Examining Small-Scale Geographic Estimates from the ACS 5-Year Data

Robert Kominski, U.S. Census Bureau
Thomas File, U.S. Census Bureau

One of the main advantages of the Census Bureau’s American Community Survey (ACS) design was that the program would be able to provide routine characteristic estimates for small geographic units such as block groups and tracts. This aspect was necessary as the ACS has been intended to replace the small scale geographic data from the now-discontinued decennial long form. This paper examines several important policy-relevant estimates for tracts from the first ACS 5-year datafile. These data, based on collections in the years 2005-9, are looked at both in terms of their estimated levels, as well as the ability to show distinguishing variation across tracts. Using data for Washington D.C., we examine two measures: the proportion of young adults who are high school dropouts, and the proportion of adults with low English-language ability. Maps will show variability of the estimates, with statistical measures providing an idea of data reliability.

  See paper

Presented in Session 75: Using the American Community Survey