| Census > ACS Main > How to Use the Data > Using Multiyear Estimates > Choosing Between ACS 1-year and 3-year Estimates |
|
In August and September 2008 the Census Bureau released 2007 American Community Survey (ACS) 1-year estimates for geographies with populations of 65,000 or more. In December 2008, the Census Bureau releases the ACS 3-year estimates based on data collected in 2005, 2006, and 2007 for geographies with populations of 20,000 or more. If data users are interested in estimates for geographies of 65,000 or more, starting in December 2008 they have two available sets of estimates to work with. These estimates differ because they are based on data from different time periods and therefore describe the characteristics of an area for two different periods of time.
There are several factors that data users should consider when choosing between the 2007 ACS 1-year estimates and the 2005 - 2007 ACS 3-year estimates.
The ACS 3-year estimates are based on a larger sample and are therefore more reliable but they use data for earlier time periods and are therefore less current. The ACS 3-year estimates include about three times as many sample interviews as the ACS 1-year estimates but a third of those interviews are three years old and a third are two years old. When deciding between ACS 1-year and 3-year estimates users must balance the need for precision against how rapidly the estimate is changing over the 3-year period.
The ACS 3-year estimates are typically preferable for estimates based on small populations. An estimate of a characteristic can have an acceptably small standard error when it applies to the full population of a published geography. However, the estimate of that characteristic may have an unacceptably large standard error when it applies to a subset of the population of the same geography. This is because the size of the population determines how large the sample is for the estimate. An example of such a situation could be the percentage of families with a female householder, no husband present, and with related children under 5 whose income in the past 12 months is below the poverty level.
Use ACS 1-Year Estimates:
Use ACS 3-Year Estimates:
For more information on multiyear estimates and choosing between the one- and three-year estimates see the following document:
Statistical Issues of Interpretation of the American Community Survey's One-, Three-, and Five-Year Period Estimates[678K]
Using Multiyear Estimates
What is an ACS multiyear estimate?
What types of geographic areas are published for the 1-year and 3-year ACS?
Four things you should know about how ACS multiyear estimates are produced