CONCEPTS OF RIGOUR: WHEN METHODOLOGICAL, CLINICAL AND ETHICAL ISSUES INTERSECT

Aroni, R., Goeman, D., Stewart, K., Sawyer, S., Abramson, M., Thien, F., Douglass, J.

AQR CONFERENCE MELBOURNE JULY6-8 1999

A number of strategies can be employed in designing studies where the primary design is qualitative but quantitative data collection methods are also included. These include nesting, concurrent data collection and analysis and independent analyses and subsequent triangulation. Some researchers presume that there is a paradigmatic divide, a chasm that cannot and should not be crossed. This paper outlines the assumptions built into the data collection and analysis processes used in a multimethod, multidisciplinary and multiuniversity study. It critically examines the difficulties and strategies adopted in a study where clinical, methodological and ethical issues intersected. This paper provides a detailed account of analyses of quantitative data and qualitative data using NUD*IST, SPSS and global analysis to assist with data management and analysis and how rigour was maintained. Data reduction, retrieval and incorporation are discussed. A model for future use is suggested.

In order to examine the notion of rigour and its construction and use in studies that use qualitative methods this paper describes three instances where methodological, clinical and ethical issues intersected in an NHMRC funded study"A qualitative study of barriers to therapy adherence by adolescents with asthma".Rather than beginning with discussions of ontology, epistemology, methodology and methods or focussing on "purity and danger" we feel it is necessary to tell a story – provide a history of the process of the project and intersperse it with interpretations and discussions of concepts and issues of rigour . Following the telling of the tale we draw on its detail to raise broader concerns about rigour and qualitative research.

PREFACE TO THE STORY

Why discuss concepts of rigour?

Some will argue that if we are concerned with rigour it is because we are being drawn into positivist, reductionist modes of thought and are therefore not maintaining integrity with our own methodological positions. Selling our souls means that in the end we go to hell because eventually "the piper must be paid" and it will all show in what must inevitably be compromises in research practice. Others argue that the concern with rigour is a pragmatic one, which is really to do with gaining access to research grants and if the traditional criteria of rigour are what the granting body committees want - then that's what we will give them. What is more if we have to prostitute ourselves sufficiently to get our money to do the research we want and can keep them happy at the same time - then so what! There are those who suggest that "the two ways" of doing research are compatible and those who do not. Finally, there are those who would argue that qualitative research is a form of science just as valid as any other form since science is based on searching for findings which can be added to the common stock of knowledge. The certainty lies in the integrity of the researchers and their findings. Does the question then become the integrity rather than rigour ? Rigour is the means by which we attempt to show integrity and competence, so rigour is really about ethics and politics - as it has always been.

Intersections between methodological, clinical and ethical issues

Intersections between methodological, clinical and ethical issues are inherent in all forms of research, whether in the health field or not. Each substantive research domain (be it education, media, interpersonal relations or health) has its own version of clinical issues. This is especially so when multi-disciplinary research teams are involved. Inevitably in multi-disciplinary teams each member is professionally socialised with a different understanding of research and what constitutes ‘good’ science. Or they have operated with different theoretical positions within their disciplines but conform to models of research practice which have the status "gold standard" within their discipline boundaries (held by the gatekeepers in their discipline's publication arena).What happens when they believe it necessary to come together to investigate a specific research question?

BRIEF HISTORY OF DEVELOPMENT OF RESEARCH DESIGN

Rather than give an interpretation of a "full" history of the research process this paper highlights three occasions where rigour was at issue.The first involves research design and the funding process, the second, an issue of research ethics and the third, the nature of analysis.

The research team included "Chief Investigators" as per initiation of the project and funding body construction of ownership and responsibility.These were:

- a health sociologist with expertise in qualitative research and multi-method research.

- a pharmacist with expertise in pharmacy practice research.

- a paediatric respiratory physician with expertise in adolescent health; and

- a respiratory physician with expertise in epidemiology.

Also included in the research team were two associate investigators who were respiratory physicians with expertise in allergies and immunology and a research assistant with expertise in qualitative computing skills. Three other research assistants were employed to conduct interviews and they received training in in-depth interviewing of young people from the sociologist.

As with other such groups, this multidisciplinary team included individuals who had each been professionally socialised with understandings of what constituted "good" research practice and research skills and experiences. They also had specific knowledge and experience of funding bodies' methodological preferences. The disciplinary and skill differences enabled and enhanced the focus on clinical issues but also involved operating with somewhat different assumptions about the conduct of research, the pragmatics involved and the meanings attached to the use of more than one method. Goodwill, respect for each other's expertise, candour and a genuine desire to answer the research question enhanced the research process in this instance.

The central question of the research project emerged from clinical concerns.How do clinicians provide quality care if they believe that non-adherence places young people’s lives at risk and they can’t be convinced to adhere to medication? Why do adolescents not adhere to preventative asthma medication regimens, when currently such regimens are their best chance for staying alive and well? We know from a review of the literature that poor adherence to asthma medications exists. The commonsense understanding based on the bio-medical view of asthma is that this is potentially dangerous.We ask what can we do if adherence to preventative medication is the only way we know to decrease death rates among young people.To answer this we have to find out what young people do about their asthma and why they do it. Why do young people think that they don’t use their preventer medication? How can we best find this out from them? How much have we found out using different methods?Not enough because young people are still dying of asthma. If we find out why young people do or do not adhere - from their perspective - will that put clinicians in a position to provide better care?

Most studies focussing on adherence have been based on "checking" on actual medication practice to determine and describe the degree of adherence. These have included studies which rely on mechanical counting devices attached to inhalers to determine how regularly participants in the study actually used their inhalers; the use of survey questionnaires and the use of diaries . In the case of studies using mechanical devices, a great deal of thought was put into developing more sophisticated models which would indicate when participants were 'dumping' medication(in order not to provide evidence of their non-adherence) rather than actually taking it (Rand,1993). However, while these studies tell us about incidence of use, explanatory data as to why young people did or did not use their medication is missing.

These studies provided a backdrop which gave a numerical profile of individual medication use patterns. They were rigorous according to traditional canons of doing good science but they answered only the question of "how often", not the question of "why".The results of such studies while useful, did not provide clinicians with knowledge that would enable them to improve their care of young people with asthma. Nor did it provide young people or their families with knowledge that would assist them in managing their asthma. How do we find out why young people do not adhere to medication regimens and what meanings they attach to them?

The NHMRC Joint Working Party on Asthma had stated publicly that the most useful question to ask at that point, was why non-adherence by young people was continuing to occur. This was a recognition that the number of hospital admissions and the morbidity and mortality rates of asthma in these age groups was relatively high and knowledge that poor adherence occurs was not enough. Survey style studies had not produced information leading to effective interventions, clinical or otherwise. A qualitative study was suggested as the best research strategy to gain access to this knowledge.

The choice of study design is inevitably tied to the nature of the research question if it is valued as a "good" study design. The team recognised that if the working party of the NHMRC was positive about the idea of using a qualitative research design then funding might be forthcoming. In addition a small pilot study using qualitative methods had been successfully carried out under the supervision of one of the team members in the previous year. The team applied to the NHMRC for a research grant to carry out a large qualitative study examining the barriers to non-adherence from the perspective of young people, their parents and their clinicians.

Even though the team members who were trained in quantitative methods were committed to a qualitative study, they were nevertheless concerned about acceptance within their disciplines of results derived from qualitative research alone. They felt it was necessary to have some means of comparing results with previous quantitative studies in the field. This is a means by which we often construct one concept of rigour - comparability with previous research, except that in this instance, the research practices guiding the project were very different to previous studies in the field, which relied on quantitative research methods alone. The quantitative instruments, the Asthma Knowledge Questionnaire and the Respiratory Health Questionnaire were incorporated into the study to satisfy the theoretical, methodological and political dilemmas. In theoretical terms it was important to compare clinicians’ and young peoples’ responses to questions about asthma knowledge, respiratory health and symptom recognition and, even though the interviews would be useful in examining these issues, the questionnaires might also provide such information in a different form. In political terms they were seen as enhancing the chances of the study being funded. They felt that the funding body's committee might "be happier" with this inclusion (the Clayton's quantitative study). In the words of one team member "it would provide them with a feeling that a safety net was in place, you know." The issue of rigour was part of the discussion but it was also a discussion of politics and ethics.

If we wanted our results to be accepted into the medical corpus of knowledge about asthma and be perceived as valuable and useful knowledge by practising clinicians then it would have to be translated into a form that could be understood as credible and legitimate. The inclusion of questionnaires to provide additional data was perceived by the three investigators with quantitative research experience as an entree into publication acceptance in their fields and a comfort zone in terms of their own involvement with the project. This was something that they were familiar with and which could also be justified in terms of rigorous research practice. It could also "fit" in the research design without substantially altering the basic parameters of the project.

From the perspective of the investigator with qualitative research experience the consequences of inclusion of questionnaires in a research design which relied on qualitative methods would not impede the fundamental mode of data collection(in-depth interviewing) and could enhance the project in a number of ways. Firstly, it could provide material to assist at the beginning of the interview sessions with establishing a first discussion point(the so-called "preliminaries" that are dismissed in most research texts). Secondly, it would put the other members of the team more at ease with the whole project. Finally, the questionnaires may be useful in highlighting the strengths of qualitative data collection methods when compared with survey questionnaires. Surely this would be the case given her faith in the richness of data that could be elicited with the use of in-depth interviews! There was also considerable debate and subsequent agreement on the issues and themes to be included in the first in-depth interviews. Mutual education was the order of the day.

When news arrived that an NHMRC interview was to be granted, the medical researchers on the team who had experience with this process suggested a practice interview to improve our performance as a team at the grant interview. It was not an inspiring performance by the team member with expertise in qualitative research. This was her first experience of such interviews and she had entered without understanding the lack of shared underlying assumptions about research methods. She had been asked questions which involved very different assumptions about the nature of rigorous research and failed to answer them well.

It was at this point that another member of the team suggested the inclusion of a third questionnaire into the study design. This was categorically refused by the researcher with expertise in qualitative research as distorting the fundamental nature of the study. It was at this point we debated the nature of the study, the pragmatics of funding and the way in which methodology and ethics intersected. It became a matter of integrity of research design and maintaining "good faith" with it and each other. Perceptions of rigour by "powerful others" rather than the actual nature or criteria of rigour was raised several times in that discussion . The team members who had quantitative research experience still had faith in the research design but were concerned at the possibility that it would not present well to an interview panel that by reputation had no empathy for such research designs or practices. The funding body was known to have committee members who had been primarily trained in quantitative methods and had very little knowledge of, or experience of, evaluating qualitative research proposals.

Assumptions

What were the underlying assumptions and points subsequently raised in this debate about the data collection and analysis processes in this multi-method, multi-disciplinary study?

They were that:

a)we are essentially engaged in a qualitative study.

b)there are quantitative data collection processes which will not disrupt the qualitative data collection processes.

c)use of the quantitative processes will satisfy the comparability criterion of rigour

d)it is likely that these responses will be understood as information for clinical practice and clinical publications.

e)by inclusion of questionnaires we could examine quality of data drawn from both collection methods

f)data drawn from questionnaires will satisfy funding body concerns about actual outcome until they learn to trust qualitative methods

g)analysis of questionnaire data will illustrate its poverty

Qualitative research is typically seen by clinicians as lacking rigour and therefore being unscientific. This of course is the most damning thing one could say in an era when scientific knowledge is regarded as the highest form of knowing and in an arena where all practice is predicated on "being scientific".

‘The most commonly heard criticisms are, firstly, that qualitative research is merely an assembly of anecdote and personal impressions, strongly subject to researcher bias; secondly, it is argued that qualitative research lacks reproducibility- the research is so personal to the researcher that there is no guarantee that a different researcher would not come to radically different conclusions; and, finally, qualitative research is criticised for lacking generalisability. It is said that qualitative methods tend to generate large amounts of detailed information about a small number of settings.’ (Mays & Pope, 1996:10)

The fear was that it would not be possible to go beyond this perception of qualitative research as lacking rigour and then the funding would not occur. There are a number of strategies that can be employed in designing studies which incorporate multiple methods where the primary design is qualitative but quantitative data collection methods and analyses are also part of the overall research design. The various models of multi-method use depend on the ways in which the research question is conceptualised, who is engaged in the process and who the audiences are perceived to be.

The original application made it clear that the quantitative components were relevant and fitting but not the central platform of the study. Therefore, only the two questionnaires which were part of the original design were included and the project was fully funded on the basis of being a qualitative study. Perhaps it was understood that ‘One of the greatest methodological fallacies of the last half century in social research is that science is not a particular set of techniques; it is, rather, a state of mind, or attitude, and the organisational conditions which allow that attitude to be expressed.’ (Dingwall, 1992:161-75)

Qualitative researchers have also grown up with quantitative models of scientific method being the gold standard. However, they have either been professionally socialised to perceive its flaws as a provider of good science, or, they have personally experienced its inadequacies while attempting to adhere to the rules which made their work rigorous but still not ‘good’ science. The alternative and more enveloping model of "good" science requires different relationships with those being studied and broader notions of what constitutes authentic and valid data.

CLINICIAN AS RESEARCHER:RIGOUR AND ETHICS

When the initial proposal was submitted for institutional ethics committee review there was concern expressed by the institutional ethics committee situated in the education rather than the clinical settings that clinicians were involved in both recruitment and treatment and that this might compromise anonymity and patient clinician interaction. We had felt that this was adequately dealt with in our proposal but we gave a more detailed explanation of the strategies we had previously outlined. In addition, we had all been taught that rigour included paying attention to and being engaged in ethical research practice.

However, there were issues which arose during the study regarding the ethics of the practice of analysis with one of the researchers being the treating clinician of patients recruited. The issues of anonymity and confidentiality as well as the possible impact on subsequent clinical practise and outcomes for patients were all raised. There were disagreements about the most ethical way to proceed which would still retain sufficient critical analytical input. The use of a diarised audit trail enabled analysis of this situation as "data".

The ethical objections were raised in terms of maintaining anonymity and confidentiality. This protection for participants is standard practice in research and regarded not only as a matter of ethical obligation but also a matter of rigour or "proper" research practice. As analysis proceeded, one of the research assistants noticed in the analysis sessions that one of the chief investigators, a clinician, recognised their own patients. Even though the transcripts were identified by number some of the transcripts had initials. This was not only an ethical problem but a slip in operating procedure or rigour. Much discussion ensued which is detailed below. Subsequent to these discussions and after discussions with the institutional ethics committee concerned, deletion of these initials was deemed sufficient to maintain anonymity of the participants. The committee's concern was to prevent "purposeful" identification as opposed to "accidental" identification. The research assistant suggested that even so, anonymity would not be maintained because the topics discussed in the interview sessions would enable the clinician to recognise patients, which was indeed the case. Perhaps this clinician and the other clinicians on the team should not be involved in analysis of transcripts which involved those participants who were known to be that clinician's patients?

During these discussions some members of the team felt that their personal integrity was being questioned and that this was unnacceptable given that they had participated in multiple research projects, knew which of their patients had participated and were treating these patients as well as researching outcomes at the same time. This was regarded as perfectly ethical within the framework of that research structure. Different ethics protocols and different protocols of "standard practice" operating for different contexts were acknowledged, yet that did not assuage the perceived insults to both personal and professional integrity. While deletions of initials reduced the likelihood of patient recognition, knowledge of, or possible recognition of participants even without any obvious identifying features may occur, if you are dealing with a doctor who is involved in, and deeply committed to, patient care. In addition, it was argued that there was no guarantee that patients would be correctly identified from numbered transcripts. However, even with all the precautions what if they were? Various proposed solutions for this problem were suggested. It was pointed out that in the informed consent forms participants had signed it was made clear that the clinician was part of the research team and that as such transcripts would be available to the clinician.Nevertheless, the institutional ethics committee required that the consent form be amended to make clear the possibility of "accidental" identification.

This incident and the discussions that ensued raised a difficult issue for this team and other teams where some members are clinicians working in the field and research participants are also their patients. Ethical objections are understandable in theoretical terms but what about in pragmatic terms and in terms of establishing and maintaining rigour in the analytical processes of the project? This was a classical confrontation between ethics and rigour in research practice. Doing good analysis is tied to having greater understanding and as much relevant information as possible as long as we do no harm. Whether research methods employed are quantitative or qualitative the dilemmas are the same. Perhaps in this case, the nature of the data collection and analysis processes highlighted the concerns.

One of the rationales for constituting research teams from different disciplines is to provide expertise from a number of different fields in order to enhance the ability to respond to the research question adequately. If we "cull" transcripts from their gaze surely this will decrease their ability to provide full analytical input. This negatively influences the clinician’s ability to participate fully as a member of the analytical research team. After all the participation of the clinicians in the team was to do far more than give access to patients but aimed to utilise their specialist knowledge in the analytical processes and enhance sensitivity to clinical issues. It was to enhance the abilities of the other team members through the provision of analytic insight drawing from their experience of the impact of medication use and the impact of asthma as lived experience from the point of view of the clinical encounter. How the clinicians would understand and interpret the transcripts was a critical part of the process of analysis, as it was itself a part of the data (the investigators had been asked to keep a diary of the research process in order to examine the impact of working in a multidisciplinary team and to act as an audit trail). Typically such debates and concerns that occur in the daily doing of research, particularly in team research, are issues that are not discussed in the research publications. We believe that there is a rationale and a place for them to be discussed in a public way.

Audit trails as a mode of transparency have been advocated by qualitative researchers as a means of satisfying the theoretical, methodological and pragmatic needs for rigour by paralleling the conventions of quantitative research reporting of data and procedures even when using different concepts for the construction of qualitative research as "good science". According to Strauss and Corbin (1990:250) "the usual canons of ‘good science’ should be retained, but require redefinition in order to fit the realities of qualitative research." The canons of "good science" in quantitative research are usually listed as "significance, theory-observation, compatibility, generalisability, consistency, reproducibility, precision and verification." (Denzin, 1994:508). Qualitative researchers have opted for concepts such as credibility, dependability, transferability and confirmability as more fitting with the nature and meaning of qualitative research. Some refer to authenticity, trustworthiness and "fit". Transparency via the use of an audit trail enables assessment of the research against the latter criteria.

The need for rigour in qualitative research is driven by both theoretical and methodological needs as well as pragmatic ones. Transparency is provided by revealing the pattern of perceived researcher choices and decision-making which is usually rendered implicit in research reports. This includes ethical issues and choices made in that domain which are often hidden in research reports and are typically seen as separate to analysis but important in participant/researcher relations. The research diary database that was used in this project incorporates an ongoing commentary of the ethical dilemmas faced in terms of ethics, theoretical significance, methodological impact and pragmatic and political impact.

Therefore we suggest that if audit trails are used as a means of enhancing rigour that they include team management issues, team work and ethics as specific components in order to provide not just transparency for the external reader but also a methodological tool for enabling the researchers to enhance the research process throughout. The audit trail use of "running diaries" provides a means of dealing with the routinised aspects of the research process as data. It also provides a means of managing the contentious or difficult research processes as data in a less threatening manner. The transparency requirement enables the possibility of defusing conflict or increasing it. In our case it has enabled a multidisciplinary team to continue to work together with mutual respect and increased ability to analyse and interpret all the data bases with significant and relevant input from all the researchers.

SAMSON AND DELILAH

The third issue about rigour raised in this research project was the use of a computer assisted qualitative data analysis system. This was a practical and political issue. One of the political aspects was tied to advice she had been given some time previously. One reason for mentioning the use of computers in a research proposal was it might increase the chances for funding because it was assumed that the members of the funding body committees were familiar with the notion of using computers to facilitate analysis of data . This familiarity may ease their concerns about rigour in analysis in a research project using qualitative methods. Was this the reason the team was considering the use of such software packages? No. When we made sampling decisions at the design stage of the project it was clear that data management was an issue. This was to be a large qualitative study and the amount of data would be overwhelming without the use of computers.

Nevertheless, there was concern by the qualitative researcher (who had been socialised in the use of qualitative methods prior to the development of software packages specifically designed for this domain) about the possible impact on analytic processes. Would the use of such a package really decontextualise data. At a recent conference, Cherry Russell (1999) discussed the problem that is encountered in thematic analysis when we engage in conventional segment and code analytical strategies, that is the loss of the story telling of informants as one focus of analysis. She revisited data from a previous study to illustrate a different approach to analysis where she refered to reinstating story telling as the focus with exploration of the cultural and interactional functions of story telling and the structures and processes of the interviews themselves. She focussed on what she states is "The fundamental and central tenet of much qualitative research...its desire to include it all, to never decontextualise, to maintain reflexivity at all costs."

Yet we use conventional segment and code analytical strategies with or without the use of computers. So , again the issue is what decisions the researcher makes in interpretation of data. As St. John and Johnson (1999) pointed out in their discussion regarding the use of computers in qualitative data analysis - "..before we wholeheartedly embrace the clutches of technology or consign all things electronic to the bin it is important that we reflect on the practical and methodological impact and implications of its use. On the one hand, should we not grasp a technology that helps cope with mountains of qualitative data in a methodical way?....On the other hand, are we in danger of reducing our thick and rich data to the decontextualised fragments we have been so critical of in other types of research." They refer to the "philistines and the faithful" where those who use computer assisted qualitative data analysis systems have been labelled as the philistines because they are perceived as not recognising the ways in which they have defiled their data; and those who do not use them are viewed as the faithful because they have maintained a purity and not contaminated their analysis by decontextualising through the use of computer assisted qualitative data analysis systems. They, in effect, highlight the way in which rigour is constructed amongst researchers who label themselves as qualitative researchers but do not use computer assisted qualitative data analysis systems. This raises issues for the funding of qualitative research as it can be difficult to have your research proposal assessed by opposing groups in this debate just as much as being assessed by those who have no claims of knowledge in this methodological field.

More recently, when reporting on the results of the research project where qualitative researchers were in the audience, there was significant concern about the fact that such a package was used. There was moral outrage at the lack of rigour implicit in the use of such a package. The argument that was put was that these packages inevitably decontextualised your data and analysis would be all the poorer with their use. The issue of the use of computer assisted qualitative data analysis systems has been subsumed under a debate about rigour, usually with the refrain that you must recognise that this will not enhance rigour but is destructive of it. It has become a statement of religious membership for what we might label methodological fundamentalists.

To illustrate the point further, in the same discussion the sample size was then called into question. Why have such a large number of participants in a qualitative study. Wasn't this study really a quantitative one in disguise? After all, wasn't one of the assumptions about qualitative research that you had to have small samples and this enabled you to spend the time necessary to do "good" analysis. There was no recognition that this interpretation of what constitutes standard qualitative research may in fact be a way of limiting and denigrating its capacities to answer research questions.

All three areas of debate touch on defining rigour from different perspectives. Perhaps the most useful suggestion that we might put is that in all these contexts there is a need to translate into the language of the assessor, the listener,the reader or the team member. Perhaps "good" qualitative research involves analysis "that accounts successfully for the way in which individuals interact with their social context. The ultimate criterion of accuracy is the ability of the theory to fit existing knowledge in a coherent manner and to explain and predict phenomena in a convincing fashion - the ultimate test of any scientific theory" (Daly, McDonald andWillis, 1992) .

We compared our needs for research reporting when preparing to present our findings at different professional conferences and realised that the requirements for presentations although similar in instructional terms, differed substantially in terms of acceptable construction of content. This was anticipated during the design stage. The investigators were all asked to keep a personal and analytical log of their research experiences in a journal or research diary. This was done to enable comparisons, provide detailed accounts for academic audiences and to enable back-tracking to check the analytical processes and how they occurred.When discussing the construction of a brief poster presentation of the preliminary findings for clinicians we pointed out that it follows the reporting format of quantitative research and that it would still read as a denuded and overly linear account of a qualitative study if the audience were to be predominantly qualitative researchers. For them the authenticity would be submerged and yet for the clinicians it is beginning to emerge as authentic and valid.

This is all to do with the essentially political and moral question of who is the research for? and why is it being done? The answer is that if we still wish to perceive research findings as having more weight than rhetoric or polemic in a debate(which our sociologist team member believes is the reason most researchers lend themselves to action research or choose not to participate in research funded or controlled by those they personally regard as morally questionable or reprehensible) then we still do need to talk about rigour (also as integrity of practice)and we are operating with a false sense of self righteousness to pretend otherwise. It is to succumb to devaluing qualitative research to not do so.

Qualitative research methods, whichever version and whichever methodology, evolved predominantly from a critique of the narrowness of what we now call quantitative methods, of their inability to capture the richness, to go from the "ground up", to incorporate the perspectives of those being researched in their voices and with them. This was a criticism of a lack of rigour by those using quantitative methods based on a passionate desire for rigorous research, research that didn't leave it all out. Rigour is all about passion, the passion for doing "good" research. What is good research and why do it ? Good research is that which makes a difference in the world just by the reading of it. This paper is a plea for the doing of good science - morally worthy , a biblical injunction to not lie, to not steal, to not murder, the participants in the research process.

The reality of lived experience and research detailing it and interpreting it purports to tell it as it really is (alternate truths and perspectives)in contrast it is supposed with medical understandings.These forms of research are predicated on the notion that participation in biomedical discourse is not telling it as it is because it is oriented to different understandings and gives different meanings to how it is from the patients lived experience -different perspective, different truth. What happens when we do qualitative research and want those with biomedical perspectives to attach significance to the other perspective, the patients' view? We must make it translatable, that is make the language understandable. How do we do this ?

What dictionary do we use? The dictionary of scientific method of the qualitative kind, our version of what constitutes good science or the doing of research with integrity. Alternatively if we are looking for a way to translate concepts of rigour to those who speak different professional languages, then we would be best to recognise and make clear that, "..it is possible to develop practical standards-workable across different perspectives - for judging the goodness of conclusions..we are still accountable for the rationality and trustworthiness of our methods" (Miles and Huberman,1994)

REFERENCES

Daly,J.,McDonald,I., and Willlis,E.(ed) 1992 Researching Health Care:designs,dilemmas, disciplines. London:Routledge

Denzin,N. 1994 The art and politics of interpretation in Denzin N and Lincoln Y(ed) 1994 Handbook of Qualitative Research Thousand Oaks:Sage pp500-516

Miles and Huberman,1994 Qualitative Data Analsyis An expanded sourcebook (2nd edition)Newbury Park:Sage

Russell, C. 1999 Interviewing old people:interpretive possibilities in different approaches to qualitative data analysis 5th International Qualitative Health Research Conference 7-10 April University of Newcastle

St John W.and Johnson, P. Using computers in qualitative data analysis:Saviour or Sin 5th International Qualitative Health Research Conference 7-10 April University of Newcastle

Strauss,A.L and Corbin, J. 1990 Basics of Qualitative Research:grounded theory procedures and techniques Newbury Park:Sage