Hanyu Sun 2015 AAPOR Abstract The Impact of Rapport on Data Quality in CAPI and Video-mediated Interviews: Disclosure of Sensitive Information and Item Nonresponse Rapport is generally described as a sense of connection, mutual comfort and ease of conversational coordination during an interaction (Foucault, 2010). Although there is no universally accepted way to measure rapport, the general consensus is that it is good for survey interviews and may affect the quality of the responses obtained (e.g., Foucault, 2010; Cassell & Miller, 2007). Technological advances in recent years have made video-mediated interviews more feasible and affordable; however, little attention has been paid to videoconferencing as a potential mode of data collection in survey interviews. In video-mediated interviews, the interviewer and the respondent can see and talk to each other via a video window. Although rapport-related verbal behaviors have been found to increase the disclosure of moderately sensitive information in face-to-face interactions (e.g., Dijkstra, 1987; van der Zouwen et al., 1991), it is unknown if rapport can be established to the same extent in video-mediated interviews, leading to similar levels of disclose. I examined the impact of rapport on data quality in CAPI and video-mediated interviews with a laboratory experiment that varies the level of rapport, the mode of data collection, and the version of the questionnaire. Eight professional interviewers and 128 respondents participated. I found that (1) there was no significant difference in rapport ratings between video-mediated and CAPI interviews; that (2) the impact of rapport on disclosure depends on question sensitivity: when questions are moderately or less sensitive, rapport does not seem to affect disclosure; whereas when questions are highly sensitive, rapport leads to more rather than less honest responses; and that (3) the effects of rapport on item nonresponse depends on question sensitivity: when questions are slightly sensitive, rapport motivates respondents to invest more effort to respond, whereas when questions are moderately or more sensitive, rapport leads to more item nonresponse.
© Copyright 2016 ExploreDoc