The SDC Steering Committee participated in a conference call to discuss the completion of the ACS Review Pilot Project. States supplied feedback on the 2009 unresolved ACS measures and this information was provided to the Census Bureau. The conference call was used to discuss the types of feedback that were supplied and the next steps for future reviews. Here are the minutes from the conference call.
The joint steering committees (FSCPE, CIC, and SDC) all signed this letter on behalf of the respective networks to Director Groves in support of a statistical calculator to be incorporated into American FactFinder. The statistical calculator requirements were also forwarded to the Bureau with the letter.
A response was received from the Director indicating that the new AFF will be adding some functionality similar to the statistical calculator recommendations in 2012.
Daniel Sheres at the D.C. Office of Planning has compiled all the equations for calculating ACS statistics (margins of error, statistical significance, etc.) into the following spreadsheet:
Doug Hillmer has prepared the following presentation and documentation on the statistical calculator. This tool will allow users to calculator margins of error (MOE) to be used in conjunction with data products published from the American Community Survey.
Click here for the PowerPoint that was presented as a webinar on October 15, 2010 for APDU (Association of Public Data Users).
Click here to view the requirements documentation for the statistical calculator.
Article written by John Blodgett of Missouri: The American Community Survey vs. the Decennial Census Long Form: Are We Better Off Now Than We Were a Decade Ago?
Spreadsheet to Calculate ACS Margins of Error and Statistical Significance for Sums, Proportions and Ratios
This calculator, updated 2/23/2009, from the New York Data Center allows users to produce the margins of error for the ACS when combining counts or proportions for a number of geographic areas. It also allows users to test significant differences between two estimates.