The Bases of Family Status Segregation: A Case Study in Exeter

Abstract

This paper employs 1971 Census data to examine the bases of family status segregation in Exeter, in an attempt to reconcile the recent research findings that there is a wide and contradictory pattern of housing use among households at similar stages of the family cycle with the frequently noted comparatively simple zonal distribution of family status groups. A classification of household types is outlined, and their segregation and distribution in Exeter described. It is demonstrated that progression through the family cycle is not associated with changes in housing tenure or accessibility to the city centre, but that households are associated with housing built at the time of, or soon after, their formation; that is to say, the distribution of family status groups is better understood in terms of inertia than changing housing needs associated with lifecycle changes. It is further shown that members of the two privileged housing classes influence ecological structure more than the disadvantaged classes. A residual segregation which cannot be attributed to housing age is partly explained by the doubling of home ownership in the last 30 years. In conclusion, simple concentric models based on housing age, with the inner zone sectorally divided on the basis of tenure differences, are tested. THIS paper has two complementary aims: to describe and understand, first, the segregation of, and second, the spatial distribution of family status groups in Exeter, a cathedral city with a population of 95 729 in I971. In contrast to the socio-economic status component, which factorial ecologies have consistently isolated as the most important dimension of residential differentiation in the western city (for a recent review see Rees, 1972), family status-the second dimension-is comparatively little studied or understood. None the less, analyses in the United Kingdom (for example, Rosing and Wood, I97I; Lewis and Davies, I974), North America (for example, Anderson and Egeland, 1961; Coulson, 1968), Australia (for example, Johnston, 1969; Timms, 1971) and New Zealand (for example, Curson, I968; Franklin et al., I963) have consistently demonstrated a zonal patterning, with generally a succession of youth-old age-middle agelate youth/early middle age from city centre to periphery. The current wisdom concerning this zonal pattern is succinctly summarized by Jones (I968, p. 429), 'The residential areas of a city, differing in the type of available residential accommodation (detached family dwellings, flats) and in access to various urban facilities (central business district, public transport, schools, recreational facilities), can be regarded as an opportunity structure, and ... competition within this structure by persons with different household requirements and different demand on urban facilities (young couples with children, unmarried workers) tends to produce a residential clustering of persons at different points in the life cycle'. He goes on to suggest that this view which emphasizes family needs seems 'to be supported by Rossi's (1955) finding that a high proportion of residential mobility in cities is generated by shifts in family composition accompanying changes in the life cycle'. Strictly, however, this is an inductive explanation of the zonal pattern in American cities. Rossi's analysis was based on intra-urban migrants in Philadelphia. The most commonly accepted model relating progression through the life cycle to changing needs in terms of accessibility, space, tenure and type of house was proposed by Abu-Lughod and Foley (I960) on the basis of cross-sectional data for all non-farm United States residents in I950. Notwithstanding apparently