Using urban county clusters in MI

[jimrtex]

The southern changes should further improve both equality and compactness.

I saw that the Grand Rapids region was at 1.991, slightly up from 1.986 and wanted to see what the effect was, before making all the changes.

[jimrtex]

The southern changes should further improve both equality and compactness.

[jimrtex]

This is a slight variant, whch improves compactness, with a tiny loss of equality.  Because the northern district has a surplus, it ends up being shifted through a district, increasing the total shift amount.

[jimrtex]

After one round the number of units is reduced to 1750, with 125 needed to comprise a congressional district.   The standard deviation of the apportionment error is 216, or 3.82% of the quota of 5648.  72,444 or 0.73% of the population are shifted across county boundaries.



Here are the groupings for the second round.   With fewer counties with units mod 5 equal to 4, the groupsing are somewhat simpler.  While many counties in the northern part of the state have less than 5 units, the distribution is fairly even.  The UCC may be forcing more complex groupings, but for example, the block of Tuscola, Genessee, and Saginaw forces the rest of the Michigan Thumb to be combined southward.