How Neighborhoods Matter for Latino and African-American Children
That neighborhood conditions affect the economic, social, educational, and health outcomes of low-income children is apparent. Determining precisely how these neighborhood aspects contribute to children’s well-being, however, is difficult. Researchers struggle to untangle the causal links between child outcomes and various dimensions of neighborhoods. Parental characteristics that are outside the scope of research, such as why parents decide to move, are particularly challenging because they obscure the independent effect of neighborhoods on children’s lives. Researchers seek to overcome this methodological obstacle, known as geographic selection bias, through natural experiments in which individuals are assigned randomly to neighborhoods. Such designs bypass parental influence, allowing a better understanding of the specific neighborhood mechanisms that affect children’s outcomes. HUD recently published the results of a research study, “Opportunity Neighborhoods for Latino and African American Children,” based on the Denver Child Study of Latino and African-American children living in Denver Housing Authority (DHA) housing. Because DHA randomly assigned families to scattered-site housing, the Denver Child Study is a rare natural experiment that offers researchers a unique opportunity to understand how neighborhoods expand or limit opportunities for low-income children.
See the Fall 2014 edition of Evidence Matters for more on the role of housing and neighborhoods in shaping children’s futures (www.huduser.gov/portal/periodicals/em/fall14/index.html).×
Denver Child Study
DHA operates both large-scale public housing developments and scattered-site, single-family and small multifamily units, assigning families on its public housing waiting list in the latter category to units throughout the city of Denver and the surrounding county. Unlike participants in voucher programs, DHA families are randomly placed in both advantaged and disadvantaged neighborhoods in an allocation process that resembles a natural experiment. Denver Child Study researchers collected comprehensive data on DHA families through a phone survey conducted between 2006 and 2008. Eligible respondents included families who resided in DHA housing for at least 2 years, were Latino or African-American, and had at least one child between ages 0 and 18. Interviewers asked caregivers to provide detailed information about their children. The final survey included 711 households who spent an average of 6 years in DHA housing.
The researchers used the survey to construct a comprehensive panel of children’s outcomes across six areas: physical and behavioral health, exposure to violence, risky behaviors, education, labor market, and marriage and childbearing. In addition to the survey, the researchers drew from various data sources for information about the families’ neighborhoods, including the decennial U.S. census, a local foundation (for information on violent crime and property crime), and the U.S. Environmental Protection Agency (for data on toxic airborne pollutants). Neighborhood indicators included safety, physical environment, social status, ethnic mix, and percentage of foreign-born residents. The Denver Child Study sought to determine which of the outcomes that Latino and African-American children experience were attributable to the neighborhoods in which they lived. Researchers also explored outcomes attributed to neighborhoods that differed by the child’s gender, ethnicity, and developmental stage.
The study finds that certain aspects of neighborhood context are clear predictors of outcomes for Latino and African-American children in Denver. The strength of the neighborhood’s influence, whether positive or negative, on a particular child outcome often depends on the gender and ethnicity of the child as well as whether the neighborhood indicator is being measured contemporaneously with the outcome or cumulatively over time.
- Positive child outcomes across physical and behavioral health, exposure to violence, risky behaviors, education, labor market, and marriage and childbearing are strongly correlated with higher neighborhood levels of foreign-born populations and occupational prestige (the distribution of occupations held across the area) and lower levels of property crime and social problems like drug use. For example, increases in a neighborhood’s occupational prestige are associated with an 81 to 99 percent lower risk of an asthma diagnosis. Researchers credit the finding on occupational prestige across outcomes with the role played in such neighborhoods by strong local networks, role models, and norms.
- Neighborhood indicators do not influence Latino and African-American children equally, and the size of the neighborhood effect for particular outcomes depends on gender and ethnicity. For example, an increase in a neighborhood’s stock of older housing (built before 1940) predicts higher odds of dropping out of school at differing rates for females (349%), for African-Americans (202%), and for Latinos (132%). Researchers suggest that this finding relates to conditions present in older Denver neighborhoods, including greater exposure to violence and greater probability of risky behaviors, which may lead to negative education outcomes. Researchers caution against the implication, however, that older housing necessarily offers a less healthy environment for children.
- The most surprising finding relates to the opposing influences of property crime and violent crime. As expected, higher property crime rates lead to a range of negative outcomes across the board such as an increased likelihood of children witnessing crime. On the other hand, violent crime rates positively impact a range of outcomes such as longer periods of time before children witness violence. Researchers believe this unexpected finding reflects caregivers’ efforts to shield their children from violent crime by limiting their activity spaces closer to home, thus increasing parental monitoring of behavior.
These findings offer insights into the lives of low-income minority children whose families apply for public housing assistance. Although researchers caution against broad generalizations given the complex web of neighborhood factors, the study nonetheless offers important lessons on how neighborhoods matter for low-income Latino and African-American children. Federal programmatic and policy initiatives, particularly those intended to improve neighborhood conditions to improve the development and life chances of low-income children, stand to benefit from the study’s findings. In addition, the Denver Child Study suggests some parallels to the Moving to Opportunity (MTO) demonstration, another much-studied natural experiment featuring the random assignment of low-income families to various neighborhoods in five U.S. cities. Although the two studies cannot be strictly compared, due to differences in measurement and analytical design, the researchers note some similarities in neighborhood effects as well as some differences that could feasibly be explained with further research.