The effect of interviewer bias on mental illness questionnaire responses.

Abstract

The purpose of this study was to investigate how interviewer behavior affects information obtained from a respondent during an interview. The five factors of interviewer behavior that were studied were social reinforcement, interviewer disclosure, verbal modeling, interviewer dress, and seating arrangement. The dependent variable was the number of symptoms reported on the 22-item Mental Health Inventory. A total of 128 male and female adult patients under treatment for mental illness were interviewed under procedures using all combinations of the five factors. Control of patient sex, age, ethnicity, and education was effected through the use of a balanced design. The results demonstrated that the behavior of the interviewer influenced the respondent's report of psychiatric symptoms. The largest number of psychiatric symptoms were reported when the interviewer gave social reinforcement while wearing a lab coat and sitting across a desk from the respondent.