Issues of reliability and validity are often ignored within qualitative research (1) because they are seen as being aligned with a quantitative approach and therefore impossible to achieve within qualitative inquiry (2). Furthermore, there are different opinions about how qualitative data can be validated.

Field and Morse (3) define validity in qualitative research as 'the extent to which the research findings represent reality' (p139).  Silverman (4) defines validity as 'another word for truth' (p 175).  Silverman (4) argues further that issues of validity and reliability are just as crucial in qualitative research, in order to avoid the problem of 'anecdotalism':

How are they to convince themselves and their audience that their findings are genuinely based on critical investigation of all their data and do not depend on a few well-chosen 'examples'? (P 176).

In response to this problem, 'triangulation' can prove useful (4,5).  A term derived from navigation (6) it describes the process of looking at the same phenomena by using various different methods such as interviews and observations, or looking at the same phenomena from different points of view.  In establishing validity, researchers generally use all or a combination of the following types of triangulation:

·        Data triangulation – multiple data sources.

·        Investigator triangulation - multiple observations.

·        Theory triangulation - multiple theoretical perspectives.

·        Methodological triangulation - multiple methods. 

Triangulation as a form of validation, however, has been critiqued as not necessarily giving a full picture of the phenomena being studied (7).  Primarily, this is because the aggregation of data or data sources can be problematic (8) and does not necessarily produce a complete picture, as Hammersley and Atkinson (9) offer: 'differences sets or types of data may be just as important and illuminating...' (P 200). 

Other means are available by which we can judge the quality of qualitative inquiry.  Lincoln and Guba for example, (2) developed specific procedures, which they considered more appropriate in qualitative studies than traditional methods of validation.  'Trustworthiness' refers to the quality of an investigation as judged by four criteria: credibility, transferability, dependability and confirmability.   

Credibility in qualitative inquiry is parallel to internal validity in quantitative studies.  This criterion can be met by providing assurances of the fit between participants' views of their experiences and the researcher's reconstructions and representations (8).  Often the iterative and inductive-deductive nature of any given qualitative research strategy can provide assurances of the fit between the raw data and what emerges as the research findings. 

'Transferability' is parallel to external validity in quantitative studies and describes the extent to which the research findings can be applied to other situations or settings (3).  To allow readers to evaluate the transferability of a study the researcher must provide a thick description of the research findings.

'Dependability' refers to the researcher's responsibility for ensuring the research process was logical, traceable and documented, whilst 'confirmability' is concerned with establishing that the data and subsequent interpretations are 'not merely figments of the inquirer's imagination' (8, p259). 

Auditing is a useful procedure for establishing both dependability and confirmability (2).  In all research an audit trail of the research process should be established.  An audit trail consists of tools for data collection, the raw data, personal notes, memos, and documented procedures for analysing the data and generating theory.  

Another process called 'reflexivity' can assist in establishing an audit trail.  Reflexivity is concerned with the researcher's responses to the various stages of the research process (6).  Reflexivity is a form of self-monitoring, a form of data analysis, and a further method of establishing credibility by showing others, that your interpretations of the data are reasonable. Qualitative researchers often keep a field diary and file any notes, hunches and ideas that arise during data collection.  These activities contribute to the establishment of the audit trail. 

Finally, the iterative debate between the data and the researcher's interpretations ensures overall trustworthiness in the final analysis.  The researcher is in the powerful position of interpreting the data in whatever way he/she wants and in accordance with what matters to them.  It is therefore important for the researcher to provide an explicit account of data collection, analysis and presentation (10). 

References

1. Benton D.C. (1996). 'Grounded Theory' in Cormack D. (ed 1996) The Research Process in Nursing. Third Edition. Oxford: Blackwell Science. 
2. Lincoln Y. S. & Guba E. (1985).  Naturalistic Enquiry.  Sage, London
3. Field P. & Morse J. (1985).  Nursing Research: The Application of Qualitative Approaches.  London, Chapman & Hall.
4. Silverman D. 'analysing talk and text' in Denzin N. K. & Lincoln Y. S. eds. (2000) 821-834.  Handbook of Qualitative Research.  London: Sage.
5. Denzin (1978) The research act: a theoretical introduction to sociological methods.  (2^nd ed.) New York: McGraw-Hill.
6. Sapsford R. & Abbott P. (1992).  Research Methods for Nurses and The   Caring Professions.  Buckingham, Open University Press.
7. Silverman D. (1993). Interpreting Qualitative Data: Methods for Analysing Talk, Text and Interaction. London: Sage
8. Schwandt T. A. (2001).  Dictionary of qualitative inquiry, second edition. London, Sage. 
9. Hammersley, M. & Atkinson, P.  Ethnography: Principles in Practice. London:Tavistock, 1983.
10. Glaser B. & Strauss A. (1967). The Discovery of Grounded Theory : Strategies for Qualitative Research. New York: Aldine De Gruyter.