The request was for consent to link data held by the Department for Work and Pensions (DWP) to survey responses.
Participants who completed in CAPI were equally allocated to one of three groups; those who complete online were equally allocated to one of five groups:
- [All modes] Standard: standard ‘BENLINK’ question, previously carried in IP, most recently at IP10.
- [All modes] Benefits/risks: version of question focusing on the benefits to the respondent of consenting and reassurance about things they might perceive as risks to them.
- [All modes] Value for science: version of question focusing on the benefits to research and policy of having people consent.
——- - [Web only] Reasons for/against consenting: respondents will be asked to list reasons for and against consenting before being asked the consent question.
- [Web only] Objective understanding: respondents will be asked knowledge-check questions about information they have been told about the data linkage, before being asked the consent question.
(CATI respondents were not asked these questions.)
Within-script randomisation was used to allocate participants at the individual level.
Allocations
The variable controlling allocation to treatments, randomised at the individual level within the questionnaire script, varies depending on whether the interview was conducted face-to-face or via the web.
Allocation variable for face-to-face respondents:
congrpftf
(1/3 allocated to each of conditions 1, 2 and 3 for face-to-face respondents)
1 Standard DWP consent
2 Benefit-risk DWP consent
3 Research and policy DWP consent
Allocation variable for web respondents:
congrpweb
(1/5 each allocated to each of the five conditions for web respondents)
1 Standard DWP consent
2 Benefit-risk DWP consent
3 Research and policy DWP consent
4 Reasons pro con DWP consent
5 Understanding DWP consent
An apparent inconsistency between these allocation variables and the mode of completion variables for 12 cases is documented in the section on known data issues, under Inconsistency between mode of allocation variable and mode of completion.
Follow-up questions
After the consent questions, follow-up questions were asked of respondents in all arms, seeking to measure subjective understanding of the task, confidence in the consent decision, the consent decision process, and subjective effort in answering the question. These were variants of follow-up questions previously asked in relation to similar consent requests (related to linking HMRC data rather than DWP data), at Waves IP11 and IP15.
In addition, questions seeking to measure respondents’ objective understanding of the consent request were asked as follow-up questions after the consent question, except for respondents in arm 5, who had already have answered such questions prior to the consent question.
Modules affected
benefitconsents_ip17, scabenefitsconsentfollowup_ip17
