Publication type
Understanding Society Working Paper Series
Series Number
2026-10
Series
Understanding Society Working Paper Series
Authors
Publication date
August 17, 2026
Abstract:
Linking administrative data to surveys provides information that is more detailed and accurate than can be collected directly from respondents. However, increasing use of online data collection may limit the potential of linked data, as respondents interviewed online are consistently less likely to consent to linkage than those interviewed face-to-face. In this research we test the effects of two interventions, designed to affect the respondent’s decision process when answering consent questions. The first intervention provided additional information about the value for science of the linked data; the second used a video of an interviewer reading the consent question, instead of showing the question text on screen. The experiments were implemented in the Understanding Society Innovation Panel, a probability sample of households in Great Britain. The value for science question wording increased consent somewhat for face-to-face respondents and web respondents who saw the consent question as a video. The video question format increased consent by 10 percentage points among web respondents and seemed to have nudged respondents towards making a more trust-based decision. These findings highlight practical approaches for improving consent to data linkage in an increasingly online survey environment.
Paper download#589144