The Understanding Society questionnaire includes questions designed to identify people who provide informal care and seeking to measure how much time they spend providing care.
The existing questions adopt an approach of asking about informal care in general. Alternative possible approaches exist, and it is important to understand whether these alternative approaches can identify carers who might not be identified using the existing approach, or more accurately measure how much time they spend caring.
This experiment aimed to study an alternative set of questions, comparing them against the existing questions.
The new questions tested can be characterised as ‘activity-based questions’. A series of questions were asked of each respondent, asking them how long they spend each week undertaking a certain class of caring activity.
To facilitate within-person analysis of the results, a question-order experiment was used. All respondents were asked both the existing and the new (activity-based) questions. Respondents were randomly allocated to be asked one set of the questions early in the questionnaire and the other later in the questionnaire. The two places where the blocks of questions were asked were spaced out within the questionnaire to lessen the impact of the same information being requested too close together.
Allocations
The variable controlling allocation to treatments, randomised at the household level:
ff_careexpw17
(1/2 allocated to each condition, allocation stratified by sampleorig, ff_incentw17, ff_gridmodew17, {recoded version of p_hhmodes grouping (1,4) (2) (3, 5, 7)})
1 Activity-based question asked late
2 Activity-based question asked early
Modules affected
caringearly_ip17, caringlate_ip17
