Do the effects of interviewer administered follow-ups of web non-respondents on dataset quality differ between overall datasets and population subgroup datasets?

Abstract

This paper investigates whether the impacts of follow-ups of web non-respondents with CATI (Computer Assisted Telephone Interviewing) on dataset quality differ between overall and substantively important population subgroup datasets. The analysis uses data from Understanding Society - The UK Household Longitudinal Study (UKHLS). The impacts of follow-ups of non-respondents on response rates, representativeness and weighted dataset quality are quantified. Implications of findings for survey practice are then discussed.