than by using pre-planned strategies for facilitating psychological and emotional safety. [34]

The nature of "sensitive" research is such that while the development of protocols and procedures to minimise the risk of psychological and physical safety may be possible, response to the needs of the individual is paramount. Each interviewer or transcriber may need to find his/her own strategies to deal with issues as they arise. What is important is that these processes are clearly documented as part of the research process, an element that has been largely omitted in research reports to date. This means that such "sensitive" research is case or study specific so that researchers must treat each circumstance on its merits and deal with interviewees as they find them, brief interviewers carefully, and consult closely with transcribers. One general strategy will not be adequate for all "sensitive" research situations. [35]

When research focuses on phenomena that may be considered "sensitive" within a cultural and social framework there may be implications for many people who are involved in the research enterprise. Much of the current literature and the approaches of Ethics Committees have been on risk minimisation for the "interviewee" or individuals who constitute the research sample. This paper has highlighted that there are others who are directly and/or indirectly affected by the phenomenon. [36]

It is important for the safety of all involved in the research process that physical and psychological safety issues for all are included. It is not sufficient to consider only the participants at the time of the ethics application submission. Part of the process includes appropriate assessment of risk and the development of clear safety plans and processes that are outlined in a research proposals. This requires the development of relevant education approaches to assist researchers and Ethics Committees to understand the ramifications for all participants. Research on "sensitive" phenomenon should be conducted in a supportive environment where the safety of the interviewee, interviewer, researcher, transcriber, supervisor, examiner and reader are considered as equal participants in the research. Issues of physical, emotional and psychological safety are issues for all participants in any research process. [37]

Recognising the fear generated in all participants involved in the study of "sensitive" research is important as different studies such as those on phenomenon like abuse, suicide or childbirth, may produce more emotional effects than research involving the study of satisfaction and acceptance of a new medical product. Methodologically these differences should be acknowledged as influencing the study, especially as they may provide insight into dimensions/findings not previously anticipated by the researcher. [38]

Examination of issues that affect the rigour and quality of the research process is essential, if researchers are to defend the outcomes of their research. Little emphasis has been placed on examining the impact of the transcriber on the quality of data, or on the lack of clear risk minimisation protocols on the sampling associated with "sensitive" phenomenon. In addition, Ethics Committees may be affecting the direction of research by imposing unrealistic or unnecessary restrictions on some research areas, while ignoring the impact of others. Each of these issues warrants further examination and possibly research in its own right. [39]

These aspects of qualitative research have not taken into consideration a series of cultural aspects that could have another set of impacts on such "sensitive" research. These comments have focused on qualitative research in Westernised societies. They have not considered attitudes and values that are held by communities and social groups in different developing countries. For example, in some of these communities, differing gender roles and attitudes to describing personal experiences can mean the collection of incomplete data or occasionally no data because the people are unwilling to share their feelings and experiences. Qualitative researchers need to situate their "sensitive" research even more closely in these developing communities. [40]

All of these points mean that doing "sensitive" research is actually more complex than is painted in many qualitative research books. It means that a successful "sensitive" research study must maintain the highest standards of quality and rigour throughout the whole research process rather than in selected parts of the study. It also places higher degrees of importance on components of the research project, e.g., transcribing the interviews, than have previously been addressed in the literature. However, it does point to improved quality in the research investigation. [41]

Burr, Gayle (1996). Unfinished business: interviewing family members of critically ill patients. Nursing Inquiry, 3, 172-177.

Cowles, Kathleen (1988). Issues in qualitative research on sensitive topics. Western Journal of Nursing Research, 10, 163-170.

Driscoll, T.R.; Hull, B.P.; Mandryk, J.A.; Mitchell, R.J. & Howland, A.S. (1997). Minimizing the personal cost of involvement in research into traumatic death. Safety Science, 25(1-3), 45-53.

Dunn, Linda L. (1991). Research alert! Qualitative research may be hazardous to your health. Qualitative Health Research, 1, 388-392.

Gerber, Rod (1994). A sense of quality—qualitative research approaches for geographical education. In H. Jager (Ed.), Liber Amicorum Gunter Niemz (pp.38-51). Frankfurt am Main: Goethe University Press.

Gregory, David; Russell, Cynthia & Phillips, Linda (1997). Beyond textual perfection: transcribers as vulnerable persons. Qualitative Health Research, 7(2), 294-300.

Kvale, Steinar (1996). Interviews: an introduction to qualitative research interviewing. Thousand Oaks: Sage.

Lane, Vivien (1996). Typists' influence on transcription: aspects of feminist nursing epistemic rigor. Nursing Inquiry, 3, 159-166.

Loff, Bebe & Black, Jim (2000). The declaration of Helsinki and research in vulnerable populations. Medical Journal of Australia, 172, 292-295.

McCosker, Heather (1995). Women's conceptions of domestic violence during the childbearing years. Masters of Nursing thesis. Brisbane, Queensland University of Technology: 139.

Morse, Janice & Field, Patricia (1998). Nursing research: the application of qualitative approaches (2nd edition). London: Croom Helm.

Parker, Barbara & Ulrich, Yvonne (1990). A protocol of safety: research on abuse of women. Nursing Research, 39(4), 248-249.

Paterson, Barbara; Gregory, David & Thorne, Sally (1999). A protocol for researcher safety. Qualitative Health Research, 9(2), 259-269.

Rowling, Louise (1999). Being in, being out, being with: affect and the role of the qualitative researcher in loss and grief research. Mortality, 4(2), 167-182.

Sieber, Joan E. & Stanley, Barbara (1988). Ethical and professional dimensions of socially sensitive research. American Psychologist, 43, 49-55.

Wilde, Verina (1992). Controversial hypotheses on the relationship between researcher and informant in qualitative research. Journal of Advanced Nursing, 17, 234-242.

Young, Marta (1997). Bearing witness to the unspeakable. Women and therapy, 20(1), 23-25.

Lecturer – Nursing and MidwiferySchool Of Nursing
Queensland University of Technology
Victoria Park Road
Kelvin Grove,
4059, Australia

Lecturer – Nursing
School Of Nursing
Queensland University of Technology
Victoria Park Road
Kelvin Grove, 4059, Australia

Dean, Faculty of Education, Health, and Professional Studies
University of New England
Armidale, 2351 Australia

McCosker, Heather; Barnard, Alan & Gerber, Rod (2001). Undertaking Sensitive Research: Issues and Strategies for Meeting the Safety Needs of All Participants [41 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 2(1), Art. 22, http://nbn-resolving.de/urn:nbn:de:0114-fqs0101220.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License