American Community Survey (ACS)

American Community Survey Alert Number 23

(released June 15, 2004)

PDF Format of Alert #23

Informing you about news, events, data releases, congressional actions, and other developments associated with the American Community Survey (ACS).

News in this Alert

• ACS 1999-2001 and Census 2000 Comparison Study Added to ACS Web Site
• Release of Series of Analytical Reports
• Release of Online Data Tables Accompanying Analytical Reports
• Contact Us

ACS 1999-2001 and Census 2000 Comparison Study Added to ACS Web Site

The ACS is designed to produce aggregated multi-year estimates for small geographic areas. These estimates are to replace the traditional long form estimates. The U.S. Census Bureau last week released results from a study comparing the ACS data from 1999 - 2001 in the 36 test counties to the corresponding Census 2000 sample data. These estimates, along with measures of data quality, were compared at the county and census tract level. In addition to several analytical reports, the associated data tables and files are available on the ACS web site. Please visit the new section at http://www.census.gov/acs/www/AdvMeth/acs_census/ to access the reports and associated county and census tract-level data.

Release of Series of Analytical Reports

This release includes several analytical reports comparing ACS 1999-2001 and Census 2000 sample data from 36 test counties. The reports are:

• Meeting 21st Century Demographic Data Needs-Implementing the American Community Survey, Report 7: Comparing Quality Measures: The American Community Survey's Three-Year Averages and Census 2000's Long Form Sample Estimates."

  Here are two key findings from this report:
  
  

  • Census 2000 long form sample households were more likely to mail back their questionnaires than ACS households. On the other hand, ACS households that did not mail back their questionnaires were more likely to respond in follow-up operations than Census 2000 long form households that had not mailed back their questionnaires. The net result was a higher overall weighted response rate in the ACS.
  
  
  
  • Questionnaires for ACS occupied housing units tended to contain higher proportions of completed items than did long form questionnaires for the Census 2000 sample occupied housing units.

• Meeting 21st Century Demographic Data Needs-Implementing the American Community Survey, Report 8: Comparison of the American Community Survey's Three-Year Averages and the Census Sample for a Sample of Counties and Tracts.

  This report is the first to compare ACS multi-year estimates and Census 2000 sample data, and is part of a larger effort to assist traditional long form data users' transition to the ACS. Here are some key findings from this report:
  
  

  • At the county level, the majority of American Community Survey estimates are in agreement with the Census 2000 sample estimates. Most of the differences between the American Community Survey and Census 2000 sample estimates are not statistically significant.
  
  
  
  • Some county level differences have been traced to specific variations in the questionnaires, edits, or other methods used to carry out the American Community Survey and Census 2000. Some of the variables that showed the largest differences, and for which at least a partial understanding of why they occurred, were race, disability status and school enrollment.
  
  
  
  • Differences at the tract level are more difficult to detect because of the smaller sample sizes. Although the general patterns observed at the county level are also seen at the tract level, the additional variability at the tract level tends to reduce the number of statistically significant differences.

The release also includes four reports prepared by local area experts that examine county and census tract-level data from a subset of test counties. Researchers contemplating analyzing these data are encouraged to examine these reports. The reports are:

• "Small Area Data Quality: A Comparison of Estimates--2000 Census and the 1999-2001 ACS Bronx, New York Test Site" by Joseph Salvo, Peter Lobo and Timothy Calabrese. To view the report, click here.



• "The 1999-2001 American Community Survey and the 2000 Census: Data Quality and Data Comparisons-Multnomah County, Oregon" by George Hough and Dave Swanson. To view the report, click here.



• "Comparison of Census 2000 and American Community Survey 1999-2001 Estimates-San Francisco and Tulare Counties, California" by Linda Gage. To view the report, click here.



• "American Community Survey and Census Comparison-Final Analytical Report-Oneida and Vilas Counties, Wisconsin-Flathead and Lake Counties, Montana" by Paul Van Auken, Roger Hammer, Paul Voss, and Daniel Veroff. To view the report, click here.

Release of Online Data Tables Accompanying Analytical Reports

In addition to the analytical reports, online and downloadable data files also are available:

Online Data: Online data profiles and quality measures data tables for each of the 36 ACS test counties are available on the Web site http://www.census.gov/acs/www/AdvMeth/acs_census/allcounty.htm. The data profiles for the ACS 1999-2001 and Census 2000 Comparison Study are similar to the standard ACS single-year profiles and to the Census 2000 Demographic Profiles. Each of the profiles contains the 3-year average ACS estimate, the Census 2000 Sample estimate, the difference between the estimates, the margin of error of the difference, and an indicator of statistical significance at the 90% confidence level.

Downloadable Data Files: In addition, the online data tables and additional data files are available to be downloaded. Access to these data files and corresponding documentation also are available on the Web site http://www.census.gov//acs/www/AdvMeth/acs_census/doc.htm.

Contact Us

If you have questions or comments about the American Community Survey, please call (888)456-7215 or e-mail cmo.acs@census.gov.

General information about this mailing list is available at: http://lists.census.gov/mailman/listinfo/acs-alert.

If you choose to unsubscribe or change your list options, you may do so by visiting this site and entering your e-mail address at the bottom of the page under the ACS-Alert subscriber's section. This will take you to an options page where you may request your password, unsubscribe, or set other options. If you have questions about this list, please send a message to acs-alert-admin@lists.census.gov.

[PDF] or denotes a file in Adobe’s Portable Document Format. To view the file, you will need the Adobe® Acrobat® Reader available free from Adobe.