After you finish conducting interviews, the first step in the post-interview phase of your research project is to transcribe your data. Transcription is the process of converting spoken word into a written form. When transcribing an interview, you simply take your recorded data and write it down as it was spoken. The transcription process will allow you to recognize any emerging patterns within and across your interviews and will be important when analyzing your data later.
Transcription is important as both a process and a product. The process of transcribing allows you to fully immerse yourself in the data and become sensitive to linguistic details within the data. The product created from transcription is a tool that can be analyzed for patterns and themes multiple times by multiple researchers (Avineri, 2017).
Types of Transcription
There are three main types of transcription: verbatim, semi-verbatim, and intelligent verbatim. In verbatim transcription, every word is recorded exactly as it was spoken. This includes filler words, verbal hesitations like “umms” and “ahhs,” and grammatical errors. In semi-verbatim transcription, verbal hesitations and pauses between words are omitted from the transcript for the sake of clarity. Finally, in intelligent verbatim, filler words are omitted, and grammatical errors are fixed in the transcript. Ethically speaking, verbatim transcription is the most accurate type of transcription and leaves the least room for bias.
The transcription process can be lengthy, especially if you do not have experience with it. For every hour of recorded data you collect, it may take between four and eight hours to transcribe fully. Luckily, there are many online transcription services available. Part of the reason transcription takes so long is that you are not only transcribing what was said, but also detailing your observation of non-verbal cues and scene descriptions into the transcribed data. For this reason, it is helpful to transcribe an interview soon after it is conducted and have your written notes on hand during the process.
You may be thinking, “why should I bother transcribing if it is such a tedious process?” Here are a few reasons why transcription is essential to the research process:
- It makes data easier to analyze and share
- It allows researchers to understand the data more fully
- It helps researchers to recognize patterns in the data
- It gets recorded data into a text-based format which will be important for presenting findings
- It preserves the accuracy and integrity of the data
Here are some tips to keep in mind when transcribing:
Ethics Check
While it may seem like verbatim transcription leaves little room for the transcriber’s own assumptions and biases, decisions are still being made in turning audio into written text. Different transcribers may even hear the same spoken material differently. The choices you make during transcription could shape how a participant is portrayed and how their knowledge is valued. You consciously and subconsciously decide how to display the spoken data. What ends up on the transcript will impact what conclusions you are able to make during analysis. Adopting standard conventions in terms of page layout, transcription symbols, and placement of verbal and nonverbal behavior will help you display the data in such a way that will not influence your analysis later on. Additionally, standard transcription conventions will let other researchers easily read, interpret, and come to their own conclusions about your transcript. For more information about transcribing procedures, conventions, and formats, take a look at Transcription in Action from the Department of Linguistics at the University of California, Santa Barbara.
Although it is helpful to transcribe the data yourself in order to become more familiar with it, there are also many technologies available to help with the transcription process. See our interview resources page for information about many of the available transcription tools. Keep in mind that the transcriptions produced by these services will still need to be reviewed by a pair of human eyes in order to input your observations of nonverbal data and make them understandable.
Personal Project
Take some time to consider how much interview data you have or will have. How much time will you have to allot to transcription? What kind of analysis do you plan on conducting, and how can you transcribe it to fit with this analysis? Then, do some research on the available online transcription tools which you can find on our interview resources page. Which one best fits your needs?
