Environmental Quality and the Wellbeing of Children

There are many reasons for believing that the environment exerts an influence (directly or indirectly) on the wellbeing of children and families. However, while clear evidence is available that low socioeconomic status is associated with lower than average levels of wellbeing, especially among adults, the evidence linking the social and emotional adjustment of children with the quality of the environment is patchy and equivocal. In this paper we focus on three levels of the family environment: the street, the home and the neighborhood. Neighborhood quality was measured by the Vinson Homel social problems index, street-type as residential or commercial/retail, and housing quality in terms of upkeep, floor occupied, availability of playspace and occupancy type. The research hypothesis was that after allowing for community selec tion processes children living in lower quality environments would be less satisfied with various areas of their lives, would experience more negative emotions, and would have more restricted and less positive friendship patterns. The sample comprised 321 families which included a 9?11 year old child, drawn from 18 neighborhoods of Sydney. Neighborhood social problem score and street-type, and some aspects of housing, predicted emotional and social adjustment. Before and after controls for family composition, social class and culture, children living in commercial streets, particularly in inner-city areas, stood out from all others in their feelings of loneliness, dislike of other children and feelings of rejection, worry, fear, anger and unhappiness. Children living in high social problems areas showed a pattern of social constriction rather than maladjustment. These results suggest not simply the influence of social class but genuine community socialization effects. Possible sources of, and mechanisms for, these effects are suggested. A large literature has noted the many ways in which the characteristics of families influence the development of children (McCandless and Evans, 1975). Less attention has been paid to other aspects of the child's environment. Recently, however, Bronfenbrenner (1979) and his colleagues (Cochran and Brassard, 1979; Garbarino, 1982; Garbarino et ai, 1980; Bronfenbrenner et ai, 1984) have drawn attention to the importance of the ecology in which child development takes place; for example the stressfulness of the physical environment, the presence or absence of supportive social networks, and the availability of services such as child care. These authors call on social scientists to pay more attention to ecological settings, noting the sparseness of empirical Social Indicators Research 21 (1989) 133-158. ? 1989 by Kluwer Academic Publishers. 134 ROSS HOMEL AND AILSA BURNS evidence linking these settings with family functioning and child devel opment. Our aim in this paper is to present evidence that features of the family environment do to some extent influence the wellbeing of children and their social adjustment. We focus on three levels of the child's immediate physical and social environment: the home, the street and the neighborhood. Our data are derived from interviews with members of 321 families living in Sydney. These families were resident in 18 neighborhoods which were selected according to a careful probability sampling strategy which maximized the differences in local community wellbeing', as measured by an index derived from 25 objective social indicators. Random selection of children from the rolls of schools servicing the areas provided a spread on the housing and street variables. Having thus selected families differing widely in terms of the quality of their environment, we ask the following questions. Do indices of children's social and emotional adjustment vary depending on the quality of the home, the street or the neighborhood? If so, is this simply because families with certain characteristics are more likely to live in particular environments, or do environmental factors have predictive power after taking account of family characteristics? Which aspects of the environment appear to be most important in influencing children, and what can be said about the processes through which these aspects achieve their impact? Local environment could influence children's wellbeing in a number of ways. Rainwater (1966) distinguishes between human and non human influences, and between direct and indirect paths of influence. Ways in which each of the four possible combinations could affect children have been proposed (e.g.: Gump, 1975; Rutter, 1981). Heavy local traffic for example could have both a direct influence (children frightened and frustrated) and an indirect influence via human media tion (parents become restrictive and nagging in an attempt to protect the child). The available evidence regarding any of these proposed effects is however somewhat limited. Where evidence exists, it refers mainly to the attitudes and behaviors of parents rather than to the experiences of children. For example, the influence of housing quality (Wilson and Herbert, 1978) and of a hostile social environment (Garbarino, 1982; Martin, 1970) on parental child-rearing practices have been described. ENVIRONMENTAL QUALITY 135 The method of our study parallels the five analytic steps proposed by Bronfenbrenner et al. (1984) for establishing a community effect on child development. Their first step, demonstrating variation in community structures, is built into our research design. Their second step, demonstrating psychological variations across communities, is investigated through correlations of environmental variables with child outcomes. By controlling for family social class, culture and family composition in our further analyses, we attempt to carry out steps three and four, namely establishing effects at a community (or environmental) level and distinguishing community processes of socialization versus selection. 'Community socialization' is defined by Bronfenbrenner et al as the effect of living under particular kinds of community conditions; 'community selection' is defined as the processes which affect which groups move into the community, which move out, and which remain. They point to variation in social class composition between commu nities as the most common source of selection effect, which highlights the need to control for differences in social class composition before drawing conclusions about community effects. Accordingly, we have included family income, education and occupational status as control variables, in addition to measures of ethnic status and other family characteristics. Although the family measures included in our models do not cover complex variables like interaction patterns within the family, we justify our variables on the basis that they have been most commonly nominated as influencing children's wellbeing (Links, 1983; McCandless and Evans, 1975; Moen et ai, 1983). The fifth and final step stipulated by Bronfenbrenner et al. (1984), identifying sources and mechanisms of community (or neighborhood) effects, is the most difficult, but we draw on details of our interviews as well as the statistical analyses to make some tentative suggestions. OUTCOME MEASURES AND INDICES OF THE ENVIRONMENT Following Bronfenbrenner (1979) and Garbarino (1982), three aspects of children*s wellbeing and adjustment are measured ? life satisfaction, emotional adjustment and social adjustment. Most measures come from the child interview schedule developed by the U.S. Foundation for Child Development for community surveys of wellbeing and adjustment 136 ROSS HOMEL AND AILSA BURNS (Lash and Sigal, 1976). This measure has been used in a number of countries and extensive normative data are now available (Zill, 1988). Four life satisfaction, four emotional adjustment and nine social adjustment indices were used. Children's satisfaction with four areas of life (self, friends, schoolwork and family) was measured using a set of five line-drawn faces with expressions ranging from very happy to sad. Once it was established that the child understood this to be a scale, he or she was asked to choose (serially) the face that best represented her feelings about each of the four areas. To measure emotional adjustment, children were presented with a set of faces representing worry, fear (being scared), anger and unhappiness. Once the correct choice had been made the question was asked "What about you? Do you (worry, etc.) about things: often, sometimes, hardly ever, or never?" A list of common sources of (worry, etc.) was then read to the child and she or he was asked to nominate which if any was true for her. In respect of fear a further question was asked: "Who is the person you are most