Subject: [EL] Attached Release Based on Final Pimary Data
From: Curtis Gans
Date: 10/14/2010, 10:55 AM
To: Election Law

This release includes a url to access all the data charts upon which it is based.

The data charts include one for the denominator I will use for the general election.  It is age-eligible citizens based on the 2000  Census' estimate of  those aged 18 and over  minus the Census' estimate of non-citizens interpolated forward based on the rate of change between 1990-2000 and interpolated forward to November rather than Census to Census or April to April.

Back before 2000, I and everybody else except Walter Dean Burnham used the Census age-eligible figure (VAP) knowing that it was a flawed figure and, at least in my case, footnoting its flaws. Thanks to the 2000 and 2002 writings of Michael McDonald and Samuel Popkin nobody uses VAP as their denominator.  McDonald uses a figure he created called VEP. I don't. I don't for three reasons. 1. VEP does not include all the factors that the make the VAP flawed and thus creates a somewhat distorted figure; 2. There is little or no data for some of these factors beyond relatively recent years; and 3. a denominator which eliminates non-citizens is the only one of use historically at least back to 1860, given that 1870 was the first year that the Census counted and published a non-citizen chart.  In addition, non-citizens is the single biggest source of distortion in the VAP.  This is not to say that there are not problems with my November age-eligible citizen figure or Burnham's Census to Census figure.  Any projection forward from a previous decade will need correction when the new Census comes out (as in next April). In some cases the rate of change interpolations will be close to what actually occurred. But in some cases, such as the District of Columbia whose rate of  change between 1990-2000 was negative and  for which there is evidence that the rate of change in this decade -- 2000-2010 is positive, there will be a need for some substantial revision which will be done next April. But in the meantime, I believe that the Census figures on age and non-citizens is still the place for this student of politics to repair.


Curtis Gans, Director
Center for the Study of the American Electorate
Center for Democracy and Election Management
American University
3201 New Mexico Avenue NW
Suite 395
Washington, DC 20016-8026
Phones: (202) 885-6295 (o); (703) 304-1283 (c), (540) 822-5292 (h)
Fax: (202) 885-6294
e-mail: gans@american.edu; curtis.gans@gmail.com
Website:
http://www.American.edu/ia/cdem/csae