Authors
Summary
The expansion of internet access in many countries raises the question of whether it is now feasible to conduct web-only surveys of the general population without compromising representativeness, i.e. how well the set of respondents reflects the study population. This study examines the representativeness of web-only surveys of the general population in the United Kingdom. Using data from the United Kingdom Household Longitudinal Study main survey (2009-2022) and Innovation Panel (2012-2023), we estimate coefficients of variation of response propensities to assess the representativeness of internet users and web respondents. The results show a significant reduction in internet exclusion. Internet users have become increasingly representative of the general population, though some gaps persist among older adults and those less educated, and other disadvantaged groups. Web survey respondents have also become more representative of the general population over the last decade, but they are still systematically less representative than internet users.