Published online by Cambridge University Press: 13 June 2016
Technology and the decreased cost of survey research have made it possible for researchers to collect data using new and varied modes of interview. These data are often analyzed as if they were generated using similar processes, but the modes of interview may produce differences in response simply due to the presence or absence of an interviewer. In this paper, we explore the differences in item non-response that result from different modes of interview and find that mode makes a difference. The data are from an experiment in which we randomly assigned an adult population to an in-person or self-completed survey after subjects agreed to participate in a short poll. For nearly every topic and format of question, we find less item non-response in the self-complete mode. Furthermore, we find the difference across modes in non-response is exacerbated for respondents with low levels of cognitive abilities. Moving from high to low levels of cognitive ability, an otherwise average respondent can be up to six times more likely to say “don’t know” in a face-to-face interview than in a self-completed survey, depending on the type of question.
Andrew Gooch, Postdoctoral Fellow, Institution for Social and Policy Studies and the Center for the Study of American Politics, 77 Prospect Street, New Haven, CT 06511 (email@example.com). Lynn Vavreck, Professor of Political Science and Communication Studies, University of California, Los Angeles, 4289 Bunche Hall Los Angeles, CA 90095 (firstname.lastname@example.org). This research is supported by a grant from the National Science Foundation (SES-1023940). The authors thank Brian Law for managing the project at the MGM Grand and Felipe Nunes, Sylvia Friedel, Gilda Rodriguez, Adria Tinnin, and Chris Tausanovitch for their participation in Las Vegas. Doug Rivers and Jeff Lewis provided programming support; John Aldrich, Larry Bartels, Alan Gerber, Gary Jacobson, Simon Jackman, Vince Hutchings, Gary Segura, John Zaller, and Brian Humes helped with the design of the experiment. Finally, the authors are grateful to Mike Thies who provided valuable feedback on drafts of the paper. To view supplementary material for this article, please visit http://dx.doi.org/10.1017/psrm.2016.20