A Dynamic Systems Approach to Patterns of Affect and Cognitive Difficulty in Interviewer-Respondent Interactions

• Activity Number: 445
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Survey Research Methods Section
• Author(s): Matt Jans*
• Companies: UCLA
• Keywords: interviewer-respondent interaction ; dynamic systems ; survey interviews

Abstract

Interviewer-respondent (I-R) interactions are dynamic phenomena at their core, making dynamic systems (DS) theory an obvious framework for their study, yet DS theory and methods are rarely applied to I-R interactions. This study uses utterance-level coded data from audio recordings of phone interviews conducted by the Reuters/University of Michigan Surveys of Consumers (SCA). Recordings were transcribed into interviewer and respondent utterances, and rater judgments of affect were applied. Using GridWare v1.15a, a DS theory-based software (www.statespacegrids.com), trajectory plotting, attractor states identification, dynamic systems parameter estimation are demonstrated. The paper also motivates a graphical and intuitive understanding of cross-utterance I-R dynamics using the software. Initial results show patterns of affect dynamics varying by responses to income, interviewer gender, and across interviews conducted by the same interviewer. Potential for DS theory and methods to inform and guide the study of I-R interactions is briefly discussed. This research was sponsored by the Charles Cannell fund in Survey Methodology and the U.S. Census Bureau Dissertation Fellowship.