Objectives: To assess the differential effects of face-to-face and Audio-Computer Assisted Self-Interviewing (audio-CASI) interview modes on different categories of questions.
Methods: Subjects were recruited from persons attending syringe exchange programs in the U.S. They were randomly assigned to face-to-face interviews or audio-CASI. The questionnaire contained items on socio-demographics, attitudes towards the program, drug and sexual behaviors, and physical and mental health. The questionnaire was abbreviated to 90 questions, which were categorized by raters into three categories: “stigmatized behaviors,” “neutral behaviors,” and “personal problems.”
Results: Mean scores were compared by interview mode. For “stigmatized behaviors”, the audio-CASI elicited a mean score of .14 (sd=.06), versus .11 (sd=.04) for the face-to-face interviews (p<.001). For “neutral behaviors,” the two modes elicited identical mean scores of .48 (sd=.11). For “personal problems,” the mean score for audio-CASI interviews was .29 (sd=.25), and for the face-to-face interviews the mean score was .32 (sd=.28, p=.046).
Conclusions: These results support the proposed hypothesis that different types of questions are affected differentially by mode of interview. Respondents disclosed more to the computer for stigmatizing questions. Conversely, when the questions addressed personal problems, the face-to-face interviews elicited more self-disclosure. For neutral questions, there was no interview mode effect.
Learning Objectives: Attendance to this session will provide the participant with knowledge of the benefits and limitations of using audio-CASI versus using traditional face-to-face interviewing techniques
Keywords: Injection Drug Users, Methodology
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.