Dear Deborah,

discussions about qualitative researches and their "truthfulness" are some kind of never-ending story so there is no definite answer to your question, on the contrary. First of all, I think you are asking incorrectly - is it really possible to verify an interpretation? I don't think so as interpretation is an explanation of researcher of what (s)he has found out, if someone else did the research or even had the same data, (s)he would probably pose other interpretation, another person would explain everything differently again, and so on. Whose interpretation is then the "right" one? Well, the one which has the most elaborated argumentation and methodology, in other words, the one which seems the most valid. For increasing the validity, there are some general suggestions.

(1) Make your research transparent - characterize how did you approaches the issue and how did you realized the research. You should state the characteristics of your respondents (social position, age, gender, and other), how did you choose the respondents (and why this population and not some other), place where the interview took place, and how did you analyzed the data. For semi-structured interviews, you should state the general topics of your interviews as well as the methods of sampling (purposive, multi-site, snowballing etc.), it depends on your research whether you want the sample (or population) more tight and restrictive or more broad and open (both approaches increase and decrease the validity and they are dependent on the researched issue).

(2) Contextualize your research into existing discourse (or point out that there is no relevant discourse) - use other researches (if there are any) and use other sources of data (documents, media representations, statistics) as instruments to support your interpretation.

(3) Keep in mind that qualitative research does not produce "truth" that can be verified, but an one possible understanding of the issue, as qualitative research is based on the social constructivist approach so you never have the complete understanding of the issue or know the truth about the issue. That's why you cannot make generalizations out of one qualitative study (you can but its validity would be very disputable). Generalizations can be made only on the basis of the outcomes of several studies - diversity of the approaches and interpretations serve as the vehicle of discussions and efforts to actually really understand the problem.

(4) You can use so called mixed methods - do part of your research qualitatively and part quantitatively, or analyze the issue (research questions) by qualitative as well as quantitative methodology and then compare the results and outcomes. That is very demanding type of research and usually it takes to make a research team to do the research.

(5) You can consider doing the research in a team as more heads know more. More opinions, interpretations, and views within the team increase the validity of the research (you discuss the data and themes between each other so that you can correct your assumptions).

(6) Validity of your research simply depends on persons who evaluate it - some people are more sympathetic to qualitative researches and know the things I wrote in the third point, some people are very strict and they are applying the verification procedures from quantitative researches on qualitative ones (although it's complete nonsense). In my experience, this situation is very often a dead end since you cannot convince the strict people about the contribution of your research, no matter how good your argumentation is.

I hope that something will prove somewhat useful to you, there si a lot more to discuss, that's for sure, so if you are interested I can recommend some interesting methodological books and texts.

Hi Deborah,

I'm a bit confused. Why would you seek a method to verify your findings of an existing qualitative study? They should be unique that do not require verification. From my experience, the only reason that you would want to 'verify' qualitative findings is within a mixed methods study. Even then, it's not verification - it is 'elaboration' and/or 'extension'. 

I think you can use focus group also or questionnaire survey where you ask them to provide their level of agreement with your finding in each case. This may be called result validation or verification.

Dear Deborah,

discussions about qualitative researches and their "truthfulness" are some kind of never-ending story so there is no definite answer to your question, on the contrary. First of all, I think you are asking incorrectly - is it really possible to verify an interpretation? I don't think so as interpretation is an explanation of researcher of what (s)he has found out, if someone else did the research or even had the same data, (s)he would probably pose other interpretation, another person would explain everything differently again, and so on. Whose interpretation is then the "right" one? Well, the one which has the most elaborated argumentation and methodology, in other words, the one which seems the most valid. For increasing the validity, there are some general suggestions.

(1) Make your research transparent - characterize how did you approaches the issue and how did you realized the research. You should state the characteristics of your respondents (social position, age, gender, and other), how did you choose the respondents (and why this population and not some other), place where the interview took place, and how did you analyzed the data. For semi-structured interviews, you should state the general topics of your interviews as well as the methods of sampling (purposive, multi-site, snowballing etc.), it depends on your research whether you want the sample (or population) more tight and restrictive or more broad and open (both approaches increase and decrease the validity and they are dependent on the researched issue).

(2) Contextualize your research into existing discourse (or point out that there is no relevant discourse) - use other researches (if there are any) and use other sources of data (documents, media representations, statistics) as instruments to support your interpretation.

(3) Keep in mind that qualitative research does not produce "truth" that can be verified, but an one possible understanding of the issue, as qualitative research is based on the social constructivist approach so you never have the complete understanding of the issue or know the truth about the issue. That's why you cannot make generalizations out of one qualitative study (you can but its validity would be very disputable). Generalizations can be made only on the basis of the outcomes of several studies - diversity of the approaches and interpretations serve as the vehicle of discussions and efforts to actually really understand the problem.

(4) You can use so called mixed methods - do part of your research qualitatively and part quantitatively, or analyze the issue (research questions) by qualitative as well as quantitative methodology and then compare the results and outcomes. That is very demanding type of research and usually it takes to make a research team to do the research.

(5) You can consider doing the research in a team as more heads know more. More opinions, interpretations, and views within the team increase the validity of the research (you discuss the data and themes between each other so that you can correct your assumptions).

(6) Validity of your research simply depends on persons who evaluate it - some people are more sympathetic to qualitative researches and know the things I wrote in the third point, some people are very strict and they are applying the verification procedures from quantitative researches on qualitative ones (although it's complete nonsense). In my experience, this situation is very often a dead end since you cannot convince the strict people about the contribution of your research, no matter how good your argumentation is.

I hope that something will prove somewhat useful to you, there si a lot more to discuss, that's for sure, so if you are interested I can recommend some interesting methodological books and texts.

Dear Debora, working with a hermeneutic approach we introduce what we call of “hermeneutic groups” as a second round of conversation. In this second round, we read to the same participants of the first focus group the narrative we have constructed based on the transcription of the first focal group. We did this in a participatory research context, looking forward to qualify and validate the narratives and having some intervention effects (mostly deepening the comprehension of the studied issue). I hope it could help. See our paper Rev Saúde Pública 2009;43(Supl. 1):16-22

Going back to your participants to hear their reactions to your Results is frequently called "Member Checking." This idea was developed by Lincoln & Guba in their 1985 book Naturalistic Inquiry. They treat it as a way to enhance the "credibility" of qualitative research (as opposed to terms from quantitative research, such as validity or verification).

Still, the whole concept does raise issues about wether you or your participants are in charge of evaluating your analyses, especially when they disagree with your conclusions. My own feeling is that when these kinds of disagreements occur, you do not need to accept their views automatically. Instead, you should acknowledge their perspective and compare it to your own.

Hello

I am not an expert in this method but it sound bad.

When someone is telling you his story by his eyes, feeling and outlook, you can not come back to him in order to verify his story, or parts of it.

the story of experiences is a subjective one, and it sometimes includs many defence mechanism or refraiming of relity.

Dear Deborah.  First, as a former professor, I tell you there is no such thing as asking  incorrectly. Second, qualitative, quantitative, interpretation, truth, etc., there is so much written on all of these and we need not be purists about any of it. 

Are you seeking to verify findings from another study and not your own?

If so, maybe what you want to do is to conduct research on some topic that is not well-known (there is not extensive research already conducted, a "new" interest/problem/issue, or is an in-depth study ) And you have found an interpretive research study (this is very specific, as not all qualitative research is interpretive) that used semi structured interviews, that were interpreted and published as "findings." (I apologize if I am wrong, here).

You want to use that research and its interpretations/findings to extend the knowledge on that particular topic by "verifying" (that word might throw off some people).  If this is so, bravo for you, for after all that is how knowledge grows and develops and what the purpose is of research.

Study the findings very carefully, even as they are indeed from interpretive research. It will probably not be possible to examine everything, so you must find a focus.  Select /extract factors that are of special interest to you or that you have reason(s) to suspect may be "accurate" (again for lack of a better term).  The semi structured interviews will probably contain some of these factors. 

Then I would recommend that you used mixed methods for your research to "verify" the findings.

I sure hope this  helps because I did make some assumptions here.

MHC

Many, many thanks for these comments covering some issues I was aware of and others not. The term 'verification' may have been a poor choice of word. My PhD study has been conducted 'aboard' and because I have not spent a extensive amount of time in the field I have been asked to 'confirm' or at least test my findings with the research participants. Member checking may be better term although I'm aware it raises a number of other issues such as participants having a mind change! Presenting a summary of my findings sounds practical and I take on board Mary-Helen's suggestion of ensuring the focus is useful.

Deborah, My humblest apologies for misinterpreting your question.  It is I, myself who misunderstood the "verify" in your question.  It is indeed used the way you state it and refers NOT to replication (as I thought) but to credibility of your conclusions/findings in your qualitative work. 

This is usually a methodological issue and there are several ways to "verify" your findings to give them more credibility.  My favorite source for this is Miles and Huberman, the most recent edition (other probably have their own). You probably do not need to re interview the participants, but if you did, it would be for clarification or elaboration.

Are there are other issues beyond verifying your conclusions? Did your advisor(s) give you  more details as useful feedback? Was rigor (or lack of ) mentioned?

Here's a couple of articles that might help:

http://edis.ifas.ufl.edu/fy394

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1120242/

Many thanks for all these sterling ideas and useful reminders.

In the event I returned to the Netherlands and conducted a Open Invitation Research Summary which included previous participants. There are lots of 'pros' and 'cons' of this approach, but it did give me some more useful data  which I have woven into my discussion chapter. Now I just need to complete my final draft.

I disagree strongly with the use of the word 'truth"  in science research, be it qualitative or quantitative. Not to mention that it is not used in the question. In science "verify findings" does not mean to find truth and no research produces truth. (Science is not / is not about truth). Science is about knowledge, acquitting and developing, discovering new knowledge, thus it is hardly ever fixed of never-changing.  Thus, it is not "truth" that is being verified when findings are being verified -- in qual or qunt research, a researcher is looking to confirm findings usually more than just once or twice and suppodt the conclusions.  Also, qualitative findings are more than mere interpretations and many times quantitative research also deals with interpreting findings.

This is not trivial, use of the appropriate word, especially in scientific research is very important.