potentially important component of a neighborhood's social organization was measured by conducting a survey of adult residents in sampled neighborhoods rather than relying exclusively on demographic census-based measures. Collective efficacy combines social cohesion (the extent to which neighbors trust each other and share common values) with informal social control (the extent to which neighbors can count on each other to monitor and supervise youth and protect public order). It is thus a capacity for collective action shared by neighbors. They find that collective efficacy so defined relates strongly to neighborhood levels of violence, personal victimization, and homicide in Chicago, after controlling for social composition and previous crime. One could imagine that lower levels of neighborhood violence and crime might change parenting practices in ways that benefit young children, although that possibility has not yet been tested with these or other data.
Taken together, this picture of at best modest neighborhood influences based on population samples is at odds with more specialized studies focused on very bad neighborhoods. For example, in a sample of patients in a Boston pediatric clinic, Taylor and colleagues (1992) found that 1 in 10 children witnessed a violent event prior to age 6, while Buka and colleagues (Buka and Birdthistle, 1997; Buka et al., submitted) estimated that about 1 in 4 urban youths reports having seen someone murdered during childhood. There are no corresponding figures for children raised in better neighborhoods. Psychiatric problems ranging from posttraumatic stress and aggression to externalizing behavioral disorders are more common among children and youth who witness violence (Singer et al., 1995). Neighborhood violence may also have indirect effects on development, if mothers in physically dangerous neighborhoods restrict their children's interactions with peers and adults (Lipsey and Wilson, 1993).
Among physiological hazards, lead poisoning continues to pose a threat to the healthy development of children, and disproportionately to low-income children of color living in central cities. As described in Chapter 8, excess lead in blood has been tied to such neurobehavioral problems as attention deficits, and poor children are disproportionately at risk for exposure to lead (Brody et al., 1994). Epidemiologists have linked the elevated levels of lead in poor urban children to old housing stock, which often still contains lead-based paint and other environmental contaminants, such as leaded gasoline. Although discontinued in the 1980s, the legacy of leaded gasoline emissions remains through elevated levels of lead in the soil, especially in central cities that are heavily congested with traffic (Mielke et al., 1997). Combining data from the Massachusetts 1990 statewide screening of children 's blood lead levels and the decennial census, Sargent et al. (1995) investigated what characteristics of communities were correlated with heightened risk for lead poisoning. They found that the odds for