Appropos of this entire discussion: it sounds to me like these problems, if they are problems, are the consequence of the  2000-09 effort to realign TIGER. That was the major MAF-TIGER upgrade effort of the decade. Presumably the new lines are (a) more accurate, and (b) won't change again.

Patty Becker



At 10:29 AM 12/7/2011, you wrote:
Content-Language: en-US
Content-Type: multipart/alternative;
         boundary="_000_C803D69C9CCBC74D8A3625CC9D78D809BBC49AE681EXCHMB1mcloca_"

We’re most interested in TAZ zone-level, or place-level stats.  (I can’t imagine needing Block-level change statistics.)
 
So, we recast all the year 2000 Block polygons as centroid points and use “spatial join” to attach a consistent-over-time TAZ code and place code (most recent jurisdictional boundaries).  Then simply roll up Blocks into TAZ totals, place totals, etc. 
 
There are Blocks that straddle TAZ and place boundaries – just have to live with that.
 
(I do repeat the step above 2 or 3 times as a quality control: I identify Block centroids using different centroid location options (thus moving the points around) and test to see whether that yields multiple membership in multiple TAZs or places.  Then I go to aerial photography and manually recode TAZ and place membership based on “rooftops assessment”: which side of the boundary line has the majority of structures, or else the majority of land acreage.  This is an issue only in ½ %  of Blocks.)
 
 
 
 
Todd Graham  |  Principal Forecaster
Metropolitan Council  | 390 North Robert Street  |  Saint Paul, MN  55101
tel: 1+651-602-1322  |  fax: 1+651-602-1674  |  e: todd.graham@metc.state.mn.us 
in:

-------------------------------------------------------------------------------------------------
Patricia C. (Patty) Becker         248/354-6520
APB Associates/SEMCC       FAX 248/354-6645
28300 Franklin Road                   Home 248/355-2428
Southfield, MI  48034                     pbecker@umich.edu