Transcribation- a useless scientific practice

Just a warning: this is about research in genereal and less about Brazil...



When I did my master´s thesis, I committed what could be the closest one can get to the conception of  ”sin” within the world of qualitative research: I did not transcribe my interviews. Now, several years later, I have fallen deeper into the pit: I see no scientific value in transcribing the interviews and see it more as a practice of honoring the professional etiquette, a reminiscence of the past and a rite de passage of an academic career.

Within the traditions of qualitative interviews there is an elusive notion that after having done an interview, you should sit down and transcribe it all, transfer the data to a sheet of paper, including remarks on how the interviewee is talking, pauses (in parentheses), interrupted sentences…like, cut of in the middle of…not completed, a kind of, orally, nothing like one normally would do in writing, you know? Faithfully copying everything that has been said. And only after having gone through this ”systematization” of your data, you can start analyzing. In perspectives of good scientific practice this is not only a waste of time, it is also a well-established practice of deteriorating the data collected.

I can understand that in the earlier days, when recording was more complicated than today, detailed transcribation was a method to have easier access to the data when doing the analyses. Forewarding and rewinding over, and over again was time consuming and it was difficult to detect patterns and structure the information. However, making a detailed copy of your data in written and analyzing these data instead of your original recordings? I don’t see the point. As a master’s student I had all my interviews on tape. I listened carefully through the tape, divided the interview into equal time intervals and made a resymé of the content (every time the counter on the side running from 000 to 999 turned another 50, I made a new passage in my notes). This gave me a pretty good overview of the content of my interviews, and a reference for going back to the important data afterwards. And then I listen to it directly, the original source so to say, instead of reading about it on a paper. In the process of dissemination, of course transcribation was done. But only the passages actually quoted in the final product.

Today the practice of transcribing all the interviews is even more dubious. Recording is done electronically and computer programs for analyzing them is accessible to everyone. Notes and quotes can be attached to the sounds, exact length and number of pauses can be quantified (if it is a point at all to do this), and all the traditional practices of transcribation can be done. And even more exiting, you can twist the voice and anonymize your data, detect speed of speech, change in tone and frequency and a lot of fancy things that was difficult with analog recordings. While doing this, you still interact all the time with the original data- the sounds from the interview, not a sheet of paper which is nothing but a bad copy.

So what is the point of transcribing all the interviews in length? Well there is an aspect of honoring those who suffered through hours of boring and tiresome work. In respect for those who have bled before, we should also do some bleeding. Because research shall be boring some times, trancribation is a fantastic way to make it boring. It makes you reflect on the hard life and sacrifices of the researcher. A kind of rites de passage.

My point is this: when students complain and ask you why they must transcribe the interview, give them a heck of a good answer.