Ethical Issues in Qualitative Research: What Would You Do?
Deslea Konza
University of Wollongong
Introduction
In any research endeavour, researchers are expected to employ high standards of academic rigour, and to behave with honesty and integrity. By its very nature, qualitative research, is immersed in a 'messy, chaotic reality of on-the-spot personal interaction....sensitivity and experience'
(Holbrook, 1997 p49 ). Rather than the objectivity and distance which characterise the more positivistic approaches, qualitative research brings with it a greater likelihood that ethical issues, such as those associated with informed consent and confidentiality, will arise. Some difficulties emerge precisely because a high level of rapport, so essential for many aspects of qualitative research, has been established between investigator and participant. That which is required for the research to be valid, can also be the catalyst for a complex array of highly charged interpersonal issues to emerge.
There has been little discussion of the ethics of educational research until comparatively recently
(Bibby, 1997; Burgess, 1989)
. Reasons put forward for this include the fact that educational research more often involves reporting existing conditions rather than aggressively experimental work; failed experiments rarely result in death or disfigurement as might be the case with medical research; and educational harm can usually be reversed, unlike some medical consequences.
This paper aims to generate a discussion of ethical dilemmas which faced the writer when engaging in research relating to students with a range of special needs. Several of the dilemmas arose when conducting interviews with the parents of these children:
• when it became difficult to determine whether or not the consent given prior to research was, in fact, 'informed', due to literacy, cognitive or psychiatric problems which became evident throughout the research;
• when the issue of confidentiality conflicted with disclosures of illegal activity, including child abuse;
• when 'reflective listening' and 'acceptance of individual realities' would have involved reflecting back damaging personal beliefs; and
• when the role of interviewer appeared to merge with that of counsellor.
It is hoped that a discussion of these issues will result in a clearer understanding of some of the difficulties associated with qualitative research and of our responsibilities towards people willing to assist us in our research, especially those who may be particularly vulnerable.
The issue of 'informed consent'
Informed consent is seen to be one of the most critical issues in qualitative research, indeed, it is regarded by some
(Bogdan, 1992; Evans, 1996)
as the 'key issue' in research with human beings. While covert research is claimed to have some validity if it is used to expose malpractice or corruption,
(Punch, 1986)
or if there is no risk to the participants
(Fine, 1992)
it is largely rejected and is seen by many as being intrinsically wrong. Kiegelmann
(1996)
stresses that the foremost consideration in research should be given to the dignity of the research participants, and that on these grounds alone, deception in research can never be condoned.
What constitutes 'informed consent?' Particular problems relating to informed consent arose on two occasions throughout the study referred to in this paper.
In one case, a prospective participant's mother had received a letter explaining the proposed project and extending an invitation to attend a meeting at which further details regarding the research would be provided. Although the mother did not attend the meeting, she sent written approval for her son to participate in the study. Subsequently it was discovered, through an indiscreet disclosure by a staff member, that this particular mother was undergoing treatment for a severe psychiatric condition. After discussion with the principal of the school, it was decided that it was not possible to ensure that 'informed' consent had been obtained, and therefore that particular child was excluded from the study. In this case, the decision appeared to be quite straightforward.
In another case, the primary caregiver was not literate and so was unable to read the explanatory note. The teacher and the investigator explained the purpose of the research personally to the mother, who expressed her willingness for both her and her child to be involved.
On subsequent occasions, when the investigator was attempting to arrange interview times, the mother was unable to participate in the discussions because of her extreme agitation about unrelated matters. The mother appeared to be totally unaware of the research under discussion. After two further attempts to explain the project and arrange interview times, it was decided that the ethical way forward would be to exclude that particular child from the study, as informed consent did not appear to have occurred. Observations on that child ceased from that point. Shortly afterward, the mother approached the investigator and asked when she could be interviewed, and appeared to be fully cognisant of the research. Discussion with the class teacher, the principal and the research supervisor lead to an agreement that the mother appeared to be genuinely consensual for her child to be part of the study and so the child was once again included.
As the study progressed, the mother would often spend considerable time talking to the researcher when she dropped off her two children at school and there was no indication that she had any concerns about the research. On one occasion during the final days of data collection, as the researcher approached the mother, she reacted in quite a startled manner and ran from the school grounds. This pattern of behaviour continued from that point on. Attempts by staff members to determine the cause of her behaviour resulted in the discovery that the mother had somehow connected the presence of the researcher with the principal's reporting of suspected child abuse, and the mother now believed that the investigator was a representative of the Department of Community Services and was wanting to interview her regarding child abuse.
The pattern of behaviour strongly suggested that the mother was not in a position to make clear judgements. While she did not officially withdraw her consent, it was clear that she was no longer comfortable with any level of involvement with the researcher. Did this mean that her daughter (the only female in the sample) should be removed from the study in its final days, and that all data collected to that point discarded? Would new understandings that may arise from analysis of data involving her daughter justify her continued involvement as a participant? If, as has been suggested by
(Fraenkel, 1990)
, the central ethical question for all reseachers is 'Will any physical or psychological damage come to anyone as a result of this?' the answer in this case is probably 'no'. The mother was not concerned about the classroom research which involved her daughter, but about possible intervention from a supposed Department of Community Services agent.
The outcome
Discussions with the principal, class teacher and other experienced qualitative researchers did not result in any clearcut answer. As pointed out by the principal, the young student played an important part in the research as the only female in the group and the investigation was likely to throw more light on a perplexing condition (Attention Deficit Hyperactivity Disorder). The mother had at no time objected to the classroom research being conducted and in fact showed a great interest in it. It was not the research or the researcher she was attempting to escape, but rather a fantasy figure and situation she had created. The class teacher believed that no harm could arise from using the data collected on the young girl, and that should be the guiding principle. Experienced researchers differed in their opinions, with some recommending use of the data, and others saying, if there were any concerns, withdrawal was the safest way forward. Data on that student was not withdrawn from the study, although information gained from the informal interviews with the mother was not used. It was also decided that that particular student should not be one of the focus students, and provide supporting evidence only.
The issue of confidentiality
The issue of confidentially is one which underpins all qualitative research
(Fraenkel, 1990; Raffe, 1989)
and is seen to be a major issue in qualitative research. Throughout this research, which was conducted over a period of one year in two different schools, the researcher became a familiar figure to many of the parents whose children were participants in the study. This arose chiefly because the parents would often linger after dropping their children off at the school, keen to hear how their children were progressing, and apparently keen to just 'have a chat' with the researcher.
While this arrangement greatly facilitated the development of a high level of rapport with the parents, on several occasions, by the time the formal interview took place, and on some occasions, merely as part of the informal chats which occurred of a morning, some highly personal revelations were made. Spontaneous and intimate disclosures are not uncommon in qualitative investigations
(Rowling, 1994)
. These disclosures included those of drug usage and distribution, unregistered firearms in a highly unstable home, sexual assault and, on one occasion, of child abuse. Suggestions that relevant agencies be contacted were not acted upon.
The outcome
Private lines of enquiry revealed that the drug usage and distribution had been reported to the local police, but was not considered to be serious enough to warrant immediate action as it was felt that their was little chance of a conviction. On the last day of the weapons amnesty, the unregistered firearms were handed in, and coincidentally, the suspected child abuse was reported by the principal of the school. In each of these cases, the researcher did nothing. Confidentiality was maintained. But in each of the cases, there could have been serious repercussions. Could it have been that these disclosures were meant to force action on the researcher's part; to, in effect, pass on the responsibility of reporting to another person? Should researchers be made mandatory reporters of suspected child abuse? As pointed out by Bogdan, researchers surely have obligations as citizens as well as obligations which their research places on them
(Bogdan, 1992)
. At what point does moral responsibility outweigh the confidentiality that we assure Ethics Committees is in place in our research?
The issue of 'individual realities'
In ethnography, the commitment is to understand and interpret processes which are occurring in their natural settings in a holistic manner, acknowledging that there are multiple realities. All efforts are directed towards reflecting the viewpoint of the individuals directly involved in the phenomena being studied. There is no attempt made to alter any participant's view of reality. Should this, then, still be the case when 'reflective listening' and 'acceptance of individual realities' involves reflecting damaging personal beliefs?
When interviewing the parents of one particular child, and questions were raised about their beliefs concerning the cause(s) of ADHD, the mother stated quite spontaneously that she always believed it had been because their son had not been planned, as their first child had been. She then became quite distressed, as did her husband who 'confessed' that he believed that also. This was apparently their first articulation of this shared belief and the level of distress was such that it was difficult for the researcher to know how to proceed. Ethnographic research strives to reflect the beliefs of the participants, but there seemed to be, in this case, needless suffering based on faulty beliefs.
The investigator was not sure how to proceed. These beliefs were the source of considerable pain to the participants. The beliefs have no scientific support. Would it be damaging or helpful to question these beliefs? To what extent do researchers have a responsibility to accept the beliefs of the participants, even beliefs that appear to be faulty or dangerous?
The outcome
The investigator made references to children with a diagnosis of ADHD who had been adopted, and so clearly planned, and to other children who were the result of IVF treatments, and so carefully planned, who also had a diagnosis of ADHD. There was also mention of the fact that unplanned did not necessarily mean unwanted.
Whether or not this was the appropriate course to take is a matter for reflection. The explanation - 'unplanned does not necessarily mean unwanted' appeared to be of comfort to them and they seemed surprised and relieved to hear of adopted children having ADHD. When the interview transcript was being read back to the parents and they were asked if they would like any material deleted, they elected to have the material remain intact. This suggests that gently challenging, rather than reflecting their beliefs had a positive outcome, although this is clearly not part of the ethnographer's normal role.
The issue of researcher versus counsellor
Participating over a period of time with a group of people in order to 'tell their stories', which is, in effect, what the ethnographer tries to do, often involves the formation of close relationships among the participants. Parents of children with special needs are often undergoing periods of intense stress and greatly in need of personal support. It is not uncommon that the researcher, a person who, by definition, has a great interest in the lives of the participants, is seen to be a source of information and particular support.
In cases where interviews are conducted over a period of time, a dependent relationship can develop. An extreme example of this occurred with the mother of one of the participants, Michael, a particularly challenging young boy. There was a high level of instability, physical violence and disruption in Michael's home. Michael's mother was genuinely concerned for the welfare of her children, but her own coping resources were severely taxed by her circumstances, leaving her little time or energy to address the problems directly. The research called for several interviews over a period of time and Michael's mother was always co-operative and helpful. Many personal issues emerged throughout the interviews and it was clear that she would benefit from professional counselling. Persistent suggestions were made by the researcher to the effect that professional help should be sought, however attempts to direct her to relevant services were not successful.
The outcome
The researcher had provided details of when she would be at the school, with the usual advice that she could be contacted at any time for information about the project. Requests to visit Michael's home became more and more frequent as his mother's ability to cope deteriorated. If the researcher was not available to visit (and the school administrative staff tried to relay this message), the mother would arrive at the school in a distressed state. The school counsellor, the school principal, various other members of the teaching and clerical staff became involved at different stages in an effort to assist the mother access approriate services. Eventually the principal had to bar her from coming to the school. This situation was not relieved until the data collection period ended.
Clearly the professional and the personal interactions had become confused, but at precisely which point, it was difficult to judge. Early interactions with the mother suggested a difficult time was being experienced but there was little indication that the behaviour would become extreme. There is also a continuing sense that the mother felt greatly let down by the whole process. The research was presented as something which was aimed at helping both parents and teachers understand the behaviours of students diagnosed with Attention Deficit Hyperactiovity Disorder, and indeed this was the intention. It was as if the mother saw this as a promised source of support which did not deliver. Her position was worse after the research, rather than better - to what extent the research process actually triggered this deterioration is difficult to say.
Suggestions for future directions
Extend discussion of ethical issues in research courses
Bibby (1997) contends that few researchers have a clear understanding of both research methodology and the moral argument or theory which should guide decision making processes when confronted with ethical dilemmas. It is important that researchers consider different moral arguments as part of their training and reflect on what is acceptable behaviour before beginning a research career.
Moral theories fall largely into three groups:
• consequential theories which state that what is good or bad can be determined by looking at the consequences of our proposed actions;
• deontological theories which expound the view that there are some duties which are absolute or obligatory despite the consequences; therefore there can be no justification whatever for deceit to be used - it is always wrong because it fails to respect the individual's right to make up his or her own mind about what should be done. This model strongly supports the rights of individuals or small groups against the interest of the majority or powerful groups; and
• Aristotlian theories which are similar to deontological theories. No action may be regarded as good if it is opposed to what most people would regard as good and proper. Thus lying, which is contrary to what most people would regard as proper behaviour, is always wrong.
Unfortunately, few guidelines can be universally applicable. There is even debate about whether it is possible to determine fundamental moral principles which should guide behaviour in all circumstances. Nevertheless, prospective researchers need to consider theories such as those briefly outlined above, even if it is only to raise awareness of the complex issues which can arise when attempting to interpret individual realities. Codes of ethics can only operate as a guide. There are no real solutions but 'a great need for researchers to regularly reflect on their work to develop their understanding of the ethical implications associated with social and educational investigation'
(Burgess, 1989)
.
Extend use of the debriefing process
Continued discussion with stakeholders in the research, supervisors and other qualitative researchers is critical and may be of assistance in coming to a decision about a particular course of action. The use of research journals as a form of debriefing
(Rowling, 1994)
may assist in the clarification of ethical issues, as different viewpoints, possible consequences and the different effects on different stakeholders are recorded and considered.
Validate interview transcripts
Rowling
(1994)
recommends the return of interview transcripts to interviewees for confirmation as one technique which is aimed at protecting their interests. This certainly seemed appropriate in the case of the parents who believed that their son had ADHD as a result of his unplanned status. They had the opportunity to state that they were happy for the transcript to remain unaltered. This does not, however, help in determining whether or not the discussion recorded was appropriate, and whether the investigator followed the appropriate course.
Collaborate with related professionals
Qualitative researchers in particular fields may need to work more closely with related professionals and to make themselves aware of services which are available to support individuals who may be experiencing periods of stress, in order that advice may be given as to appropriate follow up services.
Discuss consent issues regularly
In qualitative research, informed consent needs to be sought at regular intervals throughout a research process rather than only prior to its commencement. This should go beyond the usual statements that consent may be withdrawn at any time, but include direct discussion regarding the progress of the research and the participants' continued willingness to be involved. Cassell
(Cassell, 1982)
points out that that it is almost self contradictory to obtain informed consent when employing an emergent research design.
Conclusion
Working with children with a range of special needs and their families, particularly in the more personalised areas of qualitative research, demands sensitivity and an increased awareness of the great vulnerability of many of these families. Their life circumstances are such that they are often placed under extreme levels of stress, and therefore more likely to react in an unpredictable manner. It is often difficult to know how the longterm interests of many of these families may best be served. Many researchers, myself among them, have been quick to claim that longterm benefits will arise from the greater understanding we may derive from our research, and that this justifies short term discomfort or anxiety on the part of some participants. Yet the effects , short or longterm, that we and our research may have on the emotional well-being of participants must always be at the forefront of our consciousness as we struggle to make the day-to-day decisions that guide the qualitative researcher in the field.
This paper does not claim to have offered definitive answers to any of the complex array of ethical questions arising from this research. It seeks more to remind all researchers of the need to be continually aware of how seemingly benign research can impact on participants, and of respecting at all times the wishes of those people who allow us to share their views of the world. If we cannot be sure of what those wishes are, the best interests of the participants, rather than the best interests of our research, must always be the guiding principle.
References
Bibby, M. (1997). Introduction: Education research and morality. Review of Australian Research in education, no 4, 1-12.
Bogdan, R. B., S. K. (1992). Qualitative research for education: An introduction to theory and mehtods. Boston: Allyn & Bacon.
Burgess, R. G. (1989). The ethics of educational research. London: Falmer Press.
Cassell, J. (1982). Harms, benefits wrongs and rights in fieldwork. In J. Seiber (Ed.), The Ethics of Social research: Fieldwork, Regulation and Publication . New York: Springer verlag.
Evans, T. J., V. (1996). Research ethics in open and distance education: Context, principles and issues. Distance Education, 17(1), 72-94.
Fine, M. (1992). Disruptive Voices. New York: University of Michigan Press.
Fraenkel, J. W., N. (1990). How to design and evaluate research in education. New York: McGraw Hill.
Holbrook, A. (1997). Ethics by numbers?An historian's reflections on ethics in the field. Review of Australian Research in Education, 4, 49-66.
Kiegelmann, M. (1996). The subject writes back. Reflections on ethics in qualitative research. Paper presented at the American Educational Research Association.
Punch, M. (1986). The politics and ethics of fieldwork. (Vol. 3). California: Sage Publications.
Raffe, D., Blundell, I. & Bibby, J. (1989). Ethics and tastics: issues arising from an educational survey. In R. G. Burgess (Ed.), The Ethics of Educational Research . London: Falmer Press.
Rowling, L. (1994). Ethical dilemmas encountered in researching sensitive issues. Paper presented at the Australian Association of Research in Education.