Andrew-

One approach is to use GIS software which can give you estimates of the
percentage of a county in district x,y, and z. It does this by area and
assumptions about the distribution of the population, but is superior to
what you describe below as some counties have only a small piece in a CD.

This is one easy way to get conversion factors
http://mcdc2.missouri.edu/websas/geocorr2k.html

Which I and others have used to estimate district level data from county
sources (such as the County Business Patterns series).

best,
Dustin

Dustin Tingley
Politics Department
Princeton University
http://www.princeton.edu/~dtingley/


On Wed, Sep 30, 2009 at 10:24 AM, Andrew Jones <[log in to unmask]> wrote:

> Hi.  Has anyone come up with a technique that translates FIPS (county)
> level data to congressional districts?  My crude approximation was to evenly
> split counties that covered more than one congressional district (CD).  For
> example, where a county was in two CDs, I would split the county effect
> between the two, assigning each of the CDs 50% of the county level variable
> I was measuring.
> Of course counties are not evenly split between CDs.  Has anyone worked out
> how counties are actually split between CDs for the 2000 Census, say by
> population or by land area?
>
> Andrew Jones
>
> **********************************************************
>            Political Methodology E-Mail List
>  Editors: Xun Pang        <[log in to unmask]>
>           Jon C. Rogowski <[log in to unmask]>
> **********************************************************
>       Send messages to [log in to unmask]
>  To join the list, cancel your subscription, or modify
>          your subscription settings visit:
>
>         http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>

**********************************************************
             Political Methodology E-Mail List
   Editors: Xun Pang        <[log in to unmask]>
            Jon C. Rogowski <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************