Yes, it is always necessary to be confidential to protect the interest of the respondents, but you need to specify this clearly in the interview protocol and made it understandable by the readers why you are doing so..

Hello Elizabeth.

It is not mandatory to make audio recordings of interviews. If, as in your case, the participants do not wish to be recorded, you should never do so. This is simply good research ethics and should not vary between institutions.

Asking participants to comment on your notes or summary is often a good practice when it is feasible, but I would say it is not absolutely necessary. I would expect practices here to vary among researchers and between research projects.

For example, if participants' email accounts may be accessible by others, e.g., by other family members, emailing them interview summaries on a sensitive topic could conceivably pose a risk.

Recording interviews and verifying notes with participants can improve the validity of your data / analysis and reveal things you hadn't spotted during the interview. There are other ways to help improve validity, such as triangulation with other data sources. All your methods in combination need to be sufficient to convince you and your audience of your findings.

Yours,

steve woods

If you don't tape the interviews your analysis takes on limitations that need to be accounted for in the final write-up as Zairul pointed out.  If you dont tape and make transcriptions, you cannot do line by line analysis, and this limits the methods you can use.  If you are modeling the study based on Grounded Theory Methodology, then tape and transcribe.  I won't say it is impossible to use aspects of GTM in your analysis, but for me line by line analysis is essential in GTM.

as Steve has pointed out, with technologies such as e-mail, there can be concerns of confidence (ones the participants may not even think about.)  In the reality of the situation (my opinion), the participants worries about being taped are really unfounded, if they are given proper assurances, but this can be a delicate process.

There are a few things that i would do in a similar study to try to influence consent to taping-- one thing is to guarantee confidence and privacy--one thing i do to emphasize this with participants (even when the interview material is not ultra sensitive) is when talking about how the process works bring up that my confidence goes so far as to resist subpoena.  In other words i inform the participant that there are no special legal relationships between researcher and participant, but that I would rather go to jail than reveal anything they told me in confidence (if you have any moral exceptions to this they should be made clear to the participant).

And perhaps more importantly: what will be done with the tapes.. i use a line in my informed consent forms something like this-- the tapes will be used to makes transcripts of the interview...all identifying information will be disguised in the final transcript, and the interview tapes will then be destroyed or digitally altered so as to make identity impossible.  depending on the actual sensitivity of the material, you may want to ensure that you computer is properly protected if using digital recordings--putting them on an encrypted flash drive may be sufficient, but dont copy them to an unsecured system or any type of cloud.

you may even have to guarantee that no one besides the person doing the interview will hear the tapes..meaning you have to transcribe them yourself..lot of work but worth it if it gets consent to tape.

hope this helps,

jam

Dear Elizabeth,

I would say that taping interviews does not absolutely guarantee excellent analysis and research quality. Remember that an entire generation of qualitative researchers and ethnographers relied "just" on their notes. Of course it depends on the analytical approach you use, but your participants and their concerns always come first, and they should not feel compelled to be recorded; such a feeling could even compromise the accounts you will gather. I well understand your point because I am doing research about gender in higher education and some of my participants expressed the same concern.

Confronting your notes with participants  is a very good strategy, you could even do that face-to-face (if possible): not only it will enhance validity, but it will give you additional insights on your findings and on your research method; it will be an important part of the process of reflexivity (that is essential in qualitative research). I even think that, once you will finish your study, you could even write a piece on how to deal with interview data when taping is not possible.

David Silverman wrote extensively about qualitative research and I am sure you can find some useful reflections if you go through his work. Steinar Kvale wrote about interviews, and you may find useful to give a look to his book "InterViews", you will see how your definition of interview and interview setting greatly impact on your analytical method.

All the best for your research !

Viviana

It depends on the qualitative method that you are using and preferences of the participant.   Audiorecording is not used in certain types of grounded theory studies but is required for linguistic studies.  As an aside, I would check with your IRB about sending research data via email.   We only used secure means to share data with participants and/or members of the research team, such as FTP, drop box, or the cloud. 

First, interview studies conducted at facilities such as the Hawthorne plant (1920s-30s) did not use audio recordings, because these did not exist.  Notes would be constructed after the interviews.  The researchers also later sometimes dictated into wax cylinders. In short, managerial field research was started without our current recording and software analysis methods.  When I first started interviewing, I took notes during the interviews and over the next few days, I added to the notes the best I could from my memory.  When I showed these summaries to my participants, they were surprised by the depth of detail.  If you are going to create case examples, I think it is very important to show the participants the summaries and get their reactions.  They could add, alter or delete something.  I work for a government institution.  All of my work e-mails can be requested by newspapers or other sources.  Therefore, I always communicate with my participants via an e-mail account that is independent from my work e-mail.  When I first started taping, I noticed how much detail I had been missing when I was only using the best notes I could create.  Whether the lost detail matters to you is a good question. You obviously lose when people pause, vary their voice tone and many other things.  When you are first being exposed to different workplaces, the jargon can take some time to grasp when you are an outsider. You will probably translate things into your wording. How close this is to their wording is a key issue.  If your participants did not want to be recorded, the reason could make a big difference.  If they want the flexibility to deny something or claim they were misquoted, sending them a summary essentially puts them on the spot. They do not know how you will eventually use their words, or if something might be taken out of context.  Reviewing case summaries removes this issue.  Some participants may simply not respond to your note summary of an entire interview.  In addition to not wanting to be pinned down, some may not want to take the time to review lengthy note summaries and make numerous corrections.  If they do not reply, you are faced with the issue of whether you think you can or cannot move forward based on no comment.  On the one hand, they did not correct anything. On the other, they did not confirm anything.  Your consent form makes a big difference here.  They should know what you are going to do. If your consent form did not cover the subject, your e-mail could make your plans clear. For example, you could give someone a period of time such as two weeks to make suggested additions, alterations or deletions.

Its up to you, if u can make notes of every point in the discussion, so you will not need it, but if you can't, it will help as you can return to the recording and write down any missed parts in the discussion.

I would say it is Elizabeth unless you have eidetic memory... simply put, if you don't record the interviews you run the risk of frantically taking notes which is likely to make your participant anxious and you very tired and out of focus to be able to think of the most pertinent follow-up questions (because you are so focused on writing what has been said), not to mention the large gaps of information from the interview as no one is able to write/type as fast as free-flow speech.

Do you have any specific worries about audio-recording?

Hi Everyone

 Thank you so much for all these amazing answers - this forum is so helpful !

I suppose I am limited because I have now completed 40 interviews and none are taped.

Only 50 % said they wanted to see the transcripts and comment -all those came back with good and helpful comments.

One woman did not want to talk at all and to only write her responses on the template and send it back like a survey via email.

I have only used my personal email so I suppose that is  a plus.I see your point re other family members - however this is a matter of  physical birth injury so most family members are over keen to give advice and help no matter how inaccurate their perspective.

Nevertheless, I will  need to show that my analysis is thorough - the themes however line up with the quantitative research done by my supervisor and team on a longitudinal 7 year study re women's health issues - women have also been examined medically. At present I have only used thematic analysis as it draws on the research done by my team.

I should have noted that I am working with a team of quantitative doctors and I am a midwife. I am not sure Grounded Theory is the way to go.

Thanks again everyone - your comments are really helpful re perspective.

 Warm regards

 Elizabeth

I think the point to recording interviews is to transcribe them and to insure that the words of the participants are being analyzed, and you analyze what they said rather than what you remember them saying. Our memories have a way of not being as good as we think they are. You also would not want to misinterpret the words of the participants, and often times, depending on the qualitative methods, you will need to use quotes.  Notes, or even better a journal, (or both) are also valuable, but usually done after an interview (sometimes during).  Although valuable and useful, they do not replace the transcription of the interviews.

The sound feedback and recommendations provided in this thread provide ample support for recording the open ended interviews. A few other reasons to keep in mind are that in any dialogue there will be "cover terms" (see Spradley) that may be unfamiliar to the interviewer at the time of the interview that can be revisited through transcripts and coding for follow- up clarification if missed in the initial interview.  

Verbatim transcripts can also be important in cases of inarticulate or ambiguous verbal responses, when one is trying to get at a respondent's intention. For example a pause in the narrative - whether brief or extended - can be captured in the recording for analysis.  It could suggest that the person was gathering his/her thoughts, or that the question is overly sensitive to respond to, or the person does not understand the question but is reluctant to ask for clarification.  In each case the taped recording may help  capture the respondents intenton -i.e. pause in conversation -in context. 

As previously mentioned frantic note-taking while somewhat effective is a serious distraction to an otherwise normal "friendly conversation" as noted by Spradley on an ethnographic interview. My experience has been once I start the conversation and the tape recorder, I try my best to always maintain eye contact with respondent and not look away as it may be perceived as disinterest.  To the contrary I try to demonstrate keen interest and close attention to a person's response, only breaking eye contact to take notes and only sparingly. To this point I should add that probes are easier to review in a recording as they are often introduced spontaneously in the conversation rather than pre-determined in an interview schedule. If I had to write down my probes every time the opportunity presented itself it would feel awkward and out of sequence as would be the respondent. Recordings are a blessing in those instances. 

Just my thoughts on the value of recordings and good luck moving forward. 

It is preferable, but one must follow the participant's wishes.  Some (as in my dv and welfare study) were very wary so I just jotted notes the best I could during the session and then afterward taped my own memory of what they had said, separating it from my observations and opinions

There are many reasons that participants may not want to be recorded and as Ruth mentions you have to follow their wishes - the question you are essentially faced with is 'will my research be weaker for not having this person involved?'. For example, if I was designing a project with a broad criteria for includsion, had many participants willing to be interviewed and only one person who objected to being recorded I may decide to simply recruit someone else. But if that person can provide insight into the topic or a perspective that is otherwise missing then I would go to considerable lengths to include them (and I think your solution of writing notes and then sending them to participants by email for confirmation that you captured the details is a good one). But as Ruth states-  there are some particular projects were large numbers of people will not want to be recorded either because of the nature of the questions being asked or because particular cohorts may have a difficult history/relation with researchers, research practices and research institutions. In this case, your option is to further marginalise and silence these groups by not involving them in the research or to seek methods that are acceptable to the community - such as not recording the interviews. It is a perfectly acceptable method and should not pose a problem to disseminating the research provided you discuss your reasons for this particular project design, some of the challenges encountered, solutions found, etc. 

Thank you so much for all these really helpful responses.

@ Elizabeth Mary Skinner My experience.

I have used recordings in my recent studies. However you have to let the respondents know and clearly state why you are recording. For example I asked the respondents first if they did not mind being recorded. Some immediately said 'no', other said 'it depends' while others did not mind. I had to explain why I was doing so. I was recording because; I needed to focus on my questions, I wanted to review our interview after so it was time saving, I needed some of these recordings for future reference (but first asked permision etc.). I even told them would send their recordings back to check if I could use them elsewhere if they consented. After this many were happy to be recorded and I kept my word.

Often times recordings are really helpful when doing interviews but the respondent must be aware and first feel confortable about it. At times I would begin the interview and two minutes down I ask if I can record and explain my reasons. At times I told them your views are really important and it would be nice to have everything and not miss out anything. This way they got my trust. Moroever I kept reminding them that this was research aimed for the good and hence their identity would never be revealed unless they desired. Like the whole process of interviewing, one needs to be smart enough to make the interview as informative as possible.

Steven R Woods without recording are the results publishable for a paper.

You cannot depend on your memory alone to remember every detail of an interview. A transcript can be sent privately to the respondent for him or her to read and state if it is an accurate account of the interview or if there anything he or she stated that should be omitted. This also increases your chances of having fulfilled some of the conditions of trustworthiness and confirmability in a qualitative study although there are certainly other ones you may use. Finally, some interviews may last one hour and yield 5 pages while others 10 or 15 pages. It enables you to also verify accounts of the participants you thought were made or could have been made and verify them by playing back those segments. Although I have stated these opinions as a researcher, I see no reason if you take good notes, are attentive and after the interview is over and the respondent is gone, you do an oral recording of your memory of it, the study could not be a good one and be publishable. There are recent studies that confirm this too.

Conducting in-depth interviews with and without voice recorders: a comparative analysis.

Rwamahe Rutakumwa*, Joseph Okello Mugisha*, Sarah Bernays, ...

First Published November 7, 2019,

https://doi.org/10.1177/1468794119884806