Interviewer bias is a type of information bias that arises when an interviewer consciously or unconsciously elicits inaccurate information from study subjects. Interviewer bias can result in differential error, which can seriously distort disease- exposure associations, if the interviewer is aware of the exposure status and outcome hypothesis in a cohort study.
In the former case, the interviewer may probe more deeply for evidence of exposure among cases than among controls. In latter case, the interviewer may try to elicit evidence of health effects more assiduously in exposed than in unexposed cohort members. Methods used to minimize interviewers to follow a fixed pattern of questioning, and, where possible, keeping interviewers unaware of the disease status and exposure hypotheses of greatest interest in case- control studies, and unaware of exposure status and health outcome hypotheses of greatest interest in cohort studies.
Possibility of interviewer bias should be taken seriously in all research. Although it may not be possible to eliminate all forms of interviewer bias, there are a number of measures that can be taken to avoid and control potential biases.
Ways in which bias occurs
Number of ways in which bias occurs, namely:
- departures from the interviewing instructions
- poor maintenance of rapport with the respondent
- altering factual questions
- rephrasing of attitude questions
- careless promoting
- biased probes
- asking questions out of sequence
- biased recording of verbatim answers
One way of avoiding, or at least minimizing, interviewer bias is to require all interviewers to follow the same protocol. Hence, a set of guidelines might be drawn up which ask the interviewer to read the questions exactly as they are written, to repeat a question if asked, to accept respondent's refusal to answer a question without any sign of irritation, and to probe in a non-directive manner.
Types of interviewer bias
There are several types of interviewer bias:
- Personal interview. Facial expressions and tone of voice could affect responses. However, monitoring interview sessions (e.g., videotaping them or watching them through a one-way mirror) could prevent and detect interviewer bias.
- Phone interview. Tone of voice could affect responses. However, having a supervisor monitor interview sessions or taping sessions could prevent and detect interviewer bias
- Investigator administered Minimal interaction with investigator, so little chance of interviewer bias
- E-mail surveys, web surveys. No interviewer, no interviewer bias.
- H.T. Reis, C.M. Judd 2000, p.302
- M.H. Gail, J. Benichou 2010, p.455
- D.E Gray 2010, p.220
- M.L. Mitchell, J.M. Jolley 2010, p.270
- Gail M.H., Benichou J., (2010), Encyclopedia of Epidemiologic Methods, John Wiley & Sons, New York.
- Gray D.E., (2010), Doing Research in the Real World, SAGE, London.
- Mitchell M.L., Jolley J.M., (2010), Research Design Explained, Cengage Learning, Australia.
- Reis H.T., Judd C.M., (2000), Research Methods in Social and Personality Psychology, Cambridge University Press, Madrid.
Author: Natalia Węgrzyn