Changes in Interviewer-Related Error Over the Course of the Field Period: An Empirical Examination Using Paradata and Behavior Codes

• Activity Number: 491
• Type: Topic Contributed
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Survey Research Methods Section
• Author(s): Kristen Olson*+ and Antje Kirchner
• Companies: University of Nebraska-Lincoln and University of Nebraska-Lincoln
• Keywords: Paradata ; Total Survey Error ; Interviewers ; Nonresponse ; Measurement Error

Abstract

It is well-established that interviewers learn behaviors both during training and on the job. However, how this learning occurs has received little empirical attention. There are two competing hypotheses about what happens during field data collection - one is that interviewers learn behaviors from their previous interviews, and thus change the measurement situation in reaction to the behaviors previously encountered (i.e., a contagion model). The second hypothesis is that interviewers encounter less cooperative respondents as the data collection period progresses, and thus change the measurement situation in reaction to these types of respondents (i.e., nonresponse propensity affecting the measurement error situation). This paper will examine these two hypotheses using data from CATI surveys, paradata, and behavior codes. Findings will have implications for interviewer training and monitoring.