[Openspace] dissolve and contiguity

Julia Koschinsky koschins at uiuc.edu
Thu Sep 9 17:07:26 CDT 2004


Kevin,

After zooming in to your shapefile, you can see that several 
polygons are surrounded by polygon "slivers" or gaps between 
polygons after the dissolve process (we have seen the same 
problem with other dissolved shapefiles).

To get a shapefile that can be used to create a working 
weights matrix, use the CLEAN command in ArcInfo before you 
dissolve. Check your shapefile after the clean operation to 
make sure it worked correctly (sometimes there are problems 
with the operation; note that you can adjust your tolerance 
level). If the clean command worked, then perform the 
dissolve function, which should leave you without any 
slivers. At that point, you can create a correct weights 
matrix in GeoDa (we tested this successfully in practice).

Best regards,
Julia

---- Original message ----
>Date: Wed, 8 Sep 2004 16:05:04 -0700
>From: "Kevin Drakulich" <kmd3 at u.washington.edu>  
>Subject: [Openspace] dissolve and contiguity  
>To: <openspace at agec221.agecon.uiuc.edu>
>
>I know a similar question has been posted to this list 
before by Mike
>Sobczak.  I believe there was no answer to that, but I'm 
hoping someone may
>have one now.
>
>Like Mike, I began with a Tiger shapefile, and 
used "geoprocessing>dissolve"
>in ArcMap 8.2 to aggregate smaller polygons into bigger 
ones.  For me, the
>move was from census tracts to "neighborhoods" (collections 
of census
>tracts) within a city.
>
>Also like Mike, when I created a contiguity-based weight 
file in GeoDA,
>there are lots of mistakes in the file.  For some 
neighborhoods there are no
>neighbors (which is clearly not the case), and for others, 
the neighbors
>listed are clearly wrong (if you look into the .GAL file).
>
>None of these problems exist if you bring in the original 
TIGER files of
>census tracts pre-dissolved.
>
>Has anyone else seen this problem or know what might be 
causing it?
>
>Thanks much,
>Kevin
>
>Kevin M. Drakulich
>University of Washington, Seattle
>
> 
>
>
>_______________________________________________
>Openspace mailing list
>Openspace at sal.agecon.uiuc.edu
>http://sal.agecon.uiuc.edu/mailman/listinfo/openspace


More information about the Openspace mailing list