Publication type
Survey Futures Working Paper Series
Series Number
2025-01
Series
Survey Futures Working Paper Series
Authors
Publication date
May 7, 2025
Abstract:
Many social surveys have adopted web-first sequential mixed mode designs in which first a web questionnaire is offered, then non-respondents are followed up in interviewer administered modes, i.e. either face-to-face (Computer Assisted Personal Interviewing (CAPI) or by telephone (Computer Assisted Telephone Interviewing (CATI). Evidence suggests such designs may be less costly than CAPI or CATI only designs and may produce datasets of higher quality than web-only designs. However, with rising levels of internet access and use, the question arises as to whether this evidence is still valid. We investigate whether follow-ups of web non-respondents with CAPI / CATI are still required to maximise dataset quality, and how this pattern may have changed over time. The analysis uses data from Understanding Society: the UK Household Longitudinal Study (UKHLS). Key findings are: 1) follow-ups are still required to maximise response rates and dataset sizes, though impacts have declined over time; 2) the impact of follow-ups on representativeness (how well datasets resemble study populations) has declined over the period 2012- 2018, with web and web plus CAPI datasets from later years not differing; 3) impacts of follow-ups on the under-representation of hard-to-reach population subgroups, such as older adults and those not in work, have declined and become negligible over a similar timescale; and 4) impacts of follow-ups on non-response biases remaining after non-response weighting, have similarly declined and become negligible over this timescale. We then discuss the implications of our findings for survey practice.
Subjects
Link
https://surveyfutures.net/working-papers/
#588610