This experiment tested different ways of asking respondents to nominate a proxy who could complete a short questionnaire on their behalf in future waves, if they themselves were unable to complete it or moved into an institution. The aim was to seek for the best way to identify who the alternative person to ask is and ensure we have permission to ask about the respondent in future waves.
The experiment tested two different versions of this initial consent question. Respondents who consented were then asked a number of follow-up questions collecting information about the person nominated. These follow-up questions were based on the Stable Contacts module. The experiment also tested the effects of rotating the order of the Stable Contacts and Proxy Nomination modules, where one comes first and the other follows immediately after. The two parts of the experiment are controlled by two variables. For the wave 14 continuing sample (identified by hhorig<19) these were generated within the questionnaire script. These controlling variables are on the record n_indresp_ip:
n_prxver (1/2 each):
1 Proxy nomination question is version 1
2 Proxy nomination question is version 2
n_prxear (1/2 each):
1 Proxy nomination is early
2 Proxy nomination is late
For the Wave 14 refreshment sample (identified by n_hhorig=19), the allocations were made at the household level prior to fieldwork. This was done to test the differences in experiment variances based on whether the allocations occur at the household-level (so everyone in the same household receives the same treatment) or at the individual-level (so treatment is allocated in during the survey to each individual independent of household) impact outcomes, including precision, differently. The allocations are recorded in the following variables on record n_hhsamp_ip:
n_ff_prxver (1/2 each, allocation stratified by sampleorig, ff_incentw14, ff_gridmodew11):
1 Proxy nomination question is version 1
2 Proxy nomination question is version 2
n_ff_prxear (1/2 each allocation stratified by ff_prxver, sampleorig, ff_incentw14, ff_gridmodew11):
1 Proxy nomination before Stable Contacts
2 Proxy nomination after Stable Contacts
The variables affected by this experiment are on record n_indresp_ip:
prxconv1ear, prxconv2ear, prxnomear, prxrelear, prxconv1lat, prxconv2lat, prxnomlat, prxrellat. The variables containing the contact details are not released with the data (Questions prxnameear, prxadd1ear, prxadd2ear, prxtownear, prxcntyear, prxpcodeear, prxtel1ear, prxtel2ear, prxemlear, prxnamelat, prxadd1lat, prxadd2lat, prxtownlat, prxcntylat, prxpcodelat, prxtel1lat, prxtel2lat, prxemllat).
