Nonresponse in CATI-Surveys. An Empirical Study on the Effects of Interviewers’ Vocal Characteristics

Abstract

This study examines nonresponse in telephone surveys. Our analysis relates response rates to interviewer voice characteristics. Drawing on theory we argue that interviewer voices can affect the respondents’ utility expectations, e. g. by indicating integrity, or by providing an intrinsic reward. Based on dual-process theories, one can additionally expect the voice to be a situative cue. We use contact data generated in a CATI survey and supplement these data with information on interviewers‘ voices. The survey was conducted during 2007 and 2008 at the University of Mannheim, Germany. To obtain metadata, we recorded the interviewers‘ voices and generated phonetic measures of vocal characteristics. This enabled us to study the determinants of interviewer effectiveness and nonresponse with a special focus on objective voice characteristics (pitch, speech rate, etc.). Additionally, we accounted for a variety of subjective voice attributes (friendliness, trustworthiness, etc.). The results show that objective vocal characteristics have greater explanatory power than subjective indicators. Furthermore, the objective voice characteristics are related to the interviewers’ productivity in a nonlinear way.