¹ Townsend, P.The Family Life of Old People. Penguin, Harmondsworth, 1963.Google Scholar

² Laslett, P. ‘Mean household size in England since the sixteenth century’, in Laslett, P. (ed.), Household and Family in Past Time. Cambridge University Press, Cambridge, 1972.CrossRefGoogle Scholar

³ Wroe, D. C. L. ‘Ageing and public expenditure in the United Kingdom’, in Carver, V. and Liddiard, P. (eds), An Ageing Population. The Open University Press, 1978.Google Scholar

⁴ Abrams, M.The future of the elderly. Futures, II (1979), 178–184.CrossRefGoogle Scholar

⁵ Office of Population Censuses and Surveys. General Household Survey 1980. HMSO, London, 1982.Google Scholar

⁶ Laslett, P. ‘The history of ageing and the aged’, in Carver, V. and Liddiard, P. (eds), An Ageing Population. The Open University Press, 1978, p. 12.Google Scholar

⁷ Evandrou, M., Arber, S., Dale, A. and Gilbert, G. N. Who cares for the elderly?: Family care provision and receipt of statutory services, in Phillipson, C., Bernard, M. and Strang, P. (eds), Dependency and Interdependency in Old Age. Croom Helm, London, 1986.Google Scholar

⁸ Laslett, P., 1978, op. cit. p. 9.Google Scholar

⁹ Eversley, D. ‘New aspects of ageing in Britain’, in Haraven, T. K. and Adams, K. J. (eds), Ageing and Life Course Transitions: An Interdisciplinary Perspective. Tavistock, London, 1982, pp. 245–266.Google Scholar

¹⁰ Laslett, P., 1972, op. cit., table 4.14.Google Scholar

¹¹ Ibid. p. 152.

¹² Anderson, M. ‘The relevance of family history’, in Anderson, M. (ed.), Sociology of the Family. Penguin, Harmondsworth, 1980.Google Scholar

¹³ Qureshi, H. and Walker, A.The Caring Relationship (forthcoming).Google Scholar

¹⁴ Shanas, E., Townsend, P., Wedderburn, D., Friis, H., Milhoj, P., and Stehouwer, J.Old People in Three Industrial Societies. Routledge and Kegan Paul, London, 1968, P. 153.Google Scholar

¹⁵ Townsend, P. and Wedderburn, D.The Aged in the Welfare State. G. Bell and Sons, 1965.Google Scholar The index of physical disability used is based upon the ability to carry out six daily activities without help. These are: (1)ability to go out and walk down the road; (2) ability to get around the house; (3) ability to climb up or down stairs/steps; (4) ability to take a bath/shower or to wash all over; (5) ability to get in and out of bed; (6) ability to cut toenails. A score of two was given if a particular activity could not be done at all, and a score of one if it could only be done with difficulty. The scores for the six activities were added, so that each elderly person could score between o and 12. The items were found to form a Guttman scale. The category ‘none or slight’ indicates that all the tasks can either be done without difficulty, or that they can be done without assistance although difficulty may be encountered with one or two. The ‘moderate/severely disabled’ category includes those who have difficulty with at least two tasks and also need help with or cannot manage at all.

¹⁶ Walker, A.The social creation of poverty and dependency in old age. Journal of Social Policy, 9 (1980), 49–75.CrossRefGoogle Scholar

¹⁷ Hunt, A.The Elderly at Home. HMSO, London, 1978.Google Scholar

¹⁸ Jolliffe, I., Jones, B., Knapp, M., and Morgan, B.Classifications of the elderly population. Ageing and Society, 2 (1982), 331–355.CrossRefGoogle Scholar

¹⁹ Shanas, et al. , op. cit.Google Scholar

²⁰ Ibid.

²¹ Young, M. and Willmott, P.Family and Kinship in East London. Pelican, Harmondsworth, 1957.Google Scholar

²² Shanas, E. et al. , op. cit.Google Scholar

²³ Sheldon, J. H.The Social Medicine of Old Age. OUP, London, 1948.Google Scholar

²⁴ Hunt, A., op. cit.Google Scholar

²⁵ Eversley, D., op cit.Google Scholar

²⁶ We are able to determine whether or not those non-elderly members living within the household are sons and daughters by looking at the number of people who are in the elderly person's family unit. The GHS allocates unmarried sons and daughters sharing the same household as their parents to the same family unit; therefore where the family unit of the elderly person contains at least three people we know, by definition, that these must be the respondent, his/her spouse, and an unmarried child of the respondent. The household may also contain other members – for example a lodger – but in this case he/she would be in a separate family unit and would not be numbered in the family unit of the respondent. To be in a separate family unit, the non-elderly adult may be an offspring who is divorced or separated, a relative other than a child, or an unrelated lodger.

²⁷ Shanas, et al. , op. cit.Google Scholar

²⁸ Laslett, P., 1972, op. cit.Google Scholar