Some methodological aspects of using postal questionnaires with the elderly.

Postal questionnaires are not usually recommended for use with the elderly. This paper describes a methodological study of the use of a postal questionnaire as a method of data collection with older subjects. In particular the problems of generating an adequate response rate, of non-response bias and response quality are addressed. It is shown that postal questionnaires can generate response rates in excess of 80% with older subjects. Questionnaire length was not associated with response rate. Non-response was highest amongst the older age groups suggesting that this method of data collection is subject to non-response bias, thereby understanding the effects of age. The format of question presentation was associated with both response rate and response quality. It is concluded that postal questionnaires are a useful and cost-effective way of collecting survey data from older subjects. In particular such techniques may be effectively employed as a method of routinely screening older people living in the community.