Skip Header

Celia G. Boertlein and Alan K. Peterson
Component ID: #ti1105932096


In January through March of 2006, the American Community Survey (ACS) conducted the first test of new and modified content since the ACS reached full implementation levels of data collection. The results of that testing will determine the content for the 2008 ACS. To meet the primary objective of the 2006 ACS Content Test, analysts evaluated changes to question wording, response categories, instructions, or examples relative to the current version of the questions.


The primary objectives of the 2006 ACS Content Test work on Residence 1 Year Ago (Migration) were (1) to collect complete and appropriate address information for recentmovers within the United States and (2) to collect complete and appropriate previous residence information for movers to the United States from Puerto Rico.The 2006 ACS Content Test compared two versions of the residence one year ago (migration) question set. The Control version replicated the current ACS question. The Test version modified the migration question by including the address (structure number and street name), collecting geographic information down to the place level within Puerto Rico for persons living in the U.S. at the time of the survey whose previous residence was Puerto Rico. Primarily because of space constraints and concern for increased respondent burden, the city limits indicator in the question was eliminated from the Test version.

Based on the empirical results per the selection criteria, the Test version of the residence one-year ago question performed better than the Control version. The Test version did not meet all criteria and some criteria were met weakly, with just equivalent results on Test versus Control. However, the Test version met the minimum criteria for selection. In addition, the Test version led to coding 81.2% of the migration responses to the block level. This block-level coding is not possible with the Control version. The ability to code to the block level using TIGERLine ID and Side allows higher quality and relatively bias-free data on the Test version. Block-levelcoding also allows the ACS staff to convert previously-collected data to current boundaries in a much more accurate basis than currently available. Boundary changes, name changes, and code changes at the place and county level are ongoing and become even more problematic when combining multi-year estimates.

  Is this page helpful?
Thumbs Up Image Yes    Thumbs Down Image No
Comments or suggestions?
No, thanks
255 characters remaining
Thank you for your feedback.
Comments or suggestions?
Back to Header