American Community Survey (ACS)

Skip Navigation Menu for Multi-Year Estimates Study

• Overview
  
  
• About the Multiyear Estimates Study
  
  
• Online Data
  
  
• Downloads / FTP Server
  
  
• Statistical Methodology
  
  
• Analytic Reports and Presentations
  
  
• Questions & Answers

About the Multiyear Estimates Study

In preparation for the release of the first official 3-year estimates in 2008 using data collected in 2005 through 2007, the Census Bureau decided to conduct this Multiyear Estimates Study. The goal of the study is two-fold: to test production methods that are planned for the 2008 release and to help data users begin to understand the characteristics of multiyear estimates.

The production methods for standard products, including thresholds and data release rules were used to determine the final set of products. The data use the same disclosure limitation methodology as the original 1-year data. The confidentiality edit was previously applied to the raw data files when they were created to produce the 1-year estimates and these same data files with the original confidentiality edit were used to produce the 3-year and 5-year estimates.

In addition, data profiles for tabulation areas that contained only a small number of households are not being released. In order to prevent the disclosure of the data for these areas through subtracting estimates from nested geographic areas, some additional tabulation areas are also not being released. We are researching alternative options to address disclosure risks for these types of areas for the production of our first 5-year data product in 2010.

Data profiles that include demographic characteristics such as age and sex; social characteristics such as educational attainment and marital status; economic characteristics such as employment status and income; and housing characteristics such as housing occupancy and housing values were produced for a broad set of geographic areas in 34 of the 36 ACS test counties (excludes Fort Bend and Harris counties). The data profiles display the estimated totals and estimated percentages along with their associated margins of error for each single and multiyear estimate.

The types of geographic areas for which data are available are listed below:
(A Complete list of Multiyear Estimates published geographies [Excel 1.3Mb])

The 34 test counties were selected based on a set of pre-specified criteria. One such factor is the sampling rates for these counties are roughly comparable to the sampling rates used in the full ACS production sample, which began in 2005. Other factors include: the size of the county’s population, the proportion of the population in areas classified as hard to enumerate (based on mail response rates in the 1990 census), and the growth or decline in the population since 1990. In addition, the counties selected varied along important dimensions such as racial/Hispanic origin distribution, large numbers of non-city style addresses and presence of American Indian reservations. The 34 ACS test counties are listed below:

A total of 14 data sets were created using data from 1999 to 2005; three 5-year estimates, five 3-year estimates, and six 1-year estimates. Data profiles for the 5-year estimates are available for all geographic areas listed above, including tracts and block groups. For the 3-year estimates, data profiles are available for those geographic areas with a population of 20,000 or more and for the 1-year estimates, those geographic areas with a population of 65,000 or more. The data sets available are listed below: