You are here:
CHAPTER 9: USING CENSUS DATA TO APPROXIMATE NEIGHBORHOOD EFFECTS
Citation:
MESSER, L. C. AND J. S. KAUFMAN. CHAPTER 9: USING CENSUS DATA TO APPROXIMATE NEIGHBORHOOD EFFECTS. Edition 1, Chapter 9, Jossy-Bass (ed.), Methods in Social Epidemiology. John Wiley & Sons Incorporated, New York, NY, , 209-238, (2006).
Impact/Purpose:
U.S. Census remains a rich and convenient data source for characterizing neighborhood environments and exposures in the United States. For researchers who wish to learn how societal structures influence health and disease outcomes, sociodemograhic census data are useful because they can offer insight into aspects of community stratification, opportunity structures and social conditions (7-10).
Description:
INTRODUCTION Despite the development of innovative neighborhood data collection methods, such as systematic social observation (1, 2), and the utilization of novel administrative data sources including delinquent tax records, homelessness shelter utilization, reports of housing violations (3) and crime reports (4-6) the U.S. Census remains a rich and convenient data source for characterizing neighborhood environments and exposures in the United States. For researchers who wish to learn how societal structures influence health and disease outcomes, sociodemograhic census data are useful because they can offer insight into aspects of community stratification, opportunity structures and social conditions (7-10). Given the data’s ready availability for researchers and extensive geographic coverage, understanding how to employ census data in research on neighborhood or area effects on health is an important social epidemiology method. Drawing on literature from geography, social epidemiology and sociology, this chapter will offer an introduction to census geography and describe the various ways that the US Census divides physical space. This is followed by an overview of the various types of available census data with special emphasis on information collected in the decennial census that can be used to characterize neighborhoods. The chapter next provides a brief description of the ways in which census data have been used previously in association with health outcomes and will offer a worked example of using census data to estimate neighborhood influences on adverse birth outcomes. Significant racial disparity exists for all birth outcomes, despite considerable research and various interventions. Birth outcomes were chosen to illustrate the use of census data to approximate neighborhood effects because neighborhood-level factors may contribute to these persistent racial disparities and this topic is an active area of research in perinatal epidemiology. The chapter concludes with a discussion of the strengths and limitations of using census data to approximate neighborhood effects. This document has been reviewed in accordance with the U.S. Environmental Protection Agency policy and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.