22.214.171.124 Plausibility One view sees validity, in phenomenological research as plausibility. Hancock (2002), argued that validity is embodied in the plausibility of the relationship between data and concepts. Potential audiences have to be convinced that the interpretations of the data are compelling and convincing. Such plausibility is aided by consistency of interpretation by different researchers, and the systematic presentation of data to convince the reader.
Neuman (2006) also argues that a phenomenological approach to validity focuses on plausibility and the accumulation and interconnectedness of evidence.
Plausible means that the data and statements about it are not exclusive; they are not the only possible claims nor are they exact accounts of the one truth in the world. This does not make them inventions or arbitrary. Instead, they are powerful, persuasive descriptions that reveal a researcher's genuine experience with the empirical data. (Neuman, 2006, p. 197)
Neuman further asserts that:
Valididty arises out of the cumulative impact of hundreds of small diverse details that only together create a heavy weight of evidence. Validity grows as a researcher recognizes a dense connectivity in disparate details. It grows with the creation of a web of dynamic connections across diverse realms and not only with the number of specifics that are connected. (Neuman, 2006, p. 197)
Plausibility is enhanced if the study is compared to existing literature.
Another aspect of plausibility is theoretical validity.
126.96.36.199.1 Theoretical validity
Theoretical validity is about whether the theory that is based on the interpretations is feasible and plausible. Theoretical development usually involves analysis, critique, lateral thinking or synthesis. It is rarely a simple deduction from premises, if it is it is likely to be a simplistic theory. Becker, for example,did not come up with the labelling theory of deviance just by observing a lot of deviant acts.
Maxwell (1992) claims that theoretical 'validity' goes beyond the concrete and descriptive and concerns itself with the constructions that researchers apply to, or develop, during the research. However, again, this doesn't take into account the iterative approach of data, interpretation and theory that is characteristic of phenomenological research (see Section 2.3). What this also does is require a reflexive attitude on the part of the researcher in examining their own epistemological, ontological and grand theoretical standpoints.
188.8.131.52 Credibility Credibility is another way of projecting validity and is similar to plausibility. It focuses on making clear links between data and analysis so that the reader can recreate the line of reasoning.
Credibility is also enhanced if the data, for example, is coded by more than one person and interpretations are double-checked. (see Section 8.3.11).
An aspect of credibility is the validity of the interpretation of the data.
184.108.40.206.1 Interpretive validity (authenticity) Interpretive validity is about ensuring the interpretation is credible, reflects the data and the interpretive process is transparent, that is the process of arriving at the interpretation is clearly set out in any report.
However, interpretation is an inextricable part of the data collection process in phenomenological research. The process tends to be ongoing and interpretation at every stage informs the subsequent data collection. Thus, one should not divorce interpretive from descriptive validity (see below).
One view suggests that the interpretation needs to be tested by referenced to the research subjects. Maxwell (1992, p. 290) proposes that to be valid an interpretation must be consistent with the views of the research subjects, in short the research subjects should be able to recognise or confirm the findings of the research.
Neumann also endorses a similar approach, which he refers to as authenticity. For Neumann (2006, p. 197), authenticity 'means giving a fair, honest, and balanced account of social life from the viewpoint of someone who lives it every day'. Authenticity does not necessarily require subject confirmation, the researcher could argue that the perspective presented represents the lived experience of the subjects but would need to be convincing in making the case for authenticity.
The problem with privileging the interpretation of the research subjects is that it is far from clear than any individual has a more valid interpretation of their own actions than might be made by another person.
220.127.116.11 Trustworthiness Similar to plausibility and credibility is the notion of trustworthiness. This requires the researcher to demonstrate that their research can be trusted.
Golafshani (2003, p. 602), for example, proposes that validity is 'more appropriately described in terms of rigour, quality and trustworthiness'. Aldridge (1996, p. 125) suggests that 'the basis of establishing trustworthiness . . . is to show that the work is well grounded, (and) to make transparent the premises that are being used'.
An important element of trustworthiness is descriptive validity.
18.104.22.168.1 Descriptive validity
Descriptive validity is primarily about the data collection stage of the research. According to Maxwell, (1992, p. 287) the key issue is the accuracy of the statements describing what has been observed (remember observation covers hearing, feeling as well as seeing).
If different observers or the use of different methods result in different descriptions of the same events or situations, this appears to raise concerns about validity. However, contrary to positivist assumptions, phenomenologists recognise that research into the personal lives and experiences of people will result in contradictory accounts because there is not a single objective reality. There is nothing wrong with two different accounts that give different weights to the same events.
Validity is compromised though if some accounts are sloppy, contain errors, such as the number of people in a group at a given time, or are otherwise indicative of a lack of concentration or focus on the part of the researcher.
Genuinely different takes on the same situation are not 'errors' but research material in themselves and it is not appropriate to clean-up accounts to remove contradictions in 'genuine' descriptions. Any such 'clean up' would itself be a subjective process with no valid basis.
22.214.171.124 Techniques for establishing validity in phenomenological research There are several ways to encourage trust and credence in phenomenological research. These include époché, peer critique, structure resonance, participant verification and triangulation.
126.96.36.199.1 Epoché (or bracketing)
Epoché is the process of bracketing the researchers assumptions and taken-for-granted conceptions.
This is often taken, by phenomenological researchers to mean being self-critical and confronting their own biases and presuppositions about the subject matter of the research.
Grocke, (1999), for example regarded époche as 'the examination of bias of the researcher' to make explicit any assumptions or preconceptions of the researcher. 'The important distinction is that the researcher's biases should not negatively influence the interview process nor the analysis of data.'
(Epoché is actually much more than this; see section2.3.1 for a fuller discussion).
At a rather more substantive level, Kvale (1983, p 148) described bracketing as 'an attempt to place the commonsense and scientific foreknowledge about the phenomena into parentheses in order to arrive at the essence of the phenomena'.
Osborne (1990, p. 87) argues that by bracketing, the researcher provides the reader with an 'opportunity to understand his/her interpretations of the data'.
188.8.131.52.2 Peer critique (or debriefing)
A way to increase the plausibility and acceptability of the study is to use a peer group to critique the stages of the research, to check the descriptive validity but more importantly to comment on the transparency of the interpretation of data and theory building. This is also known as the 'juridical process', peers act as a jury that have to be convinced by coherent arguments. The jury must 'be able to follow the thought processes that have led to the conclusions and to accept them as valid' (Polkinghome, 1983, p. 57). Normally peers are other researchers or scholars. It is, though, not always easy to arrange such critiques and it helps if you have a ready group of informed peers to call on.
184.108.40.206.3 Structure resonance
Validity can also be attested by getting other people who have not participated in the study, but who have experienced the same phenomenon to comment on the findings. This is known as 'structure resonance': do the findings resonate with another similar group to the subjects? Although this may give credence, seeking approval from similar groups may also lead to rather conservative interpretations.
220.127.116.11.4 Participant verification
As noted above (descriptive validity) the participants in the research could offer a perspective on veracity and credibility of the interpretation. This procedure is also referred to as 'member checking' (Aigen, 1995;Creswell, 1998) or 'goodness of fit' (Osborne, 1990). Borimnejadet al. (2006, p. 339) used this approach in their study women suffering from vitiligo. They recontacted each participant in the study and gave her 'our actual manuscripts to make corrections or changes as she wished'.
One has to note, though, that is but one perspective and subjects' interpretations should not necessarily be privileged over anyone else's. It is unlikely, should they be asked, that all the participants would agree on the extent to which the research outcomes fit their view of reality.
18.104.22.168.5 Triangulation and validity
Validity is supposedly enhanced if the research is approached or verified from a number of different direction or sources. Using a variety methods of data collection, different researcher perspectives and different conceptual frames are all forms of what is called triangulation(see section 1.15), which give greater credence to research outcomes.
Kvale (1983), for example, is of the view that trustworthiness of phenomenological research is enhanced when several people conduct the interviews as this supposedly results in a broader and more richly nuanced picture of the themes.