Residential Turnover and Family Composition of Home Owners in Four Subdivisions in Natick, Massachusetts

Abstract

T HE study here reported* examines the interrelationship of three variables in residential subdivisions-age of dwelling, length of residence, and age distribution of children-and traces certain forces operating to modify family composition of the residents as the subdivisions age. New subdivisions attract families with very young children. length of residence distribution changes as the subdivision ages, causing marked changes in the age distribution of children 0-17 years. findings have implications both for a theory of residential mobility and the practical problem of predicting school enrollments. Four subdivisions in Natick, Massachusetts were selected for study. This residential community, with a 1950 population of 20,000, had over a 40 percent increase in population in the last decade. It is located in the Boston Metropolitan Area, about 20 miles west of Boston. Housing to accommodate this rapid growth of population has been built in the form of the residential subdivision. Four of these subdivisions, similar in size and type and value of dwelling, but varying in age, were selected for study.' For purposes of replication, the following analysis is carried through for two cases. Case I includes a subdivision built in 1943 and an adjacent older subdivision built in 1937. Case II includes a subdivision built in 1947 and an adjacent older subdivision built in 1933. See Table 1. relation between two of the variables-age of subdivision and age distribution of children of occupants-is pictured in Figure 1. Basic data are presented in Table 2. In both Case I and Case II, occupants of the subdivisions built about five years ago (dotted lines in Figure 1) have more children aged 0-5 years than occupants of the adjacent subdivisions built about 15 years ago.2 In both cases, occupants of the subdivisions built about five years ago have fewer children aged 12-17 years than occupants of the adjacent subdivisions built about 15 years ago.3 number of children aged 6-11 years per 100 dwelling units does not differ in fiveand fifteen-year-old subdivisions in either case. One of the forces operating to change the age distributions of children in subdivisions as the subdivisions age is the change in the length of residence distribution associated with aging of the subdivision. Table 3 shows the effect of aging of subdivision on the length of residence distribution. When the subdivision is new, all of the occupants are new residents. As the subdivision ages, the proportion of new residents falls. In the subdivisions which are about 15 years old, approximately half of the occupants have lived in the subdivision less than five years. Characteristics of the out-migrants are not known. But it is probable that some have lived in * Point of departure for the research was a project directed by William L. C. Wheaton, Department of Regional Planning, Harvard, and sponsored by the Housing and Home Finance Agency, Office of the Administrator, on the subject, The Cost of Municipal Services for Residential Areas. author is solely responsible for statements or interpretations appearing in this paper. I All dwelling units are owner-occupied, 1I-story single family, and valued at between $10,000 and $15,000. 2 Differences between means significant at .05 level. 3 Differences between means significant at .05 level.