Estimating the prevalence of disability in the community: the influence of sample design and response bias.

An estimate of the prevalence of physical disability in the community based upon a sample survey may be influenced by the sample design and the response to the method of data collection employed. In this paper we describe a postal survey of a sample of households in the London borough of Lambeth and the procedures used for calculating the influence of these factors on the estimate produced. These procedures can be used to adjust the estimate to take account of the relative chance of households falling into the sample and to correct for non-response bias.