¶ … Falter in Quality
The strength of the use of interviews as a data collection device is that they ensure a high rate of response; that the interviewer can tailor his or her response to the subjects' responses over the course of the interview; and the interviewer can clarify any ambiguous responses. Problems with interviews can arise when the subjects make their responses conform to the unintentionally communicated expectations of the interviewer, based on a desire to 'please' the researcher. To ensure quality control in my process of interviewing I tried to adhere to a relatively similar script between interviews for comparison of subject responses. I did not want to be overly directive during the interview and only intervened if the subjects were confused and had no response to my queries. I transcribed and recorded the interviews to ensure I was not missing any critical word choices that might be revelatory of the subject's attitudes. I also asked a limited number of questions given the time limit I had to ensure accuracy and so I could go over each subject's response very thoroughly.
To reduce bias, I tried to ask general questions rather than leading questions. I focused on the perceptions of the subjects to trafficking in general rather than specifically asked them about my research question. I also tried to proceed from general to more specific questions once again so I did not appear to be 'leading' the subjects in a particular manner, thereby tainting the results. I was relatively non-directive, even when it was tempting to sway subjects down a different path than the one they chose to take (for example, when someone did not understand what human trafficking meant). In the future, however, I think I would introduce some sort of 'mechanism' to ensure that in case the subject did not understand the first question, the subsequent questions would clarify the original term, although I did find it interesting after doing so much research on the subject that the phrase 'human trafficking' was not a universally known construct.
You’re 84% through this paper. Sign up to read the full paper.
Sign Up Now — Instant Access Already a member? Log inAlways verify citation format against your institution’s current style guide requirements.