Relatively Accurate Needs to be Accurate

The most difficult part of this assignment was understanding the interviewee. Because of his thick accent and imperfect enunciation, it was very difficult for my unfamiliar ear to make out all his words and syllables at first. As I listened to the clip more often, I became familiar with certain variations his accent put on certain English words. I also noticed that it was much easier for me to fill in the blanks for certain words and phrases once I had listened and recorded in detail, the interview. Clues and hints in the conversation made these things clearer. The interviewee also had a habit of trailing off and not finishing sentences which made it difficult for me to understand where one thought ended and another began.

Indexing this was not as difficult as I expected it to be. I quickly developed a system of listening to about 7-10 second clips, writing down what I could remember, and then replaying it to verify what I had recorded was what I heard on the sound bite. Although this took a while, about two hours total, it enabled me to record a relatively accurate translation.

While recording the clip, I used commas where both participants paused and hesitated while speaking. The interviewer was very easy to understand and record because her portion of speaking came mainly in the form of questions. With the interviewee, he spoke in fragmented and run on sentences. So, the grammar I used to record him is less than correct, but depicts the rhythm of his speech well. I was concerned about not using good grammar when I began transcribing, but I became more comfortable with it as the process progressed. I used the words gonna, ‘em for them, and incorrect syntax to emphasize the interviewees accent. When I encountered words and phrases that I could not decipher, I used the words that sounded closest to what I heard followed by a question mark in parenthesis. In other cases where I could make no sense of what was said, I used three underscores followed by a question mark in parentheses. I think the root of my inability to understand certain words was the interviewee’s accent.

I think the interviewer did a good job of speaking clearly and asking pointed questions during the interview. The only thing I found a bit offsetting, especially for the interviewee, was the lack of emotion or engagement I heard from her on the sound bite. Although she may have been very physically engaged in the conversation, it did not appear to be that way on the clip. Overall I think she was very composed and prepared for the interview but I would personally approach the situation with a bit more vocal engagement. I think it is important for the interviewee to know that you are physically and mentally focused only on them and what they have to share.

My overall reaction to this assignment is that it is important to fully engage with the interviewee, remain composed and come prepared, and finally to ensure the clarify words and phrases during the interview if you do not understand because listening to the clip will make it all the harder. Clarifying words and phrases, especially if someone has a thick accent, is important so that future researchers can access the material and understand it with ease.

1 Response to “Relatively Accurate Needs to be Accurate”

  1. 1 mal2013 February 17, 2014 at 1:09 pm

    I appreciate your detailed description of how you planned the transcription process. Do you have thoughts on how you will avoid seeming emotionally detached from your interview subject?

Comments are currently closed.


The Williamsburg Documentary Project (WDP) strives to collect and preserve the rich past of Williamsburg, Virginia. By conducting oral history interviews, building physical and digital archives, and creating online exhibits, the WDP interprets Williamsburg’s recent past. The WDP works towards developing a better understanding of Williamsburg by bringing together individuals, local groups, Colonial Williamsburg, and the College of William & Mary.

Add Users

If you want to add yourself to this blog, please log in.