In the initial wave (Wave 1 for GPS and EMBS, and Wave 6 for IEMBS), the issued sample comprised of residential addresses [selected from the Postcode Address file (England, Scotland, Wales) and the Land and Property Services Agency national list of domestic properties (Northern Ireland)]. Only addresses that receive fewer than 50 letters per day were included in the sampling frame to exclude institutions. After the addresses are selected and interviewers visit the addresses, they are instructed to select only private residential addresses and exclude businesses and abandoned addresses (the interviewer outcome variable w_ivfho in datafile w_hhsamp can be used to identify these addresses, w represents the wave prefix). If there were individuals in the responding households who were currently living in an institution (e.g., care homes, University hall of residence), then they were identified as a core member of the sample (Original Sample Member, OSM) but coded as being absent. For example, a university student living in rented accommodation would not be included as an absent household member in their parents’ house because they have a chance of being selected directly as they would be living in a residential address. But a university student living in a college dormitory would not have a chance of being selected as they would be in an institution but would be included as an absent member in their parents’ household. Similarly, someone living in a care home would only be included in their family members’ household as an absent member, if considered by their family member to be a part of that household. However, if the person in the care home does not have any family member who considers them part of their household, they would not be included in the survey. Absent household members may nominate others in the household to provide a proxy interview on their behalf. As explained in the Wave 1 Technical Report, “In such instances where the individual was enumerated as part of the household but were away or could not be contacted during the survey period, a proxy interview was sought, as a last resort.” Of the 298 people marked as absent in Wave 1 (169 of these were away at college or university), a proxy interview was completed on behalf of 108 of them. Those who are absent can be identified using the variable w_absflag, the reason for absence using the variable w_absreason and whether they were interviewed by proxy can be established by using the variable w_ivfio in the datafile w_indall, which provides information from the household grid (where w represents the wave prefix). At the next wave, if they were still in the same institutional accommodation, these absent members were still treated as absent and continued to be included as absent household members for as long as they stayed there. However, if they moved out of that accommodation, even if it was to another institutional accommodation, they would be counted as a “split-off household” from the original issued household, i.e., they will be considered to be in a new separate household. At this point attempts would be made to contact them (if contact information is provided by members of the original household) and the usual following rules will apply. So, in case of students, whenever they move out of the university accommodation (of Wave 1), they are considered to be a “split-off household” and as they will be eligible for interviews with the same following rules as the rest of the sample. Note, even if the student only lives at the residence hall during term-time, and goes home for the holiday, they should still be treated as a “split-off household”. Please note, as explained in the example earlier, if at the initial wave, a student was living in a private residential accommodation (like in a rented apartment by themselves or flatmates), they would have been eligible for selection into the sample in that address and so would not have been included as part of the parental address. In other words, while the sample will initially not represent people in institutions, over time coverage will improve and should eventually, i.e., over the maximum period of stay at such institutions, be complete. There may be larger under-coverage of international students as they will have no chance of selection via their parents’ households, and so would only be included if they are living in private accommodation selected in the first wave. The sample is chosen from private households and so institutions are not part of the core sample. However, if sampled individuals move into institutions after the first wave, they are interviewed if it is possible to do so. The only exception to this is prisons, where we do not seek an interview.
FAQ
