Using Within- and Across-neighborhood Change to Assess How Neighborhood Conditions Affect Children’s Achievement

• In Event: New Approaches for Understanding Neighborhood Effects in the Developmental Sciences

Abstract

Children in the U.S. are living in increasingly divergent and economically segregated communities, with levels of neighborhood concentrated wealth and concentrated poverty growing dramatically in recent decades. Empirical evidence, including experimental work, has identified effects of neighborhoods on children’s development, but has provided limited insights into the specific characteristics of neighborhoods that drive these effects. To move this field forward, it is essential to delineate the specific characteristics of more and less advantaged neighborhoods that drive benefits and risks to children. Investment and stress theories suggest that children’s access to promotive neighborhood resources (such as high quality schools and ample cultural and recreational resources) as well as their exposure to neighborhood stressors (such as crime and pollution) are important forces.

In this paper we use a unique dataset created by combining 6 waves of repeated assessments of children’s academic skills (math, reading and science) across a nationally representative sample of children from the ECLS-K followed from kindergarten through grade 5 (N≈17,600) with a broad range of administrative data assessing neighborhood conditions. Neighborhood measures include concentrated advantage and concentrated disadvantage from the American Community Survey; cultural/education/health resources from the Economic Census; food availability, walkability, and measures of educational quality (educational achievement, high school graduate rate, and AP course enrollment) from the Child Opportunity Index; violent crime from the FBI; and airborne pollution from the EPA. We exploit the fact that neighborhood characteristics shift over time both within place, due to the changing conditions, and across place as families move. We use fixed effects models to help adjust for measured time-varying characteristics and unmeasured time-invariant characteristics of children and families that may bias measured links between neighborhood conditions and children’s achievement.

Initial results suggest that increases in concentrated advantage, cultural/educational/health resources, educational quality, walkability, and food availability were associated with improvements in children’s academic achievement. In contrast, increases in pollution and crime were associated with declines in children’s achievement. Further analyses will assess whether effects vary based on shifting neighborhood conditions versus shifts in neighborhood location. Results help to inform both person-based and place-based efforts to improve the neighborhood conditions of children and families to enhance child development.

Authors

• Rebekah Levine Coley, Boston College
  Presenting Author

• Portia Miller, University of Pittsburgh
  Non-Presenting Author

• Bryn Spielvogel, University of Utah
  Non-Presenting Author

• Laura Betancur, Iowa State University
  Non-Presenting Author

• Lorraine Reese Blatt, University of Pittsburgh
  Non-Presenting Author

• Elizabeth Votruba-Drzal, University of Pittsburgh
  Non-Presenting Author