At the end of the interview at each wave, contact details for the respondent are collected, making sure that the ones on file are up to date and correct. In past waves, a list of several contact details is asked about simultaneously: home landline, mobile number, work number, and email address. Given the potential importance of contacting respondents on their mobile phone in upcoming waves, these experiments explored alternative designs to better ensure mobile numbers are collected.
At Wave 12, in the control group, participants were asked the questions as these have been in past IP waves. The alternative condition asked respondents specifically about mobile phone numbers separately from all the other contact details normally asked about. After asking for mobile phone number, these respondents were asked for all the remaining contact details normally asked about (home landline, work number, and email address) in a grouped manner similar to the normal design.
Allocation to this experiment was done at the household level; all respondents within a household received the same set of contact detail questions. Households were allocated randomly and equally to conditions (50% each). The controlling variable for this experiment is l_ff_mobexp_w12 on record l_hhsamp_ip:
1 Control, usual contact detail questions
2 Mobile phone focused contact detail questions
The variables affected by this experiment are in the record l_indresp_ip:
l_rphmob_code, l_rhland_code
At Wave 15, an experiment tested amendments to questions in the Contact Details module, with the aim of increasing the reporting and updating of mobile phone numbers. Households were randomly allocated to one of two treatment groups: one group received the ‘mobile first’ version of the questions that was asked in IP12/IP13; the other group were presented with amended wording aiming to justify/increase compliance with the request for mobile numbers. This affected the first questions in the ‘prioritise mobile number’ version of the Contact Details module used in IP12 (for half the sample) and IP13 (for everyone).
The wording asking for mobile contact details was experimentally allocated equally among households.
The controlling variables are on record o_hhsamp_ip:
ff_mobnow15 (1/2 each, allocation stratified by sampleorig ff_gridmodew15 ff_incentw15 ff_consentcasiw15)
1 = Mobile first version
2 = Increase compliance version
Modules affected:
contactdetails_ip15
