Little is known about the characteristics of respondents and non-respondents to electoral surveys, which is an issue of growing concern as survey response rates fall. Using a procedure that allows all of the addresses sampled in the 2005 British Election Study to be located within the small areas defined for data dissemination with the 2001 census, as well as relevant electoral areas, this study reports on ecological analyses of where those who responded and those who declined to be interviewed lived. It finds that although certain types of area were under-represented in the sampling procedure and others were over-represented, there were only slight differences in the geographies of survey respondents and refusers: people living in low-density areas were slightly more likely to respond; those living in high-density areas with high proportions of their populations suffering from socio-economic disadvantage and with high levels of population mobility were slightly more likely to refuse to participate in the survey.
|Number of pages||27|
|Journal||International Journal of Market Research|
|Publication status||Published - 2006|