| Census > ACS Main > How to Use the Data > Using Multiyear Estimates > How Multiyear Estimates are Produced |
|
The American Community Survey (ACS) 1-year estimates are based on data collected in one specific calendar year. For the 2007 ACS estimates, the data from all sample interviews that were obtained between January 1, 2007 and December 31, 2007 are pooled to produce the 1-year data products. Multiyear ACS estimates are based on the data collected over multiple consecutive calendar years. Therefore, using the 2005 - 2007 ACS estimates as an example, the data from sample interviews collected between January 1, 2005 and December 31, 2007 are pooled to produce this 3-year product.
For the ACS 1-year estimates, the geography is based on the boundaries defined on January 1 of the year being tabulated. For the ACS 3-year estimates, the geography is based on the boundaries defined on January 1 of the final year in the 3-year period. Likewise for ACS 5-year estimates, boundaries are defined as of the final year in the 5-year period. For example, Amarillo City, Texas annexed some territory in both 2005 and 2006. The 2005 1-year estimates were published without the boundary changes because they all occurred after January 1, 2005. The 2006 1-year estimates were published with the 2005 boundary changes because they were added prior to January 1, 2006. The 2007 and the 2005 - 2007 estimates were published with all the boundary changes. Note that the Census Bureau does not plan to update the 2005 and 2006 1-year estimates using the January 1, 2007 boundaries.
The ACS 1-year estimates of total housing units are controlled to conform to estimates of housing units from the Census Bureau's Population Estimates Program. The ACS 1-year estimates of total population are controlled by age, sex, race, and Hispanic origin to conform to the official population estimates from the Population Estimates Program. These controls are for the reference date of July 1 of the tabulation year and are updated annually. The ACS 3-year estimates are controlled to conform to a simple average over the 3-year period of the housing and population estimates from the Population Estimates Program. The version of the Population Estimates Program estimates used to obtain these averages are the ones corresponding to the last year of the period. For example, the controls used for the 2005 - 2007 weighting would be equal to the simple average of the 2005, 2006, and 2007 estimates that were produced for the year 2007. Thus, the controls for the 3-year period may not equal the simple average of the controls used to produce the ACS 1-year estimates.
Monetary values for the ACS 1-year, 3-year, and 5-year estimates are inflation adjusted to the final year of the period. For example, the 2005 - 2007 ACS 3-year estimates are tabulated using 2007 adjusted dollars. These adjustments use the national Consumer Price Index (CPI) since a regional-based CPI is not available for the entire country.
Using Multiyear Estimates
What is an ACS multiyear estimate?
What types of geographic areas are published for the 1-year and 3-year ACS?
Choosing between ACS 1-year and 3-year estimates