Total Survey Error in Practice

19Using Doorstep Concerns Data to Evaluate and Correct for Nonresponse Error in a Longitudinal Survey

Ting Yan

Methodology Unit, Westat, Rockville, MD, USA

19.1 Introduction

Doorstep concerns are one type of paradata collected at the “doorstep” in a face‐to‐face survey. These data capture the interactions between interviewers and potential survey respondents during the survey introduction, and reveal the concerns sampled members have expressed about the survey request and also their reasons for refusing the survey request when refusal occurs. Survey organizations vary in how they collect doorstep concerns data and how much they collect. Some tape or record the actual interactions between interviewers and respondents at the survey introduction (e.g., Campanelli et al., 1997; Couper and Groves, 2002; Groves and Couper, 1996; Maynard and Schaeffer, 1997). These recordings contain rich and in‐depth information on what exactly happened. However, the key challenge with this type of data collection method is the qualitative nature of the resultant data—the recordings have to be transcribed first and then subjected to further coding or content analysis before they can be analyzed using quantitative methods. For instance, Campanelli and colleagues asked interviewers to audio‐tape full interactions occurring at the doorstep (Campanelli et al., 1997). They then transcribed all recordings, developed a coding scheme aimed to investigate interviewer persuasion strategies, and coded recordings ...