How Does the ACS Measure Up in the Bronx, New York, Test Site?

• Activity Number: 170
• Type: Invited
• Date/Time: Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
• Sponsor: Social Statistics Section
• Author(s): Joseph J. Salvo*+ and Arun P. Lobo and Timothy Calabrese
• Companies: NYC Department of City Planning and New York City Department of City Planning and NYC Department of City Planning
• Keywords: American Community Survey ; data quality ; small-area data ; item allocation ; city planning ; neighborhoods

Abstract

The American Community Survey (ACS) is slated to replace the 2010 Census long form, based on the premise that the ACS is capable of creating an accurate and useful socioeconomic picture of sub-county areas, such as neighborhoods in large cities. This study compares estimates from the 1999-2001 ACS and the 2000 Census for 88 neighborhoods in Bronx County, New York. It asks whether estimates from the ACS can replace those from the census sample by drawing a picture of similarities and differences; meaningful differences are defined in this study as those that can affect the work of a large city planning agency. The results show that, despite some important conceptual differences, most estimates from the 1999-2001 ACS do not differ from those in the 2000 Census in a meaningful way; however, there are some important exceptions. Examining differences in the context of the data collection effort (mail return rates, household nonresponse rates) and in terms of data quality (item allocation levels) leads to a better understanding of differences between the two surveys.