Introduction This paper provides background information for a series of new household-level classifications which will be available for the first time in 2001 UK Census output. These will complement existing household classifications proposed for Standard Tables for the 2001 Census. At the time of writing, it is anticipated that versions of three new classifications will be released as univariate tables as part of standard output (Tables UV66, 67, and 69 (1) ). More detailed versions of these three, as well as a fourth, will be available for bespoke tabulations. The concept for a fifth classification is also outlined. In this paper we describe the classifications, summarise results of the consultation exercise that directed their development, and clarify definitions used. Standardisation of data is an essential part of data collection, if the data are to be used by anonymous researchers, without prior knowledge of the individuals from whom the data are collected (Porter,1995, page 46). Classifying data, through grouping different attributes together on the basis of some perceived common experience, mean- ing, or interpretation, is an important part of standardising and objectifying data (page 42). For example, classifications of social class in Britain were first derived by the General Statistics Office in the mid-19th century for the purposes of deriving occupational-based mortality rates, in order to explore the impact that occupation had on individuals' health, and in so doing establish the aetiology of occupational diseases (Higgs, 1991). These early medicalised classifications were substituted by the `professional' model of occupations in the beginning of the 20th century, which not only had greater economic utility, but was also thoroughly empirically tested to ensure that it adequately elucidated observable social and biological diversity (Szreter, 1996). The development of social classifications of this type has focused on summarising the characteristics of individuals, and the derivation, production, and utility of individual- based classifications are key components of social science research. This paper is Alternative household classifications for the 2001 Census Jo Wathan Cathie Marsh Centre for Census and Survey Research, University of Manchester, Oxford Road, Manchester M13 9PL, England; e-mail: jo.wathan@man.ac.uk Clare Holdsworth Department of Geography, University of Liverpool, Liverpool L69 3BX; England; e-mail: clareh@liv.ac.uk Rachel Leeser Greater London Authority, City Hall, The Queen's Walk, London SE1 2AA, England; e-mail: Rachel.leeser@London.gov.uk Received 23 November 2002; in revised form 6 October 2003 Environment and Planning A 2004, volume 36, pages 1101 ^ 1123 Abstract. This paper anticipates the release of a series of new household classifications as part of the standard output from the 2001 UK Census. It contains essential background information for census users interested in using these classifications. In addition to describing the classifications, we outline the background to the classifications, and the consultation undertaken to ensure that each classification reflects real user demand. DOI:10.1068/a35304 (1) Previously given as UV61 ^ 63 in ONS (2001a).