Interviewer Effects on the Process Leading to Cooperation or Refusal Using Interviewer Call Record Data
Gabriele B. Durrant, Julia D'Arrigo
Southampton Statistical Sciences Research Institute (S3RI), University of Southampton, United Kingdom

It has long been recognised that interviewers play an important role in gaining contact and cooperation from sample survey members. In particular, it has been hypothesised that the interaction process between householders and interviewers plays a key role in persuading sample members to take part in a survey. This paper investigates the influence of the interviewer on the process leading to cooperation or refusal with focus on the interaction between interviewers and householders at the doorstep. The role of interviewer attitudes, strategies and behaviours are examined, controlling for other interviewer characteristics, such as socio-demographic characteristics, work experience and pay grade. Possible interaction effects between socio-demographic characteristics of the householder and the interviewer are also investigated.

A key advantage of the study is that unusually rich information is available on interviewers sample members and call histories. The interviewer data comprises socio-demographic characteristics, attitudes, behaviours and strategies from a survey of interviewers in the UK. Contact history records from six UK household surveys are available capturing key features of each call to a household, including date, time and outcome of each call and information of the initial reaction and demographic characteristics of the householder at the doorstep. This information was linked to UK Census records and interviewer observations, providing detailed information on both responding and nonresponding sample units to the six surveys.

A multilevel multinomial model is used to examine jointly the different possible outcomes at each call, the influence of interviewer characteristics, call histories and sample member characteristics. The modelling approach accounts for clustering of interviewer calls within households and interviewers. The paper highlights potential implications for survey practice, such as for adaptive and responsive survey designs.

Biography: Gabriele B. Durrant is a Senior Lecturer at the Southampton Statistical Sciences Research Institute (S3RI) at the University of Southampton. She is the principal investigator of two 3-year ESRC-funded research grants in the UK, one on paradata and one on nonresponse analysis. She has research interests in analysis of paradata, interviewer effects, nonresponse in sample surveys, measurement error, and statistical modelling in the social sciences, in particular multilevel modelling. She is guest editor of a special issue on paradata of the Journal of the Royal Statistical Society, Series A, Associate Editor of JRSSA and member of the committee of the Royal Statistical Society Social Statistics Section. She is also the programme coordinator of the UK-based programme of short courses CASS (Courses in Applied Social Surveys).