A comparison of computer-assisted and hard copy telephone interviewing

Area of Science:

• Epidemiology
• Health Services Research
• Survey Methodology

Background:

• Traditional telephone interviewing relies on paper questionnaires, which can introduce variability.
• Computer-assisted telephone interviewing (CATI) offers a standardized approach to data collection.
• Evaluating CATI's impact on interview dynamics and data quality is crucial for research.

Purpose of the Study:

• To compare computer-assisted telephone interviewing (CATI) with traditional paper-based telephone interviewing.
• To assess the impact of CATI on interview length, interviewer actions, and data completeness.
• To evaluate the quality of interviews conducted using CATI versus paper questionnaires.

Main Methods:

• A randomized trial was conducted comparing CATI with paper questionnaires for telephone interviews.
• Respondents were surrogates for 400 deceased Florida residents from a colorectal cancer case-control study.
• Key outcomes included participation rates, interview duration, comments, probes, 'don't know' responses, and interviewer evaluations.

Main Results:

• CATI interviews were, on average, 3.4 minutes (14%) longer than paper-based interviews.
• The number of interviewer comments decreased by 25% and probes by 19% in CATI.
• Observed differences in comments and probes were smaller than variations between individual interviewers.

Conclusions:

• CATI is a viable method for telephone data collection, with minor increases in interview length.
• CATI standardizes interviewer behavior, reducing variability in comments and probes.
• The study highlights CATI's potential for improving data consistency in epidemiological research.