I find the whole idea of 'checking' results with participants a bit strange because it implies that participants have an answer my research question against which my own answer can be 'checked'. This is most certainly not the case. Feedback by participants on results creates additional data that need to be analysed, nothing more (but also nothing less, i.e. getting additional data is always great).

Margaret - You can deal with member check conflicts in qualitative research through the method of triangulation. In qualitative research, a member check, also known as informant feedback or respondent validation, is a technique used by researchers to help improve the accuracy, credibility, validity, and transfer-ability of a study. The aim of triangulation is to obtain accuracy in measurement between two points for which you require a way of measuring that is reliable. Validity and reliability concerns in triangulation need thorough and careful explanations because the objectivity of qualitative research is evaluated in terms of the reliability and validity of its observations. By reliability is meant the extent to which a measurement procedure yields the same answer however, and whenever it is carried out. Validity is the extent to which it gives the correct answer. This can be done through consistency and precision: It is important to obtain consistently similar responses to the same questions in similar circumstances. In precision, one has to know how systematic is the form of data that is based on asking people questions about things that they know little about.

Dr Manzoor Hussain

Thank you so much for your answer!

Can you please clarify further by giving me an example of how to use triangulation to deal with one or more participants if they disagreed with one of the themes?

To give you some context, I interviewed 24 professionals in the health care field and discussed with them their experiences and views regarding their appointments with a certain population of patients.

Many thanks :)

I think the most important things in member checking are not only to get agreement, but also for you to respond reflexively to any disagreements as part of your analysis process. This means defining the points of disagreement in ways that you can go back into your data and systematically examine them. Also, during the member checking, make sure that you get as much depth and detail as you can on any disagreements, so that you can use that information as effectively as possible when you question your own tentative conclusions.

I find the whole idea of 'checking' results with participants a bit strange because it implies that participants have an answer my research question against which my own answer can be 'checked'. This is most certainly not the case. Feedback by participants on results creates additional data that need to be analysed, nothing more (but also nothing less, i.e. getting additional data is always great).

Margaret - In triangulation, three are three types of "Reliability" which can be used: 1). "Quixotic" reliability: This refers to the circumstances in which a single method of observation yields the same measurement over and over again. 2). "Diachronic" reliability: This refers to the stability of an observation over time. 3). "Synchronic"' reliability: This refers to the similarity of observations within the same time period, which can be evaluated by comparisons of the same data by different methods. An example of the use of "Quixotic" reliability: Suppose a study is conducted on gender discrimination, and the subjects are asked the question "Do you believe in the equality of men and women?" Unfailingly, the answer obtained is "Yes". However, the reality observed around us is actually quite different. We may then conclude that the finding has only "quixotic reliability", because people are giving the answer they think is "correct", because they do not wish to offend anyone. So, it is probably a good idea to ask a different kind of question, like, "Do you think that women professionals are as competent as their male colleagues?" or "Would you like to work under a female boss". Perhaps the answers to these questions will be more varied and reflect reality better.

Jochen Glaser, you are not the only one with doubts about member checking. Morse (2015) "Critical analysis of strategies for determining rigor in qualitative inquiry," is another source on this topic.

Assuming you are doing more than simply summarising the data at a superficial level (which might be ok for some purposes), member-checking is not likely to identify a big issue. Remember, your participants (probably) don't have access to all the data or time to analyse it and your findings are your own: your analysis - your findings - your interpretation. Member-checking is however, useful. Apparently minor 'disaggreements' would help you refine or eleborate your thinking. If there were bigger inconsistencies this might suggest you did not collect sufficient data in the first place - i.e. you did not get saturation, in which case you may need to gather more data.

How member checks are used, and valued, depends upon your philosophical stance as a researcher that you disclosed (I assume) at the outset of your study. More post-positivist oriented researchers use members checks more as a procedure to be "checked off" to show participants degree of agreement/ disagreement. In these case transparency is essential to communicate to the reader where agreement/disagreement is present. I agree w/ David Morgan as member checking is not necessarily done "after" data analysis, but it is an essential component of data analysis. For more constructivist/post mod/post structuralist/participatory action-oriented researchers, member checks are active periods of data analysis viewing participants as co-researchers, not simply a relatively "passive" component of triangulation. My summary statement would be: do member checking in a way that is congruent with your philosophical stance as a researcher that you, hopefully, made transparent during the proposal.

Member checking can be applied in qualitative research with the understanding that you are the primary researcher and, therefore, you make the final determination regarding what makes the final cut in your findings/results and in your manuscript. In your role as primary researcher, you are the only person who has been involved in every step of your study since its inception. Information regarding how data that will be included/excluded should be disclosed to all co-researchers/participants upfront (beginning with your first contact) so there are no conflicts. If you have conducted the research fairly and allowed each co-researcher/participant an opportunity to fully express her-/himself, member checking has already been assumed within the process of inquiry, i.e., during your interviews. Post data collection member checking becomes, at that point, a formality. One final point: not all themes will emerge across the board with all of your co-researchers/participants. It is not uncommon in qualitative research for there to be "unanticipated themes" or "unique themes". These speak to the uniqueness of each co-researcher and can be included in the results with clarification that, while they were not prevalent, they are still significant because they were part of one or more of your co-researchers' rich experience (Sultan, 2018).

For content analysis, the use of NVivo software is very useful and effective. However you will need training to learn how to work with the program.

Member checking is indeed a powerful tool in promoting the trustworthiness of your study, but it needs to be done the right way. Asking participants to comment on your interpretations to "get agreement" is not the purpose of this practice. Instead, asking participants to comment on your interpretations allows them to make sure that what you've abstracted through your analysis is faithful to what they believe. If disagreement occurs, great! Now you have motivation to return to your data with a new perspective toward understanding your analysis. Keep in mind, too, that member checking does not need to occur once the final research report is concluded. You can ask members to read over your typed transcripts to see if they find any discrepancies, or you can ask them about your initial interpretations before you move to deeper analysis. What's most important, though, is that you present this process with transparency within your final report. The more open you are about your process--even admitting when member checks found your interpretations unsound--will lend credibility to your work.

Check this article, I guess it is very helpful for your case!

Member Checking: A Tool to Enhance Trustworthiness or Merely a Nod to Validation?

Article Member Checking: A Tool to Enhance Trustworthiness or Merely...