Participation in surveys such as the British Gambling Prevalence Survey (BGPS) 2007 is a sequence of many steps – from the selection of a sample of addresses to the transferral of full interview data from respondents to the survey organisation. Each step of the process represents an opportunity for non-response to occur - potentially introducing bias into key survey estimates as the behaviour and views of those who do not respond may be different from those who do participate. The aim of this report was to examine potential non-response biases to the BGPS 2007 in three main areas. These were: 1) Household non-response - where the selected household does not participate for whatever reason. 2) Individual non-response - where an eligible respondent aged 16 and over, within a co-operating household, does not participate. 3) Item non-response – where a respondent chooses not to answer certain survey questions. In addition, this report examined the adjustments that were made to the BGPS 2007 data to account for non-response and looked at where improvements could be made for future studies within the BGPS series.
