Within-household selection methods for probability web surveys

As part of our survey methodology seminar series - organised alongside City, University of London and NatCen Social Research - a webinar was held on 5 November 2020.

This event was held in association with GenPopWeb2 - a network of UK-based academic and non-academic partners who share knowledge and collaborate in the area of online data collection in social surveys.

Speakers:
Kristen Olson (University of Nebraska-Lincoln), Patten Smith (Ipsos MORI) and Joel Williams (Kantar UK Public) presented their research in a seminar chaired by Peter Lynn, (Institute for Social and Economic Research). Debbie Collins (NatCen) opened proceedings.

Event Series:
City, University of London - European Social Survey HQ - NatCen Social Research survey methodology seminar series

Abstract:
Within-household selection is a challenge for address-based push-to-web surveys. Here in the UK, we have moved away from (quasi) random methods of respondent selection because postal instructions are not necessarily followed and a wrong selection is made in a large proportion of households. Instead, the most common methods are (1) asking all eligible adults to take part and (2) asking up to any two eligible adults to take part. However, the methods we are using have their own challenges. This seminar, a joint venture with GenPopWeb2, brings together leading methodological experts to discuss the pros and cons of the current methods of choice in the UK, consider these alongside other methods, and identify areas for further research and experimentation.

The seminar will be chaired by Peter Lynn, Professor of Survey Methodology and Director of the Institute for Social and Economic Research (ISER), University of Essex. Peter’s research interests include survey sampling, weighting, non-response, measurement and survey quality. He is Chair of the European Social Survey Sampling and Weighting Panel, Editor of "Methodology of Longitudinal Surveys" (Wiley) and "Improving Survey Methods" (Routledge) and was awarded the Royal Statistical Society Guy Medal for services to survey research.

About the Speakers:
Kristen Olson is Leland J. and Dorothy H. Olson Professor and Director of the Bureau of Sociological Research at the University of Nebraska-Lincoln, USA. Kristen’s research is focused on why measurement, nonresponse, and coverage errors occur in surveys. Kristen and her colleagues developed the “confirmation question” approach to within-household selection in household mail surveys. Kristen recently chaired a Task Force for AAPOR on surveys that transition from interviewer-administered to self-administered or mixed mode data collection, is lead editor of the recently published book Interviewer Effects from a Total Survey Error Perspective, and is Editor-in-Chief of the survey methodology side of the Journal of Survey Statistics and Methodology.

Patten Smith (FAcSS) is Director of Research Methods at Ipsos MORI and has worked on high quality social surveys for around 40 years. Patten is visiting professor at Surrey University and was Chair of the SRA from 2011 to 2017. He and his colleagues developed the ‘up to two adults’ selection methods, first used in the Active Lives Survey.

Joel Williams is Head of Methods within Kantar’s UK Public division. He and his colleagues developed the first official statistics push-to-web survey in the UK - the Community Life Survey - and Joel has written widely on the topic, including “An introduction to address-based online surveying” for the SRA journal Social Research Practice, and “Five methods of within-household sampling: does it matter which one we use?” for DCMS.