- internal validity, which depends on the strength of the relation between cause and effect
- external validity, which refers to the possibility to generalize findings.
Validity is a complex issue in qualitative research. Traditional standards used in quantitative research cannot easily be applied. If there is no truth or only one that cannot be reached, and knowledge is subjective, criteria for validity can only be very generic or/and subjective. Lincoln and Guba (1985) adapted empiricist criteria for qualitative research (See http://www.socialresearchmethods.net/kb/qualval.php for a detailed description of these criteria)
- internal validity translated as credibility. This has partly to do with being systematic. It is important to have clear descriptions of processes used, as well as of the reflexivity applied.
- external validity as transferability. Generalisability is not easy in qualitative research. It requires strategic planning – research project needs to be framed accordingly and processes made visible to convince.
- reliability became dependability
- objectivity became confirmability
Later the concepts of authenticity and morality (Angen 2000) were also added as validity criteria
- Real purpose of the research is clear
- Research serves participants
- Gives voice to, empowers participants
- Reciprocity : what do we give back to community?
- Respect of relationship
Morality or ethical validity
This refers to the recognition that the choices we make through the research process have political and ethical consideration
- Is research helpful to target population?
- Are there alternative explanations?
- Did we really learn something?
Post structuralism considers the need for validity criteria itself as flawed
Concepts and ideas which are important when considering validity
- Etic (a researcher taking an outsider position) and emic (insider)Methodology as ideographic (research aims to go inside object of research, about individual experience) or nomoethic(researcher remains objective)
- Reflectivity: looking back, what other possibilities exist
- Reflexivity: looking inside – crucial in ensuring awareness of how subjectivity impacts on research.
- Reliability: is there a clear process – does the research conform to methodological expectations
- Standpoint position or epistemic privilege: refers to the idea that certain researchers are in an epistemologically advantageous position and therefore likely to generate less partial research. For instance women are in a better position to research women issues than men. Standpoint position is not unchallenged (see Rolin, 2006 for more on this topic)
Possible validity checks at different stages of research
You need to make clear to the reader of your reserach that you have been systematica and at all stages considered
- The relation between researcher(s) and participants
- How to substantiate what is done
At level of data collection
- Sampling strategy
- Participants – researcher relationship. Researcher(s) need to take responsibility for the relationship established with participants, need to be reflexive and transparent about this. Is distance between researcher and participants necessary, what does it guarantee? distance?
- Carrying out research in a respectful manner
- Prolonged engagement with participants (insures credibility)
- Record keeping (standardization of fieldnotes, recording, transcribing)
- Refer back to sources to let them read what is recorded
- Collaborative work – cross checking
At level of analysis
This is about demonstrating that your interpretation is valid.
- Methodological strategies for producing more credible or rigorous qualitative research
- Awareness of choices made and taking responsibility for these choices
- Triangulation: refers to the use of different methods to investigate the same phenomena – allows to judge efficacy or validity of methods and sources used. But triangulation implies a belief in objectivity and truth and may not always be useful or coherent with an approach undertaken. Different ways to measure may show things from a different angle
- Check validity with participants – give epistemological privilege to others. Who judges validity? In some cases it is possible to go back to people who provided the information or to a similar group of people for feedback
- Negative or deviant case analysis
- Check alternative explanations
At level of discussion and conclusion
- Clear demonstration of how conclusion, interpretation was achieved. What is interpretation based on? How has data been woven together? Steps taken need to be continuously noted and justified, nothing should be taken for granted, logic of progression has to be explained. Enough evidence should be provide for the audience to judge how convincing it is (Mason 2006).
- Clearly indicate choices made
- Assess biases
- Self reflect
Validity of final work
- Peer review
- Dissemination of results can reflect quality
- External audits
Resources and references
Angen,M.J. (2000). ‘Evaluating interpretative inquiry: reviewing the validity debate and opening the dialogue’, Qual Health Research 10/3, 378 – 395.
Lincoln, Y and Guba, E (1985) (2nd ed). Naturalistic Inquiry. Newsbury Park : Sage.
Mason, J. (2006) (2nd ed). Qualitative researching. Thousand Oaks: Sage.
Maxwell, JA. (1992). ‘Understanding validity in qualitative research’. Harvard Educational Review. 62/3, 279-300.
Richards, L. (2009). Handling qualitative data: a practical guide. Thousand Oaks : Sage.
Rolin Kristina (2006). ‘The Bias Paradox in Feminist Standpoint Epistemology’, Episteme: A Journal of Social Epistemology 3/1, 125-136 available at http://muse.jhu.edu/journals/episteme/v003/3.1rolin.html
Web Centre for Social Research Methods (2006). Qualitative validity http://www.socialresearchmethods.net/kb/qualval.php
Winter, Glyn (2000), ‘A Comparative Discussion of the Notion of ‘Validity’ in Qualitative and Quantitative Research’. The Qualitative Report 4/3&4, available at http://www.nova.edu/ssss/QR/QR4-3/winter.html