Experiments on multiple requests for consent to data linkage in surveys

Abstract

Surveys are increasingly seeking informed consent from respondent to link to several administrative data sources. This paper explores how best to administer consent requests for multiple data linkages in a survey. We experimentally varied the format of the consent request and the order of five data sources in two surveys. The format of the request did not have much effect on consent rates. The order does matter, but in complicated ways that depend on the particular outcomes of interest. Objective knowledge and subjective comprehension of the consent process, and confidence in the consent decision are unaffected by format or order.