Interactive Voice Response (IVR)

Interactive voice response (IVR) is a data collection technology that uses a recorded voice to ask survey questions by telephone, in place of live interviewers. Respondents enter their answers by pressing the buttons on the keypad of their touchtone telephone. An IVR system controls the presentation of the survey questions, captures the responses entered via touchtone, prompts respondents to answer questions, and offers automated help to respondents. IVR is also known as telephone audio computer-assisted self-interviewing (T-ACASI) and [Page 343]touchtone data entry. These terms all refer to computerized telephone data collection systems where respondents answer survey items via automated self-administered procedures, as opposed to giving verbal answers to live interviewers.

IVR has two primary uses in survey research. First, IVR can be used to replace interviewer-administered telephone data collection. Potential respondents may be contacted first by telephone interviewers and then switched to IVR, or they can be contacted by another mode (e.g. mail or Internet) and provided a call-in number to use for completing an IVR interview. Another use of IVR is to provide telephone survey respondents greater privacy in responding to questions of a potentially sensitive nature. When used for this second purpose, IVR interviews are typically conducted through initial contact from telephone interviewers who then switch respondents to IVR for the sensitive items. The respondent is then switched back to the interviewer after answering the sensitive questions. Regardless of the specific purpose, IVR data collection typically involves a relatively short, simple interview or a brief module that is part of a longer interview.

IVR offers a number of potential advantages over interviewer-administered modes of telephone data collection. First, because a pre-recorded voice is employed to administer all survey items, IVR respondents are all read questions, response options, and instructions in the exact same way. This provides a higher degree of interview standardization than interviewer-administered telephone data collection, where interviewers' vocal qualities, reading skills, and presentation skills vary.

A second advantage of IVR is providing greater privacy for answering questions, especially survey items of a potentially sensitive nature that could be affected by socially desirable reporting. Because IVR respondents enter their answers by pressing touchtone buttons, they do not have to be concerned about giving their responses to a live interviewer or about others (e.g. family members) hearing their responses. Research indicates respondents in IVR mode give more nonnormative answers to questions of a sensitive nature compared to respondents in computer-assisted telephone interviewing (CATI) mode. Examples include greater reports of illicit substance use, certain sexual behaviors, and negative satisfaction ratings.

Third, when IVR surveys are conducted with no interviewer involvement, this mode can be a much more cost-effective method than CATI. This version of IVR data collection saves on both the costs of recruiting, training, and supervising interviewers as well as telecommunication costs. Because telephone charges are incurred only when potential respondents call in to complete IVR interviews, these costs are lower than costs in the interviewer-administered mode where multiple outbound calls are typically made to contact each sampled unit.

IVR does have some potential limitations compared to interviewer-administered modes of telephone data collection. First, IVR mode provides unique opportunities for unit nonresponse or incomplete interviews. When respondents are left to complete an IVR interview on their own time, they are less likely to participate without the motivation provided by interviewers. Similarly, in surveys where interviewers first recruit respondents and then switch them to IVR mode, respondents have an opportunity to terminate the interview at that point, without being exposed to the persuasive efforts of an interviewer.

...