Must it be a Quarrel for it to be a Qualitative Interview?

Are you a project supervisor, transcriber, or at any point responsible for handling qualitative interview audio? Have you had an experience where the interview sounded like a street conversation rather than a meaningful engagement that addressed the research/evaluation questions?

I understand that there are many interview styles, and they differ by field; these include job (human resources management), research, psychology, medical, sociology, and others. All these interviews have different structures and styles, but mainly fall into formal, informal, structured, unstructured, semi-structured, or group interviews.

At different times, I have handled audio for different kinds of interviews, but this time I was astonished. I could not fathom the fact that it was a research interview, more or less a focused group discussion, because it was a chaotic collection of information, despite the fact that it was supposed to be a semi-structured kind of conversation.

The interviewer mostly used jargon and informal street language; I suppose it was not to build rapport with the participants, because it could have been done differently. This simply distorted the meaningful collection of information. As a transcriber, I was forced to work on changing these words to meaningful Swahili terms so that they would become relevant when translating to English and make sense during data analysis and report writing.

Most of the time it sounded like the interviewer is quarreling with the participants, forcing them to answer questions and belittle them when they do not respond despite repeating "all answers are correct, just say what you know", in most cases this statement sounded like you have to say anything regardless, and even if your answer do not make sense, just say it.

It is not the statement itself; it is the way the interviewer said it. The tone did not sound right.

I understand the use of that type of language could mean something in the research, like the kind of community interviewed was of a certain calibre; however, the constant use and repeatedly bringing in informal terms in the conversation during the interview could mean something else to participants and bring about answers that are irrelevant as well.

When you introduce yourself as a researcher, participants tend to view you in a certain way, especially in our communities. Despite going down to their level of understanding, when using some type of words, they might rethink whether you are truly a researcher or someone like them. It is different when the research involves observation, as it is a community-based participatory approach of research; the grounds are different.

Unless the research methodology agreed to use informal questions, it would have made sense even during the analysis of such data. But when the interview was supposed to be structured or semi-structured, and all I heard during transcription was a series of street terms, most of which were brought up unnecessarily by the interviewer rather than the participants, it was disheartening. The transcription process was painful, from the length of the audio to the constant thought process of writing these words into meaningful way.

For the interviewer, it was not the first nor the last time they used such language, which could indicate that the supervisor did not review the audios, and previous transcribers never mentioned it. It is time to emphasise that, regardless of your enumerators' experience, they should be trained accordingly and, to ensure data quality, the audios should be reviewed, with a room for the transcribers and translators to provide constructive feedback. This will improve data quality and benefit subsequent projects that rely on such data.

Qualitative interviews are beautiful when done in the right way. Quality data is crucial; this all comes from careful execution of the methodology selection, recruitment of enumerators, training them, and reviewing audios before the data analysis process starts, from transcription and so forth. It does not have to be a quarrel for one to get data from participants.