Volume 4, No. 1, Art. 12 – Januar 2003
Qualitative Research in Sport Sciences: Is the Biomedical Ethics Model Applicable?
Steve Olivier & Lesley Fishwick
Abstract: Research in sports science has historically been grounded in positivist traditions. This means that ethics committees may not be adequately sensitized to the ethical problems posed by qualitative research. Qualitative researchers may thus be disadvantaged in the research approval process. Our paper argues that the traditional biomedical ethics model may not always be appropriate in evaluating qualitative proposals. Due to the nature of its methods, qualitative work may have emergent and ongoing ethical issues that require consultation and resolution. We argue that, contrary to the judgements of many ethics committees, methods such as deception and covert observation can be justified if certain conditions are met. In reaffirming a commitment to the overarching ethical principle of respect for persons, we conclude that researchers need to recognize and plan for ethical issues in their work. Likewise, ethics committees need to recognize that qualitative work poses unique problems, but that these need not necessarily be insurmountable obstacles to project approval.
Key words: ethics, quantitative research, qualitative research, ethics committees
Table of Contents
2. The Biomedical Ethics Model
3. Ethics Review and Qualitative Research
4. Emergent Research Design
5. Informed Consent
7. Power and Trust in Researcher/Participant Relationships
9. Benefits to Participants
10. Qualitative Research Ethics: Diverse and Emergent Issues
Despite initial resistance from traditional Sports Science practitioners, the role of qualitative research within the discipline is being increasingly accepted and valued. BIDDLE, MARKLAND, GILBOURNE, CHATZISARANTIS and SPARKES (2001, p.778) note that qualitative research in Sports Science is a relatively new field of inquiry, but that it is gaining credibility. Alternative methods such as case studies, observational studies, ethnographies, action research and personal narrative histories are being used with increasing frequency in areas such as the sociology of sport, sport psychology, and sport management. 
Qualitative research utilizes methods that are different to those employed in quantitative studies. Given that methods impinge on ethical issues, it follows that there may be some differences in the ethical problems faced by the two sets of researchers. In fact, RAMCHARAN and CUTCLIFFE (2001, p.358) contend that qualitative researchers may be treated unfairly by ethics committees1). Most ethics committees are more familiar with quantitative methodologies, which could lead to inequitable treatment for qualitative researchers. 
In the current paper, we introduce and discuss some of the ethical issues raised by the use of different research approaches, as qualitative methodologies in some cases pose a different set of problems. We debate the applicability of the commonly applied biomedical ethics model for qualitative research, and take the perspective that judgements cannot simply be applied from a positivist perspective. 
The abuses perpetrated in invasive biomedical experiments have been well documented (CAPRON 1989, p.137, & OLIVIER 1995, p.136). The response from the research community to these abuses was the formulation of the Nuremberg Code, the Declaration of Helsinki, and the many variations of ethical review/IRB approval processes in current operation. As an aside, it is worth noting the wide variations in operation of these processes. OLIVIER has expressed reservations about how the review process operates in Sports Science, contending that local variation makes judgements inconsistent, and that the process may at times be bureaucratic and restrictive. We concur with this disquiet with regard to the ways in which qualitative research is assessed (OLIVIER 2002, p.196). 
The ethics review process initiated in response to research malpractice is founded on the basic principles of nonmaleficence (not harming research participants), justice, and autonomy. Collectively these principles aim to protect people from harm, to treat people equitably, and to empower potential and actual participants. The principle of autonomy enshrines an individual's right to self-determination, and is practised through the insistence on obtaining first person, written, informed consent. Issues such as anonymity, coercion, and the right to withdraw from a project without sanction, are underpinned by the commonly accepted ethical principles mentioned above. 
Methods presented to ethics committees in Sports Science are traditionally expected to provide evidence of control of independent and extraneous variables, to describe relatively inflexible procedures (for good reasons of validity and reliability, it must be noted), and to present predetermined methods of analysis. As such, research proposal submissions in Sports Science tend to follow the tradition of the biomedical model. However, the positivistic perspective as represented by the biomedical model and guidelines of Institutional Review Boards (IRBs) may be too inflexible for qualitative studies. One of the key aspects of much qualitative work is an inductive approach and emergent design of studies (including the methods of sampling and the actual direction of the study). Also, qualitative researchers present their findings in a variety of ways which differ markedly from the presentation of quantitative investigators. Examples might include ethnodrama, and poetic representations. A different perspective on ethics, one that suggests a more flexible approach and appreciation of ongoing decision-processes, may be more applicable for the challenges facing qualitative researchers. 
What is the remit of ethics committees in general? Most fundamentally, they serve to evaluate the ethical acceptability of proposed research projects. OLIVIER (2002, p.197) holds that "... such committees are of crucial importance in regulating research and preventing abuses, since investigators should not be the sole judges of whether their research conforms with generally accepted ethical codes and practices". In addition to considering ethical issues, the role of committees has expanded to include a broad range of design issues. It is this shift from a narrow ethics evaluation to a broader methodological scrutiny that may present difficulties for qualitative research. This is particularly the case if a committee is dominated by people immersed in positivist paradigms. 
Several issues are common to both quantitative and qualitative research. These include covert observation, power relationships between researchers and participants, cultural factors, emotional involvement, benefits to participants, and public versus private behaviour. We would contend that some of these have greater prominence as ethical issues in qualitative work. 
Even so, it could be argued that the principles of nonmaleficence, justice, and autonomy apply equally to quantitative and qualitative research. We would argue that of course we ought to be beneficent when conducting our research. Similarly, we ought to respect an individual or groups' right to self-determination where it is relevant in the context of a particular research project. With regard to the latter principle, we would however argue that ethics committees should not unthinkingly insist on following the autonomy model in cases where it might not be relevant. 
Unqualified acceptance of the autonomy model disqualifies, by definition, covert research and deceptive research. Indeed, informed consent and deception are mutually exclusive concepts. It is not that we are suggesting that the principle of autonomy does not apply to qualitative work, but rather that we believe it can be justifiably overridden in certain instances. Some qualitative researchers probably agree that covert or deceptive work may be justified in some cases, but there is ongoing debate in this area. The lack of agreement over the ethics of covert work is captured by the divergent reactions to HUMPHREYS' (1970) "classic" study on homosexuals. He acted as a "watch queen" at public toilets, thus befriending the men. Later, he traced them from their car license plates, and ultimately questioned them in their homes under the guise of a different project. The work provided important information on stereotypes of homosexual men, but questions were raised as to the ethics of the project. PUNCH (1998, p.168) notes that on the one hand he received a prestigious award for his work, whereas on the other there were efforts made to revoke his Ph.D. 
Such examples serve to illustrate the quagmire of dilemmas facing Ethics Committees. These committees are often guided by the codes of ethics of professional organizations. This is in itself a potential problem, as almost all codes are deontologically based. What this means is that they are rule-based. Rules can be useful in that seemingly different moral problems are in fact similar, and the rules can then be applied universally. Also, rules are useful in providing "instant" answers. Qualitative work, particularly that which explicitly employs an emergent design, is however by definition flexible. Inflexible rule-based approaches may thus not always be appropriate for the questions posed by qualitative projects. 
Codes of ethics published by recognized associations, such as the British Psychological Society and the British Sociological Association provide general guidelines on the obligations of researchers on confidentiality, informed consent and use of deception. However, such codes are either too specific (for example "you must obtain written, first-person informed consent"—thus leaving no opportunity for alternative methodologies) or too general (for example "respect the rights of research participants"—thus giving inadequate guidance to those unfamiliar with ethical issues). In Sports Science, there is very little overt (in the form of ethical codification) recognition of the ethical issues that affect qualitative work. 
Further, there is little evidence of overt discussion of ethical considerations in the vast majority of published qualitative research reports within Sports Science. PELED and LEICHTENTRITT (2002, p.147), in their review of qualitative studies in social work, concluded that the lack of overt discussion on ethical consideration implies that the responsibility for "proper ethical conduct" lies within the individual researchers. There are inherent dangers with such a reliance on individual researchers, who have a vested interest in the research. Self-interest, ego, and career demands mean that we are almost never the best judges of our own work. Independent review, whilst not sufficient, is necessary. Nevertheless, it may be that the relative youthfulness of qualitative traditions in Sports Science means that it is being disadvantaged by ethics committees, with these committees being strongly influenced by the practices of biomedical ethics. 
There is certainly room for discussion and debate about ethical considerations in qualitative work within sport sciences. Such debates should focus on key principles including not doing harm (nonmaleficence), justice, autonomy (where relevant and appropriate), research related benefits for participants and for others (beneficence), and researchers' technical competence. Recurring themes for ethical issues in relation to qualitative work in Sports Science include the role of the researcher; the desirability and necessity of informed consent; deception; covert research; the researcher's responsibility to informants, sponsors and colleagues; risks versus benefits; reciprocity and intervention; issues of relationships and "leaving the field"; how participants are represented in reports; and how to deal with unforeseen ethical issues that emerge during and after the research. 
One of the potential pitfalls for qualitative research subjected to a qualitative tradition of ethics review is that the research design is often emergent. The inductive approach, of, for example, much ethnographic research means that developing research procedures is an ongoing process. The nature of the problem to be investigated is fluid, incompletely determined at the beginning of the study, and subject to change as the study progresses. So, intrinsic to this process is that the design cannot be fully specified in advance, but rather emerges over time. 
This is in sharp contrast to experimental research, which requires, by its very nature, detailed planning and control. In qualitative research however, it is often neither possible nor desirable to provide ethics committees with concrete numbers of participants in advance of the study (nor the specific questions that will be asked). 
In qualitative research, the nature of the procedures means that questions or lines of questioning may change according to the preliminary responses received. Indeed, the very focus of the project may change, with a new fundamental direction being pursued. For example, in a project investigating power relationships among sports coaches and young children, the initial focus might be on the coach-athlete relationship. It is conceivable though, that during the conduct of the research, it becomes apparent that the real determinant of the power situation is a notion of power being transferred from parents to the coach, and consequently exercised by the latter. Thus, it might emerge that there are multiple sources of power, and the emphasis may switch to examining the alternatives, rather than solely focusing on the original supposition. So, at any stage of a qualitative project, the information received and concurrently evaluated can influence either or both the immediate procedures and the ultimate direction of the project. This is in contrast to more traditional areas of study in Sports Science. 
For example, in a physiological study investigating time to exhaustion during energy drink and placebo conditions, the physiological risks are well established, and risk can be managed through careful monitoring and appropriate emergency medical provision. If the subject has provided voluntary consent and has adequately comprehended the relevant information, the project is likely to be deemed ethically acceptable. 
A qualitative project may however be different in that the problems are perhaps more difficult to foresee. This provides a challenge for qualitative researchers. For example, in examining group cohesion in sport teams, FISHWICK (1983) found that a particular clique's fondness for illegal drugs was a major source of discord for team morale. This led to a series of unanticipated ethical issues concerning informant confidentiality, trust, and decisions about what to include in the final report. Such unforeseen circumstances are not limited to the extended fieldwork of ethnographic research. Similar, types of issues can also be revealed during in-depth interviewing. For example, an investigator wants to examine the influence of media images on perceptions of body image. During an interview, it becomes clear that a participant suffers from a serious eating disorder. The researcher is not trained to deal with this. What ought s/he do? 
There is of course an answer to this moral issue, but this paper is not the place to explore it2). Rather, the example serves to illustrate the differences in planning and procedure that could be experienced by quantitative and qualitative researchers. Researchers should plan adequately, and ethics committees should recognize that different approaches present different solutions and problems. One ethics process model will not, in cases such as these, cater adequately for all research proposals. Generally speaking, if procedures and ethical issues are adequately considered and catered for in advance, then the project is more likely to be deemed ethically acceptable. For example, in his participant observation work on soccer hooligans it is likely that GIULIANOTTI (1995) considered in advance his guidelines for action if he himself was faced with actual involvement in physical violence. One of the rare examples of a more detailed and considered approach to guidelines of personal responsibility is given by BRACKENRIDGE (2001, pp.156-157). In her account of how she managed the interview process (which focused on sexual exploitation of athletes) she explicitly mentions oral consent procedures, confidentiality agreements, storage of data, participant input, follow-up counselling arrangements, and her stance of non-involvement in reporting on behalf of a participant. 
We are not arguing here that qualitative researchers should be singled out in terms of attempting to predict potential issues in studies, as such forethought is also needed for quantitative studies. We do not believe that ethical problems should be approached differently for different types of research. Whatever relativists would have us believe, we hold that there are basic ethical principles that ought and must be considered when planning and evaluating a project. This does not mean that a principle can never be overridden, but this would need justification. These principles include those mentioned earlier, namely nonmaleficence, justice, and autonomy. Particularly for qualitative research, beneficence (in terms of the immediate participants) could be added here. These are all subsumed under the basic principle of respect for persons (leaving aside, for the purposes of this paper, the issue of animal research). 
Specific issues that all researchers need to consider are the nature of confidentiality agreements, anonymity, privacy, risks and benefits (physical, social, psychological), consent and deception, covert observation, cultural and/or gender factors, using vulnerable populations, coercion and sanction in the participation process, the researcher's response to harmful/stressful situations (for both participant and researcher), the desirability and nature of debriefing, and how emerging and ongoing ethical issues will be dealt with. 
The principle of autonomy, embracing an individual's right to self-determination, underpins the notion of informed consent. Informed consent implies that a participant freely agrees to participate (without coercion or threat of sanction being applied), and that the relevant consequences of such an agreement is understood by him/her. In studies where the notion of informed consent is considered appropriate, it is preferable that it is obtained in written form, and that it is given by the person concerned (first-person consent). This is intended to protect both the investigator and the research participant. 
Many ethics committees, in applying the biomedical ethics model, will insist on written, first person informed consent being obtained. This may not always be appropriate, however, even in quantitative type studies. For example, if researchers wish to investigate energy expenditure among illiterate isiZulu speaking cane cutters in Africa (perhaps with the laudable aim of improving working conditions), they would find that written first person informed consent is inappropriate. The first and obvious problem is the one of illiteracy. The second issue is that, generally speaking, a community such as the research population described above subscribes to a notion of community based rights and decision-making, rather than the essentially Western notion of individualism. In this case, individually witnessed oral consent may be appropriate, if obtained in conjunction with permission from, for example, a tribal elder. It is worth noting that such consent needs to be contemporaneously recorded in writing, with this recording being witnessed if possible. 
The applicability of informed consent may vary according to the characteristics of the participant characteristics. When young children are involved, parental (proxy) consent plus the child's assent (agreement) is necessary. For older children, proxy consent and a modified (comprehensible)3) consent form should be employed. Vulnerable populations such as prisoners, or people with learning disabilities (for example), need special considerations to be applied (such as paying particular attention to comprehensibility; being aware of the potential for coercion; and so on). 
A marked difference between qualitative and quantitative consent requirements pertains to whom to ask for consent and whether consent is needed at more than one point in time. When using open-ended interviewing, questions often go down avenues not anticipated by the researcher or the participant. An example of research of this nature would be SPARKES' (1996) work on life history and narratives of self. Whilst work such as this may seem to be "unplanned" to quantitative researchers, it is of course an integral part of the exploratory, emergent nature of some qualitative work. In cases such as this, or if using covert observation, behaviours may take an unexpected directions and the distinction between public and private behaviour may become very blurred, such as in WHYTE'S (1943) classic Street corner society study. The data obtained may be extremely valuable, but does the original consent agreement cover unsolicited and unanticipated disclosures? Will such unexpected directions increase the likelihood of participants making further unwanted disclosures? Also, qualitative researchers need to consider whether or not their consent agreement includes issues relating to participant involvement in the transcription and reporting process. Will participants have the opportunity to check transcripts, and what will be their rights in terms of deleting (perhaps sensitive) information? 
The possibilities outlined in the preceding paragraph suggest that in qualitative studies such as ethnographic work, action research, and narrative life history (involving a series of interviews), informed consent is not a single event. Rather, obtaining informed consent is an ongoing process in which the researcher has to be sensitive to participants' reactions during data collection, and be prepared to renegotiate consent every now and then. Ethics committees need to be aware that in qualitative work the nature and direction of a project can change, thus changing the nature of the consent requirements. Both ethics committees and researchers must identify mechanisms whereby such changes can be communicated and facilitated. Given the relative youth of qualitative inquiry in Sports Science, we suggest that it might be prudent for researchers themselves to lead the way by including ethics review and monitoring procedures in their research proposals to committees. 
Obtaining informed consent in qualitative research sometimes poses problems. Many of the classic sport ethnographies such as FINE'S (1987) study of little league baseball, KLEIN'S (1993) study of male body-builders, and CROSSETT'S (1995) study of women's professional golf, would not have been conducted if written consent was required of all the participants within these sport subcultures. Deception and informed consent are, as previously mentioned, mutually exclusive concepts. So, by definition, if an ethics committee insists on consent, it cannot logically approve studies involving deception. This is a line that is increasingly taken by ethics committees. It may thus be up to researchers to convince a committee that deception in a study can be justified. However, as PUNCH (1998 p.171) points out, "a strict application of codes will restrain and restrict a great deal of informal, innocuous research in which students and others study groups and activities that are unproblematic". We believe that there are four basic conditions that may justify the use of deception in a study. Firstly, the results of the study must be important. Secondly, participants should not be likely to suffer physical, social, or psychological harm. Thirdly, the results could not be obtained in any other way. Finally, where appropriate, debriefing should take place. If followed these conditions should clear the way for research that might involve, inter alia, covert observation. 
For example, a study might be intended to examine the extent and antecedents of racist attitudes amongst football fans. The researcher would spend time gaining access to a group on a deceptive basis, perhaps even by deceptively revealing racist tendencies him/herself. In terms of contributing towards a just society, the results are clearly important. Participants would not suffer harm in the process of the research (this assumes that the researcher does not provoke or encourage actions, and that consideration is given to the safety of the researcher as well). Responses from the group in question could not be obtained in any other way, and the causes could not be inferred by simple observation, thus necessitating joining the group under false pretences. Debriefing would of course be problematic, but if handled correctly, could result in overall benefits. 
This does not mean that we think that deceptive studies should be blithely accepted by ethics committees. On the contrary, it needs to be stringently justified according to the four conditions presented earlier. In presenting a study for approval, researchers must be aware of the all-embracing principle of respect for persons. Deception/covert research can easily violate an individual's autonomy. Nevertheless, practising deception in research does not necessarily negatively (in practical terms) affect an individual's right to self-determination. Researchers must nevertheless remain aware of the rights of potential participants, bearing in mind ZELAZNIK'S (1993, p.68) contention that the rights of participants outweigh the rights of researchers to conduct research. By the same token, ethics committees should be aware that deceptive research can be both valuable and nonmaleficent. 
Qualitative work poses potential problems for researchers, particularly when the project focuses on vulnerable groups. Concerns regarding power, trust, confidentiality, anonymity, disclosure, and so on, are heightened. 
Given that in qualitative research the researcher is the main data collection instrument, obtaining valuable data depends on the researcher-participant relationship, as a climate of trust is a basic element of the successful data gathering process. The ability to establish a sense of trust and maintain a fine balance between objective and empathetic involvement, and the taking of a non-judgemental stance, are key skills and abilities of qualitative researchers. 
Sociological research in sport, in particular, focuses on disenfranchised and vulnerable groups. Examples of such research include BRACKENRIDGE'S (2001) work on sexual abuse victims, CLARKE'S (1997) study of lesbian physical education teachers, and PRONGER'S (1990) work on gay men in sport. The well being of participants in vulnerable circumstances (e.g. children, abuse victims, gay athletes, drug-users) is of particular concern. SWAIN, HEYMAN and GILLMAN (1998, p.22), in their work with people who have learning difficulties, note that the essentially political act of research can exploit the vulnerable and powerless groups within society, further their disempowerment, and contribute to their oppression. Also, when leaving the field, qualitative researchers need to reflect on the relationships that have developed, and they should consider their ethical obligations in this regard. 
For example, participants may feel, due to their particular circumstances, that they cannot withdraw from a study. This may merely be because they perceive that a particular power situation exists, or that they feel coerced, or that they fear some sort of sanction. Perceived sanctions may be intangible, such as a loss of "face", embarrassment at the negative perceptions of "dropout", or potential loss of self-esteem. It is incumbent on researchers to provide the correct climate for participants, so that they feel empowered. 
At the risk of playing semantic games, we feel that it is worth, at the outset, establishing what we mean by anonymity and confidentiality. 
Anonymity means that individuals, groups, or situations cannot be identified by the way in which the research is disseminated. Confidentiality is sometimes confused with anonymity. If we offer unconditional confidentiality, we cannot disseminate the outcomes of our investigations at all, as confidentiality means secrecy. Some researchers such as BRACKENRIDGE (2001, p.153) make the decision to "never tell", explicitly stating that she would rather face contempt of court charges rather than reveal her data sources. 
The crux of the anonymity and confidentiality issue is to safeguard against the invasion of privacy by assurance of anonymity. What we actually mean, of course, is that we will only disclose results in a manner in which participants cannot be identified. Safeguarding the identity of participants is often more difficult than it seems at first. Some attempts at anonymity in published research are superficial and inadequate. For example, an article might state that, in a psychological intervention program, the batsman with the highest average on the South African cricket tour to India was prone to crises of confidence. Leaving aside professional obligations between the psychologist and the client and focusing purely on the publication/dissemination issue, it would clearly not need a particularly good detective to find out who the batsman concerned was. 
Identifiers and characteristics make anonymity difficult, so the temptation is to not include them in the reports. On the other hand, they provide context and relevance to the findings and the discussion, so to leave them out renders the work sterile and perhaps lacking relevance. As pointed out by an anonymous reviewer of this paper, "The dilemma is that good qualitative case studies require 'thick description', and the better this is, the more identifiable the participant becomes". It is a fine line to tread, and perhaps we need to again return to ZELAZNIK'S (1993, p.68) injunction to give precedence to the rights of the participant. For example in a case study approach, FISHWICK (1990) had to select very broad categories for the origin of quotes (such as female, upper management) rather than specific job titles to protect the identity of the respondents with the specific organizations. 
Participants in qualitative studies are particularly vulnerable to invasion of privacy, unwanted identification, breach of confidentiality and trust, misrepresentation and exploitation. Safeguarding privacy as well as assuring anonymity is one of the key issues within preventing harm. In using quotes, interviews in life history research often reveal biographical details, and this makes protecting identities extremely problematic. Changing names and places is no guarantee of anonymity and this may be against the wishes of participants. 
One of the key ethical issues in ethnographic research for example is how participants are represented and what their rights are in the research process. Even in such an apparently straightforward convention of using pseudonyms to name informants in ethnographic research raises further ethical dilemmas. Do the participants prefer such names? How much say do interviewees have in the overall research process? Should interviewees view final interview transcripts and quotes? If participants disagree with a researcher's interpretation of events, who has the final say? These issues point to a broader concern in terms of the relationships formed within research. 
A further problem with confidentiality is that there is not, or at least shouldn't be, absolute confidentiality. What we mean here is that confidentiality agreements can, in appropriate circumstances, be overridden. 
An example from a fictitious interview study may illustrate this. A researcher is attempting to investigate reasons for teenage dropout in gymnastics. S/he agrees with all participants that their anonymity will be respected, and that any unwanted and personal disclosures to her/him by participants will be treated as strictly confidential. So, what does s/he do if a participant reveals that her/his reason for thinking about leaving the sport was persistent, ongoing and serious sexual abuse by her/his coach? What is the researcher, who is not a trained counsellor, to do? Without going into the specific courses of action open to the researcher (which are not relevant here), the answer is that the researcher must do something. That is, s/he is morally obliged to act. In this case, there is a justification for overriding confidentiality in the interest of the athlete concerned, and indeed other athletes. Whilst the example is a difficult one, with all sorts of arguments involved, it serves to illustrate that confidentiality is not absolute. 
Generally speaking, however, anonymity and confidentiality are cornerstones in solidifying the relationship between researcher and participant in the qualitative research process. Trust is implicit in the relationship, and must be maintained unless there are exceptional circumstances as described above. Where possible, the possibility of instances such as the gymnastics example should be identified in advance, and appropriate measures put in place. Ethics committees are more likely to approve projects that provide evidence of such planning than if a response to situations is perceived as ad hoc. 
PELED and LEICHTENTRITT (2002, p.149) contend that providing participants with some research-related benefits is a minimal requirement. In much feminist work, and action research, a basic premise is that participants should emerge from research with greater benefits than which they entered the project with. Such "... values direct us to go beyond fairness in our relationship with research participants, and to use the research to contribute to personal and social empowerment of vulnerable and disenfranchised groups" (PELED & LEICHTENTRITT 2002, p.148). For example, CLARKE'S (1997, p.36) research on lesbian physical education teachers provides an opportunity for the participants to tell their story, and reveals something of their lived experience, which serves to challenge the oppressive structures that "force" them to conceal their lesbian identities. By making the women the subject and not the object of analysis, she does much to make these encounters more accessible, helpful, empowering, and respectful for lesbians. 
At present, beneficence is an issue that researchers will need to make individual value judgements on. Ethics committees dominated by the biomedical tradition are unlikely to insist on benefits for participants, as they will probably view increased knowledge as a sufficient good in itself. What they might however question is the notion of external validity (generalizability), or the lack thereof (see SPARKES 1992 for a discussion of this issue). This is a debate too wide for the purposes of this paper. The reason for raising the issue here though, is that ethics committees nowadays concern themselves with more than ethical matters, and both the design and "value" of projects may be debated. In the context of committees influenced by the positivist tradition, qualitative researchers would be well advised to prepare reasons that advance the claims of the authenticity, credibility, trustworthiness, and value of their work. This will assist in accelerating the education process, whereby ethics committees need to be more cognizant of issues faced by qualitative researchers. 
Given the diverse approaches to qualitative work in Sport Science it would be neither feasible nor desirable to provide set ethical guidelines to cover all eventualities. Qualitative studies on sport range from structured interviews, content analysis and pre-determined sample sizes at one extreme, to unstructured sport ethnographies. Given this range it would be unlikely that any code of ethics could address key dilemmas in a meaningful way. If a code could address all the issues, it would probably be too general to be of any practical use. If it were specific, it would not cover the diverse range of issues confronting qualitative researchers. 
Also, we have earlier touched on the emergent nature of qualitative methods and procedures. The same applies to ethics. It is not always possible to identify and quantify risks in qualitative work. Unsolicited self-reflection is one such issue. To ensure that ethical problems that emerge during the conduct of research are dealt with, researchers need to establish a mechanism of referral at the outset. This will not necessarily be the same as the "oversight" system demanded by some biomedical ethics dominated committees. The oversight model might in fact be inappropriate for some types of qualitative research, particularly those that require privacy, anonymity, and the establishment of trust. Nevertheless, researchers and committees should identify the means to report ethical issues, and the means to solve them or receive guidance about them. 
In support of the concept of qualitative research ethics as an ongoing process, SWAIN, HEYMAN and GILLMAN (1998, p.33) hold that qualitative studies are inherently fraught with ethical dilemmas that cannot be predicted at the outset. They argue that there is a need for ethical guidelines that focus on the research process and which complement codes concerned with the planning stage. This frames ethics as a continuous process of decision-making. 
Research in Sports Science has historically been dominated by quantitative methods and traditions. This is changing, but it is not clear that the mechanisms for evaluating the ethical merits of studies are keeping pace. The methods employed in qualitative work mean that researchers face different ethical problems compared to their quantitative colleagues. Researchers and ethics committees need to be aware of the differences, and projects must be planned and presented to committees accordingly. The wide range of methods in inductive approaches makes it difficult to formulate specific guidelines for ethical conduct. Nevertheless, qualitative researchers should attempt to foresee both obvious and emerging ethical problems when they plan their research. Having done so, they should set support, monitoring and reporting mechanisms in place. Whilst we have argued that some ethical principles are not cast in stone, researchers ought to follow the overarching principle of respect for persons. When confronted with ethical dilemmas, the rights of the participants should be seen as outweighing the rights of the researcher to conduct research. 
1) Recognising the international readership of this Journal, the terms Ethics committees and IRBs are used interchangeably. Ethics Committees in the USA serve to resolve dilemmas regarding patient care (clinical settings), but similarly named committees elsewhere serve the same function as IRBs do in the USA (research settings). In this paper, the term "ethics committees" refers to the process of research approval. <back>
2) This paper is not the appropriate place for a full debate on this moral dilemma. Nevertheless, it would be remiss of the authors to not give an indication of what they believe to be the moral action to perform. Very briefly, the imperative is to act. In this case, an appropriate authority must be notified. The notion of confidentiality is not absolute, and considerations of harm to the individual, possible harm to others, and the potential inability of the individual to act on her own behalf, all combine to override confidentiality in this instance. <back>
3) For guidelines on the comprehensibility of informed consent forms, see OLIVIER and OLIVIER (2001). <back>
Biddle, Stuart; Markland, David; Gilbourne, David; Chatzisarantis, Nikos & Sparkes, Andrew (2001). Research methods in Sport and Exercise science: Quantitative and Qualitative Issues. Journal of Sports Sciences, 19, 777-809.
Brackenridge, Celia (2001). Spoilsports: Understanding and preventing sexual exploitation in sport. London: Routledge.
Capron, Alexander M. (1989) Human experimentation. In Robert M. Veatch (Ed.), Medical Ethics (pp.125-172), Jones and Bartlett Publishers: Boston.
Clarke, Gill (1997) Playing a Part: The Lives of Lesbian Physical Education Teachers. In Barbara Humberstone (Ed.), Researching women and sport (pp.36-49). London: Macmillan.
Crossett, Todd (1995). Outsiders in the clubhouse: the world of women's professional golf. Albany, NY: SUNY Press.
Fine, Gary (1987). With the boys: Little League baseball and preadolescent culture. Chicago: University of Chicago Press.
Fishwick, Lesley (1983). Cohesion in interacting sports: A preliminary qualitative study using three women's teams. Unpublished M.Sc. thesis, Dalhousie University.
Fishwick, Lesley (1990) A sporting chance? Resegregation of coaching jobs in women's intercollegiate athletics. Unpublished Ph.D. thesis, University of Illinois.
Giulianotti, Richard (1995). Participant observation and research into football hooliganism: reflections on the problems of entree and everyday risks. Sociology of Sport Journal, 12, 1-20.
Humphreys, Laud (1970). Tea room trade: Impersonal sex in public places. Chicago: Aldine.
Klein, Alan (1993). Little big men: Bodybuilding subculture and gender construction. Albany, NY: SUNY Press.
Olivier, Steve (1995). Ethical Considerations in Human Movement Research. Quest 47, 135-143.
Olivier, Steve (2002). Ethics review of research projects involving human subjects. Quest 54, 194-204.
Olivier, Steve & Olivier, Amanda. (2001). Comprehension in the informed consent process. Sportscience, 5(3), http://www.sportsci.org.
Peled, Einat & Leichtentritt, Ronit (2002). The ethics of qualitative social work research. Qualitative Social Work, 1(2), 145-169.
Pronger, Brian (1990). Gay jocks: a phenomenology of gay men in athletics. In Michael Messner & Don Sabo (Eds.), Sport men and the gender order (pp.141-152). Champaign, IL: Human Kinetics.
Punch, Maurice (1998). Politics and ethics in qualitative research. In Norman Denzin & Yvonna Lincoln (Eds.), The landscape of qualitative research: Theories and Issues (pp.156-184). London: Sage.
Ramcharan, Paul & Cutcliffe, John (2001). Judging the ethics of qualitative research: considering the "ethics as process" model. Health and Social Care in the Community, 9(6), 358-366.
Sparkes, Andrew C. (1992). The paradigms debate: an extended review and a celebration of difference. In Andrew Sparkes (Ed.), Research in Physical Education and Sport: Exploring Alternative Visions (pp.9-60). London: Falmer Press.
Sparkes, Andrew C. (1996). The Fatal Flaw: a Narrative of the Fragile Body Self. Qualitative Inquiry 2, 463-494.
Swain, John; Heyman, Bob & Gillman, Maureen (1998). Public research, private concerns: ethical issues in the use of open-ended interviews with people who have learning difficulties. Disability and Society, 13(1), 21-36.
Whyte, Willian-Foote (1943). Street Corner Society: The social structure of an Italian slum. Chicago: University of Chicago Press.
Zelaznik, Howard (1993). Ethical issues in conducting and reporting research: a response to Matt, Safrit, and Kroll. Quest, 45(1), 62-68.
Dr Steve C. OLIVIER
Research interests: research ethics, human performance
Dr SC Olivier
Head of Sport Sciences
Wynne Jones Centre
Newcastle upon Tyne
NE1 8ST, UK
Dr Lesley FISHWICK
Research interests: identity, culture, sports ethnography
Dr L Fishwick
Wynne Jones Centre
Newcastle upon Tyne
NE1 8ST, UK
Olivier, Steve & Fishwick, Lesley (2003). Qualitative Research in Sport Sciences: Is the Biomedical Ethics Model Applicable? [49 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 4(1), Art. 12, http://nbn-resolving.de/urn:nbn:de:0114-fqs0301121.