Exploring the Effects of Interviewer- and Self-Administered Survey Modes on Record Linkage Consent Rates and Bias

Abstract

In an effort to reduce data collection costs survey organizations are considering more cost-effective means of data collection. Such means include greater use of self-administered interview modes and acquiring substantive information from external administrative records conditional on respondent consent. Yet little is known regarding the implications of requesting linkage consent in self-administered (as compared to interviewer-administered) surveys with respect to consent rates and consent bias. To address this knowledge gap we report the results of a linkage consent study in which sampled employees were randomly assigned to an interviewer-administered (face-to-face) or self-administered (mail or Web) interview in which consent to link federal employment records was requested. We observed a strikingly lower consent rate in the self-administered (53.9 percent) versus the interviewer-administered (93.9 percent) survey mode. However, the impact of mode on consent bias was much less severe as survey-measured correlates of linkage consent did not interact with interview mode. Moreover, while self-administration yielded larger consent biases in the linked administrative variables, on average, compared to interviewer-administration, the average relative magnitude of these biases tended to be small (less than 6 percentage points). We conclude by discussing these findings in the context of survey practice and speculating on their possible causes.