Computer Audio-Recorded Interviewing as a Tool for Survey Research

Abstract

Developers must understand the needs of user populations and the potential benefits of the new software to them, so that the development team can create an effective system. Especially for applications employed outside the computing profession, it is important for the software team to learn context and workflow and to understand the value that their development work will bring to the users. This article discusses design and implementation considerations for computer audiorecorded interviewing (CARI), a method coming into widespread use for survey research in the social sciences. When implemented as part of the data collection process, CARI allows a survey manager to listen to the exact circumstances of how questions were asked and answered during the interview, a much more powerful approach than prior indirect methods of quality control and improvement. Design considerations can be complex when planning an integrated system. Based on a decade of experience and prior implementation of several distinct CARI systems, this article explores a part of the operational world of survey research from the eye of the system developer. It offers context for those developers who are unfamiliar with survey research or for anyone who is unfamiliar with CARI operations. Discussion focuses on benefits, requirements, user goals, system design challenges, and options.Keywords audio recording, system design, interface design, CARI, computer audio-recorded interviewing, survey research, qualitative researchWhat do the following have in common: making policy decisions for handling child abuse, estimating cocaine use in cities and states, knowing if the cost of living is going up or down, and setting the fees for hunting and fishing licenses? All of them depend on data collected through research surveys, underlying decisions made by federal, state, and local agencies. Nearly every step of gathering, managing, and delivering survey data occurs electronically, shepherded by software and hardware

How Well Do Interviewers Record Responses to Numeric, Interviewer Field-code, and Open-ended Narrative Questions in Telephone Surveys?

“…We need a fundamental rethinking of whether and how we ask these questions and record or code the responses we receive. One obvious avenue for the future is to take advantage of audio-recording capabilities to capture full responses (Hicks et al 2010;Thissen 2014), which can then be processed by coders postsurvey. Our coders could reliably code responses to these questions from recordings and transcripts, in large part because they did not have to also do the many other tasks interviewers do (Japec 2008).…”

Bridging anthropological theory: Accumulating and containing wealth in World of Warcraft landscapes

“…I excluded players I believed were under 18 years of age. I conducted interviews in Skype (Deakin and Wakefield, 2013) and recorded interviews using MP3 Skype Recorder (Thissen, 2014).…”

Applying Machine Learning to the Evaluation of Interviewer Performance

“…Computer-Assisted Recorded Interviewing (CARI) has long been used by survey organizations to monitor interviewer performance (e.g., Hicks et al 2010;Thissen 2013). CARI produces question-level audio recordings that capture the interactions between the interviewer and the respondent.…”