I’ve received several requests regarding racial maps for the 2005-2009 American Community Survey 5-year estimates. Having looked over the data, I still remained concerned over the use of 5 year estimates because of the reliability of the estimates. Estimates at the census tract level even for large population groups (i.e., blacks, whites) are not so good in certain areas. To illustrate this point, I’ve produced a presentation containing several maps with explanation, as well as an accompanying data file.  The maps can be downloaded in the Seniors Count! download center, and the data is available upon request.

I have chosen race to illustrate my point about unreliability because race is a basic demographic indicator, involves large population groups, and therefore should be expected to be one of the more reliable measures contained in the ACS.  Furthermore, there has been a flurry of activity over the past week regarding the debate over whether segregation is increasing or decreasing in the Tri-County Southeast Michigan area. Unfortunately, based on my review of the data, we will have to wait for the 2010 US Census data to make a determination, because there are too many unreliable estimates at the census tract level to make this determination using ACS data alone. Please download the presentation for further explanation.