Using Urban County Clusters To Guide Redistricting

[jimrtex]

Moderator's note: The discussion about LA and VRA requirements have been split off into its own thread (12/8/13).


The following maps illustrate how urban county clusters could be used in the redistricting process.

An Urban County Cluster is a group of counties from a Metropolitan Statistical Area which have 25% or more of their population in Urbanized Areas (urbanized areas are urban areas with a population of 50,000 or more).  So urban county clusters are characterized by significant levels of population concentration and significant commuting ties, which form the basis for defining metropolitan statistical areas.

There are three urban county clusters in Michigan:

Detroit: Wayne, Oakland, Macomb, Livingston, and St. Clair.   This is the entire Detroit MSA, with the exception of Lapeer, which has little population in urbanized areas, and is the most peripheral county.  Genesee, Washtenaw, and Monroe are not in the Detroit MSA, as Flint, Ann Arbor, and Monroe plus Toledo, provide an alternative focus for employment so a smaller share of their population commutes towards Detroit.

Grand Rapids: Kent and Ottawa counties.   Barry and Montcalm are also in the MSA, but do not contain portions of urbanized areas.

Lansing: Ingham, Eaton, and Clinton.  This is the entire Lansing MSA.

There are other single-county clusters, including Kalamazoo (Kalamazoo), Battle Creek (Calhoun), Muskegon (Muskegon), Ann Arbor (Washtenaw), Jackson (Jackson), Monroe (Monroe), Flint (Genesee), Saginaw (Saginaw), Bay City (Bay), Midland (Midland), and Niles-Benton Harbor (Berrien).  For redistricting purposes, these are treated no differently than other counties.



An Apportionment Region is a group of contiguous counties that has a population approximately equivalent to an integer number of congressional districts (for Michigan in 2010, 705,974 persons).  Ideally an apportionment region would have a population equivalent to one congressional district within an error of 0.5% (one-half of one percent), or  (702,444 to 709,504).  Such a region may directly form a congressional district comprised of whole counties, within acceptable levels of population deviation.

Apportionment regions with somewhat larger errors may be converted to congressional districts with relatively small shifts of a portion of one county.  Apportionment regions of higher magnitude will require significant divisions of counties.  Of course, some such division is unavoidable in Michigan, since Wayne, Oakland, and Macomb all have populations greater than that of one congressional district.

In the First Round, the goal is to create the most apportionment regions, which will in turn tend to create more regions that can form, with possible minor adjustments, a single district.

To qualify, an apportionment region must have an error of less than 5% of the ideal congressional district population.   That is an apportionment region with a population approximately equivalent to four districts, must have a population in the range 3.95 to 4.05, rather than 3.80 to 4.20.

An urban county cluster must be contained within a single apportionment region, unless the entirety of one or more apportionment regions may be created within the urban county cluster.  In such a case, the remaining counties, if any, of the urban county cluster must be contained in a single apportionment region.

More than one urban county cluster may be contained within an apportionment region.  However, because of the population of the urban county clusters, such an apportionment region may be rather populous.

While an apportionment region may have a deviation as much as 5%, closer is better, and plans that have unnecessarily large deviations may be excluded from consideration.

The length of the boundaries between apportionment regions is also a consideration.  Highly convoluted, or elongated regions are less compact.  The length of a boundary between two counties is considered to be the direct distance between the end points of that border.  



Portions of the counties within the Great Lakes and Lake St.Clair are not considered as forming boundaries or adjacency.  The only way (for a district) to cross from the Upper Peninsula to the Lower Peninsula is over the Mackinac Bridge.  For purposes of contiguity, both Emmet and Cheboygan counties are considered to be adjacent with Mackinac County, with a zero length border.

Point and near-point connections may not be used to establish contiguity.  A near-point connection is one where the length of the border is less than 10% of the square root of the area of the smaller county, excluding any territory within the Great Lakes.  In particular Gratiot and Shiawassee, and Livingston and Jackson may not be placed in the same apportionment region, unless other counties provide connectivity.  If the border between these counties coincides with boundary between apportionment regions, its length is included in the total length (in both instances, the connection is less than one mile).

[jimrtex]

The following maps illustrate how urban county clusters could be used in the redistricting process.

An Urban County Cluster is a group of counties from a Metropolitan Statistical Area which have 25% or more of their population in Urbanized Areas (urbanized areas are urban areas with a population of 50,000 or more).  So urban county clusters are characterized by significant levels of population concentration and significant commuting ties, which form the basis for defining metropolitan statistical areas.

There are three urban county clusters in Michigan:

Detroit: Wayne, Oakland, Macomb, Livingston, and St. Clair.   This is the entire Detroit MSA, with the exception of Lapeer, which has little population in urbanized areas, and is the most peripheral county.  Genesee, Washtenaw, and Monroe are not in the Detroit MSA, as Flint, Ann Arbor, and Monroe plus Toledo, provide an alternative focus for employment so a smaller share of their population commutes towards Detroit.

Grand Rapids: Kent and Ottawa counties.   Barry and Montcalm are also in the MSA, but do not contain portions of urbanized areas.

Lansing: Ingham, Eaton, and Clinton.  This is the entire Lansing MSA.

There are other single-county clusters, including Kalamazoo (Kalamazoo), Battle Creek (Calhoun), Muskegon (Muskegon), Ann Arbor (Washtenaw), Jackson (Jackson), Monroe (Monroe), Flint (Genesee), Saginaw (Saginaw), Bay City (Bay), Midland (Midland), and Niles-Benton Harbor (Berrien).  For redistricting purposes, these are treated no differently than other counties.

I think this makes a lot of sense as a definition. I like the restriction to counties within the MSA which avoids the whole merger question. For clarification, is the 25% rule based on some usage within the Census or is it arbitrary? I'd like to make the definition as defensible as possible. I know that the states vary on their definition of the threshold for urban vs rural. For example in PA the state average population density is used as the threshold for urban classification.

It appears that for the purposes of applying the 25% rule, the relationship of other urbanized areas within the county to the principal urbanized area is not a factor. Is there a table that clearly delineates the urbanized area population as opposed to the urban area population in the county? Again, I'd like to clearly source the definition, preferably from the Census web site.

The 25% threshold is somewhat arbitrary, but going down to 15% or up to 50% doesn't make a large difference.   There is a spreadsheet on the census bureau website that has the population of each urban area within each county, as well as the non-urban population of each county.  I used that file in combination with the census bureau definition of CBSA to make my calculation:

Using urban area population would have captured a great number of micropolitan areas, as well as fringe areas of metropolitan areas.   Atlanta MSA is up to 29 counties, and while the Atlanta  urbanized area does not reach that far, there are likely to be urban clusters in many of those counties.

There is a time lag problem.  Urban areas are not delineated until a few years after the census.  It might be possible to update populations of the 2000 urban areas, since there are equivalency files relating 2000 census blocks to 2010 census blocks.   It is the intent to update CBSA delineations based on commuting data from the 2011-2015 ACS, but the initial central counties will be those from the 2010 census.

These are the changes if the threshold was increased to 50% or reduced to 15%.

Alabama:
Daphne-Fairhope-Foley (x Baldwin County) 35%
Montgomery (trim Elmore) 34%

Arizona:
Lake Havasu City (x Mohave) 27%
Prescott (x Yavapai) 40%
Sierra Vista (x Coshise) 40%

Arkansas:
Little Rock (trim Lonoke) 45%
Fort Smith (trim Crawford) 48%

California:
Sacramento (trim El Dorado) 34%
Chico (x Butte) 45%

Connecticut:
Worcester, MA (trim Windham) 28%
Torrington (x Litchfield)

Florida:
Crestview-Fort Walton Beach-Destin, FL (add Walton 18%)

Georgia:
Atlanta (add Barrow) 17%
Atlanta (add Carroll) 19%
Atlanta (add Dawson) 20%
Atlanta (trim Walton) 33%
Chattanooga, TN (trim Walker) 45%
Chattanooga, TN (trim Murray) 30%
Macon (add Jones) 16%
Warner Robins (add Peach) 23%
Savannah (add Long) 19%
Savannah (trim Bryan) 31%

Hawaii:
Kahului-Wailuku-Lahaina (x Maui) 36%

Chicago (add Grundy) 24.6%
St.Louis, MO (trim Monroe) 29%
Carbondale (trim Jackson) 47%

Indiana:
Cincinnati, OH (add Dearborn) 20%
Indianapolis (trim Boone) 38%
Indianapolis (trim Morgan) 32%
Terre Haute (trim Clay) 39%

Iowa:
Des Moines (add Warren) 19%

Kansas:
St.Joseph, MO (trim Doniphan) 30%
Wichita (add Butler) 17%

Kentucky:
Clarksville, TN (trim Christian) 28%

Louisiana:
Baton Rouge (trim Iberville) 34%

Maine:
Portland (trim York) 34%
Bangor (x Penobscot) 40%

Maryland:
Balltimore (trim Queen Anne's) 26%
California-Lexington Park (x St.Mary's) 38%
Washington, DC (Calvert) 21%

Massachusetts:
Pittsfield (x Berkshire) 45%

Michigan:
Lansing (x Clinton) 36%
Lansing (x Eaton) 45%
South Bend, IN (add Cass) 17%

Minnesota:
Mankato (x Nicollet) 41%
Minneapolis-St.Paul (trim Sherburne) 29%
Minneapolis-St.Paul (trim Wright) 27%
Duluth (x St.Louis) 47%
Grand Forks, ND (trim Polk) 26%
La Crosse, WI (trim Houston) 28%

Mississippi:
Gulfport-Biloxi-Pascagoula (trim Hancock) 40%
Hattiesburg (trim Lamar) 49.6%

New Jersey:
Allentown, PA (trim Warren) 48%
New York-Newark, NY-NJ (trim Hunterdon) 46%
New York-Newark, NY-NJ (trim Sussex) 25%
Philadelphia, PA (trim Sussex) 40%

New Mexico:
Farmington (x San Juan) 41%

New York:
Kingston (x Ulster) 49%
Rochester (trim Ontario) 32%
Utica-Rome (x Oneida) 49%
Watertown (x Jamestown) 49.8%

North Carolina:
Asheville (x Haywood) 45%
Greensboro (add Randolph) 15%
Winston-Salem (add Davie) 17%
Winston-Salem (add Stokes) 24%
Myrtle Beach, SC (trim Brunswick) 37%
New Bern (x Craven) 49%
Raleigh (add Johnston) 22%
Rocky Mount (trim Edgecombe) 31%
Goldsboro (x Wayne) 49.8%

Ohio:
Cleveland (add Geauga) 22%
Columbus (trim Fairfield) 34%
Dayton (trim Miami) 46%
Toledo (trim Wood) 42%
Wheeling, WV (trim Belmont) 39%

Oklahoma:
Oklahoma City (trim Logan) 25%
Tulsa (trim Creek) 36%
Tulsa (add Rogers) 21%

Oregon:
Albany (x Linn) 40%
Salem (x Polk) 35%

Pennsylvania:
Bloomsburg (trim Montour) 46%
Gettysburg (x Adams) 31%
Johnstown (x Cambria) 43%
Allentown (trim Carbon) 27%
East Stroudsburg (x Monroe) 33%
Pittsburgh (trim Butler) 32%
Pittsburgh (trim Fayette) 47%
Chambersburg (x Franklin) 40%
Williamsport (x Lycoming) 48%
Youngstown, OH (trim Mercer) 33%

South Carolina:
Charlotte, NC (add Lancaster) 19%
Columbia (add Kershaw) 20%
Florence (add Darlington) 20%
Hilton Head Island-Bluffton-Beaufort, SC (x Beaufort) 43%

South Dakota:
Rapid City (add Meade) 24.9%
Sioux City, IA (trim Union) 39%
Sioux Falls (trim Lincoln) 49%

Tennessee:
Kingsport (trim Hawkins) 31%
Morristowm (add Jefferson) 20%
Sevierville (new Sevier) 17%
Nashville (add Robertson 20%)
Nashville (trim Wilson 37%)

Texas:
Corpus Christi (trim San Patricio) 26%
Dallas-Fort Worth (trim Johnson) 29%
Houston (trim Chambers ) 36%
Beaumont (trim Hardin) 39%
San Antonio (trim Comal) 49%

Utah:
Ogden (trim Box Elder) 49%

Virginia:
Staunton-Waynesboro (x Augusta + IC) 48%
Kingsport-Bristol-Bristol, TN-VA (add Scott) 18%
Kingsport-Bristol-Bristol, TN-VA (trim Washington + Bristol) 46%
Virginia Beach-Norfolk (add Gloucester) 24%
Washington, DC (trim Fauqier) 32%
Lynchburg (trim Amherst) 36%
Lynchburg (add Bedford) 18%
Lynchburg (trim Campbell) 32%
Richmond (trim Dinwiddie) 29%
Washington, DC (trim Prince George) 47%
Roanoke (trim Botetourt County) 36%

West Virginia:
Huntington (trim Wayne) 33%
Beckley (trim Fayette) 35%

Wisconsin:
Eau Claire (trim Chippewa) 43%

I don't see a compelling reason to drop down to 15%.  If the threshold were increased, I would change the definition to: county whose largest urban area is an urbanized area, and which has an urban population greater than 50%.  Many of the areas that would be disqualified based on a higher threshold, are because they have larger urban clusters that have not connected to the main urbanized area.

Ellis County, TX is at 56% because there is no Waxahachie UC, while Johnson County, TX is at 29% because Cleburne UC still exists.

[jimrtex]

EDIT: Corrections to maps.

Mixed up counties in Appleton and Oshkosh WI MSAs
Mixed up Lousville suburbs in Indiana (Floyd and Clark)
Showed Ottawa rather than Wood as part of Toledo, OH cluster.
Included Chatham, NC as part of Durham-Chapel Hill cluster.
Showed Goliad rather than Victoria, TX.
Omitted York, VA from Virginia Beach-Norfolk cluster.

No changes to SW and NW maps.

Yes, I wonder how well this definition will work in other states, and where the data is.  And what and where do you look to know if more than 25% of the county is in "urbanized areas?"

The lighter shaded areas (eg Washtenaw, Genesee, and Monroe) are separate Metripolitan Statistical Areas within a Consolidated Statistical Area (CSA).  The darker area is the "main" MSA for the CSA.  The original version used CSA, but these were too extensive (Allentown MSA is part of the New York CSA),

With the latest definition, the association between Ann Arbor, Flint, and Monroe and Detroit is simply to illustrate the evolution of definition.  The single county "clusters" are treated as ordinary counties.









Link to 2010 Urban Area to County Relationship File

Link to Core based statistical areas (CBSAs) and combined statistical areas (CSAs)

Some definitions:

Urban Area: Blob of continuous dense residential population, which are delineated (for the most part) without regard to political boundaries and have at least 2500 persons.  The Urban population lives in urban areas, the Rural population lives outside urban areas.

Urban Areas are classified as Urbanized Areas and Urban Clusters based on their population.  Urbanized Areas have a population of 50,000 or more.  Urbanized Areas have been defined since the mid-1900's to recognize suburban growth outside incorporated cities.  The concept of Urban Areas based on population density is new with the 2000 Census, though intended to provide continuity with the 1990 delineations.  Before 2000, the Rural population lived outside cities of 2500 persons.

Core Based Statistical Areas (CBSA) are groups of counties defined based on a core urban area and commuting patterns.  Metropolitan Statistical Areas are associated with Urbanized Areas, while Micropolitan Statistical Areas are associated with Urban Clusters.

Urban Areas with a population greater than 10,000 (that is, all Urbanized Areas and Urban Clusters with more than 10,000 persons) are potential cores of a CBSA.  A county is an initial central county if the urban area with the most population in the county, has greater than 10,000 persons total AND 5,000 persons or more, or 50% or more of the total county population.  Counties that share the same potential core, form an initial central county cluster.

Initial central county clusters are treated as a unit for determining outlying counties.  Outlying counties may be either initial central counties, or counties without a core area.  In the Detroit area, Wayne, Oakland, and Macomb form a central county cluster based on the Detroit Urbanized Area.  St. Clair, Lapeer, Genesee, Livingston, Washtenaw, and Monroe form separate initial county clusters (of one) based on Port Huron UA, Lapeer UC, Flint UA, South Lyon-Howell UA, Ann Arbor UA, and Monroe UA, respectively (Detroit UA extends into most of the counties but is not the dominant urban area).

St.Clair, Lapeer, and Livingston are converted to outlying counties of the Detroit-Warren-Dearborn Metropolitan Area based on commuting into the central counties of Oakland, Macomb, and Wayne (25% or more of workers resident in the outlying counties work in the central counties).   Genesee, Washtenaw, and Monroe remain independent MSA.

What my definition is intended to capture is counties with dense population, and strong commuting ties, but that happen to have a different urban core.

[jimrtex]

Yes, I wonder how well this definition will work in other states, and where the data is.  And what and where do you look to know if more than 25% of the county is in "urbanized areas?"

The lighter shaded areas (eg Washtenaw, Genesee, and Monroe) are separate Metripolitan Statistical Areas within a Consolidated Statistical Area (CSA).  The darker area is the "main" MSA for the CSA.  The original version used CSA, but these were too extensive (Allentown MSA is part of the New York CSA),

With the latest definition, the association between Ann Arbor, Flint, and Monroe and Detroit is simply to illustrate the evolution of definition.  The single county "clusters" are treated as ordinary counties.

I think I understand, and am still looking to put a firmer basis on the 25% rule. I assume this is the excel file you are referring to on this Census page.

Comma delimited text file on this Census Page which also has the explanation of the file, as well as a similar file for relationships between urban areas and minor county divisions.   The latter file may be of use in New England, and possibly throughout the northeastern US.  For example, it might provide a better way to split Ottawa County between Muskegon and Grand Rapids.

Re 25%.

Typically, the percentage is up around 80% to 100%, but perhaps drops down to the 70% for smaller urbanized areas.  A lot depends on the size of the county.  El Dorado, CA fell be 50% because it has a substantial population around Lake Tahoe in addition to Sacramento.  The same is true for St.Louis County, MN where the Iron Range population is significant relative to that of Duluth.

The actual footprint of the urbanized area might be relatively small.  Lansing covers parts of 13 townships, in an area of the state where counties are 16 townships (4 x 4), 5 in Ingham, 3 in Eaton, 4 in Clinton, and 1 in Shiawassee.  In cases like that, the percentage of the county population might be somewhat low, because there is substantial area for smaller towns and urban fringe - areas where the residents commute but otherwise live in a rural setting.   20 ppm is probably a high number for purely agricultural area (4 largish families with 160 acres each).   But you could easily get to 200 ppm without any real crowding or need for streets.

The Lansing UA portion of Clinton County has 36% of the population in the county, in 5% of the area.  The density of the Lansing UA is 1018 ppm; of St.Johns UC is 1992 ppm.  And for the rural area it is 74 ppm.  It is denser than it would be without Lansing.

The Census Bureau definition of central county is not particularly helpful, since it appears to be intended for use with urban clusters and micropolitan areas.  It only requires 5,000 persons from the UC to be in the county.  But that is between 10% and 50% of the UC population.  But 5,000 persons is at most 10% of an urbanized area.

In the meantime, I see Lucas and Ottawa, OH in the same cluster. However, they are not in the same MSA, but Wood is 41.97% UA in the above table so it looks like it should be with Lucas instead of Ottawa.

It should be Wood.  (note Wood fails because of the presence of Bowling Green)

I am preparing an alternate definition.   County in MSA, urbanized area has largest urban area share of county population, and urban population > 50% of county population.  That should be a more limited trimming than my list.

[jimrtex]

These changes are based on the following definition:

County is (1) constituent of a metropolitan statistical area; (2) an urbanized area is the most populous urban area in the county; (3) and 50% or more of the county population is in urban areas (alternatively, less than half the county's population is rural.

Thus qualifying counties are either (a) a central county of the MSA (1 and 2 are necessary conditions) and have a mostly urban population; or (b) an outlying county, that had been initially identified as a central county of a metropolitan statistical area, but that had been merged into another CBSA based on commuting, and has a mostly urban population.

Counties of type (b) had either initially been identified by a pseudo core (eg South Lyon-Howell UA) or a weak core (eg Port Huron UA). 

Counties that are excluded by the 50% urban requirement were misidentified as urbanized counties based on the Census Bureau using a single criteria for both Metropolitan and Micropolitan statistical areas (ie 5,000 persons in the county in the largest urbanized area).   5,000 persons is between 10% and 50% of an urban cluster that is a core.  It is less than 10% of an urbanized area that is a core, and thus some counties that are fairly peripheral to the urbanized area, are classified as being "central" to the metropolitan area.  These counties are really bedroom counties, but have some people sleeping on the couch in the living room.

Alabama:
Montgomery (trim Elmore) 46% urban.

Arkansas:
Fort Smith (trim Crawford) 48% urban.

Connecticut:
Torrington (eliminate Litchfield).  Bridgeport UA, Danbury UA, Hartford UA, and Waterbury UA come into Litchfield County, but Torrington UC is the most populous urban area.  In part because of multiple target counties for commuting, Litchfield is not an outlying county of any.

Delaware:
Salisbury, MD (add Sussex)  Salisbury, MD urbanized area is the largest urban area, in a county with a bunch of small urban clusters, such that Sussex County is 59% urban.

Georgia:
Athens (trim Oconee) 49.6% urban.
Dalton (trim Murray)  30% urban.   I had misstated that Dalton was in Chattanooga, TN MSA.

Indiana:
Terre Haute (trim Clay) 39% urban.

Kansas:
St.Joseph, MO (trim Doniphan) 30% urban.  The portion of St. Joseph UA in Doniphan is so small (2368) it does not even qualify as a central county under the Census Bureau definition.

Kentucky:
Clarksville, TN (trim Christian) Hopkinsville UC is largest UA in county.

Louisiana:
Baton Rouge (trim Iberville) 41% urban.

Maine:
Portland (trim York) 43% urban. 
Bangor (eliminate Penobscot) 42% urban.

Maryland:
Balltimore (trim Queen Anne's) 46% urban.
California-Lexington Park (eliminate St.Mary's) 49.6% urban.

Michigan:
Lansing (trim Clinton) 47% urban.

Minnesota:
La Crosse, WI (trim Houston) 43% urban.

Mississippi:
Hattiesburg (trim Lamar) 49.6% urban.

North Carolina:
Asheville (trim Haywood) 45% urban.

Ohio:
Wheeling, WV (trim Belmont) 45% urban.

Oklahoma:
Oklahoma City (trim Logan) 45% urban.
Tulsa (trim Creek)  46% urban.

Pennsylvania:
Bloomsburg (trim Montour) 46% urban.
Gettysburg (eliminate Adams) 46% urban Hanover Urbanized Area qualifies the county as a central county of a metropolitan statistical area, but Hanover is in York County with a long flanking arm down the York Pike into Gettysburg.   York (City) Urbanized Area is the much larger urban area that qualifies York County.  York-Hanover MSA consists of York County, while Gettysburg MSA consists of Adams County (MSA are named based on the largest city, regardless of the urban area that forms the core).

South Dakota:
Sioux City, IA (trim Union) 39% urban.

Tennessee:
Kingsport (trim Hawkins) 42% urban.

Texas:
Beaumont (trim Hardin) 48% urban.

Virginia:
I had to restore 3 independent cities to their original county to maintain the counties as part of the urban county cluster.  I suspect that best solution would be to treat all IC as part of their original counties.

Kingsport-Bristol-Bristol, TN-VA (trim Washington + Bristol) 46% urban.
Washington, DC (trim Fauqier) 43% urban.
Lynchburg (trim Amherst) 36% urban.
Lynchburg IC added to Campbell to maintain its status.
Petersburg IC added to Dinwiddie to maintain its status as part of Richmond cluster.
Hopewell IC added to Prince George to maintain its status as part of Richmond cluster.
Roanoke (trim Botetourt County) 36% urban.

West Virginia:
Huntington (trim Wayne) 35% urban.
Beckley (trim Fayette) 42% urban.

[jimrtex]

So does this definition put Wood, OH with Toledo? The percentages are 41.97% UA, 28.48% UC, leaving 29.55% rural. You said it would fail before because of Bowling Green, but this new definition appears to simplify things so that one doesn't need that level of detail.

I started looking at some of the OH competition plans, since minimizing county fragments and increasing compactness were both goals. There was a 0.5% population deviation maximum for the districts. Instead of using our concept of microchops which don't count, there was a rule that allowed a chop without penalty that exactly included the part of municipality that crossed the county line. It is possible to view the top scoring plans in terms of apportionment regions, and the 1st and 2nd place plans are shown below. The plan on the left was better for compactness and the plan on the right had fewer fragments seen as more regions. It is interesting that neither plan preserves the metro areas we are identifying.

 

With my alternate definition the urban county clusters would be:

Youngstown: Mahoning, Trumbull
Akron: Summit, Portage
Cleveland: Cuyahoga, Lorain, Lake, Medina
Columbus: Franklin, Delaware, Licking, Fairfield
Dayton: Montgomery, Greene, Miami
Cincinnati: Hamilton, Butler, Warren, Clermont
Toledo: Lucas, Wood

Single county clusters:
Lima: Allen
Springfield: Clark
Mansfield: Richland
Weirton-Steubenville WV-OH: Jefferson
Huntington-Ashland, WV-KY-OH: Lawrence

What I was trying to capture with the 25% threshold was where urbanized areas were a significant part of a county's population, and implied additional urban population or urban fringe population (not classified as urban, but much higher density than economically-viable agriculture will support).

The latest definition requires that an urbanized area be the largest urban area in the county.  That is, it would be the urban area that we would choose if we had to select a single  urban area that best characterized the county's settlement pattern (and is quite similar to the procedure used by the Census Bureau when defining the initial set of central counties).

For the most part, the 25% total urbanized area threshold assured we had picked the largest urban area.  Exceptions were Sussex, DE, where Salisbury UA had 21% of the population, along with a bunch of urban clusters, and Litchfield, CT, where several urbanized areas produced a relatively high percentage, but not in any single one, such that Torrington UC was the largest core.

But adding the 50% urban criteria actually demonstrates a concentration of spatially-associated population, rather than the previous implied settlement.

The counties that are trimmed under my latest standard are typically low population (less than 100,000) and where an urbanized area has crossed the boundary, but not had a strong enough presence to produce related population growth.   A county  with less than 50% urban population simply is not a concentrated high density population.  Though it might qualify as a "central county" it functions as an "outlying county".

What I was say with regard to Wood, was that Bowling Green UC with the 24% of the population, was keeping Toledo UA at 42% from reaching a higher share of the population.

The district along Lake Erie in the contest plans could be considered a Bowling Green-Port Huron to Elyria-Lorain district, but using the urban county definition it is a Toledo to Cleveland district.

I didn't really like the Ohio rule for cities crossing county lines.   It was too much a gimmick, and basically said it was OK to split the city, but you could also split the county.   If you wanted to consider splitting cities and townships something to have priority over splitting counties (which is certainly the priority expressed with regard to House districts in the constitution), I would rationalize the political geography.

A city that crossed a county line would be considered to be in the county in which it had the most population (eg Franklin County would be extended outward to encompass all of Columbus, for districting purposes).   The same would be done for municipalities in multiple townships, and a municipality that had absorbed a township in part, would absorb all townships throughout its territory.

[jimrtex]

I believe my final Ohio plan did preserve all the urban clusters no? I am a bit confused about what you guys are talking about, about Wood. I get confused a lot on these matters.

I think this is the OH plan you refer to.

There are quite a few chops and splits of counties in the clusters that I see. In order to test urban cluster preservation the plan has to be reduced to a set of whole county apportionment regions. County clusters may not span different regions in a state.

Ideally there should be no population shifts between regions while providing districts within 0.5% of the quota. In practice up to one microchop may be used between any two regions to get districts within that limit. If a regular chop is needed, then those two regions are a single larger apportionment region. Chops used within a region should be kept to a minimum.

Maximizing the number of valid apportionment regions is equivalent to minimizing the number of regular chops. That chop count has to be balanced against erosity and the court will expect that for a given set of criteria (ie chop count and erosity) the population range between the most and least populous district will be minimized.

I think his Columbus cluster is OK.  The counties added to get to three districts don't have to be together.  The idea is to keep the 3 districts Columbus-centric.

It should pretty easy to get Lucas+Wood into a 1-district region.

Cincinnati+Dayton can probably go into a 4-district region.   4 plus 1, is just as good as 3 plus 2, and arguably better since the one is a non-metro district rather than putting a larger area under the domination of the large cities.

In northeast Ohio, Summit+Portage might have to be used.

[jimrtex]

I believe my final Ohio plan did preserve all the urban clusters no? I am a bit confused about what you guys are talking about, about Wood. I get confused a lot on these matters.

Wood County is 42% Toledo Urbanized Area, 24% Bowling Green Urban Cluster, and the rest rural or small Urban Clusters.

Under my first definition it qualified because it was part of the Toledo MSA, and more than 25% of the county was in urbanized areas (urbanized areas have a population greater than 50,000, urban clusters have less than 50,000.  Size is the only distinction between urbanized areas and urban clusters.

Muon had questioned where the 25% threshold came from.   Answer: it seemed to be a reasonable number to me.   Typically, if you have a large urban area, it has a much higher share of the population, since at 1000 person per square mile, a small area of a county, say 10% can have a lot of people.   So most counties with part of an urbanized area have most of their population in that urbanized area, or very little.  25% was in sort of a middle area, not including too many counties which really weren't densely populated.  Or in actuality populated densely.  For example, San Berndardino County is not densely populated, but the areas that are populated, are populated densely).  25% also did not exclude many counties where the urbanized area was really characterizing the population distribution.

I had also reasoned that if an urbanized area had 25% of a county's population, that there would be additional population nearby, which while not in the urbanized area, but was closely tied economically to a city where they did everything but sleep and mow the lawn.  They not only worked in the city, but shopped in the city.

But I had examined what would happen if I had increased the threshold to 50% or reduced it to 15%.  Wood County would have been excluded from the Toledo cluster if the threshold had been increased to 50%,

But under my new proposal it would qualify.  The largest urban area in the county is Toledo Urbanized area.  If you were characterizing where the population lives, "Toledo suburbs" is an accurate answer.  If you said "a mix of Toledo suburbs, Bowling Green, and rural", you would also be accurate, possibly more accurate.  But we are only permitting one answer.

Which in this case is "Toledo suburbs and most of the population is urban."

If Ohio were to create metropolitan governments, and did it in a rational fashion, it would likely split Wood County. 

We could do the same with respect to redistricting, but it is unlikely to be done in a way than wouldn't be interpreted as meaning, "so us pols can put Kaptur and Kucinich in the same district after all."

[jimrtex]

I believe my final Ohio plan did preserve all the urban clusters no? I am a bit confused about what you guys are talking about, about Wood. I get confused a lot on these matters.

I think this is the OH plan you refer to.

There are quite a few chops and splits of counties in the clusters that I see. In order to test urban cluster preservation the plan has to be reduced to a set of whole county apportionment regions. County clusters may not span different regions in a state.

Ideally there should be no population shifts between regions while providing districts within 0.5% of the quota. In practice up to one microchop may be used between any two regions to get districts within that limit. If a regular chop is needed, then those two regions are a single larger apportionment region. Chops used within a region should be kept to a minimum.

Maximizing the number of valid apportionment regions is equivalent to minimizing the number of regular chops. That chop count has to be balanced against erosity and the court will expect that for a given set of criteria (ie chop count and erosity) the population range between the most and least populous district will be minimized.

I think his Columbus cluster is OK.  The counties added to get to three districts don't have to be together.  The idea is to keep the 3 districts Columbus-centric.

It should pretty easy to get Lucas+Wood into a 1-district region.

Cincinnati+Dayton can probably go into a 4-district region.   4 plus 1, is just as good as 3 plus 2, and arguably better since the one is a non-metro district rather than putting a larger area under the domination of the large cities.

In northeast Ohio, Summit+Portage might have to be used.

I think we agree on this, let me confirm that. The multi-county clusters in OH and their number of districts is:
Youngstown: Mahoning, Trumbull (0.623)
Akron: Summit, Portage (0.975)
Cleveland: Cuyahoga, Lorain, Lake, Medina (2.751)
Columbus: Franklin, Delaware, Licking, Fairfield (2.289)
Dayton: Montgomery, Greene, Miami (1.108)
Cincinnati: Hamilton, Butler, Warren, Clermont (2.192)
Toledo: Lucas, Wood (0.787)

In general one anticipates that each county cluster would go into a region made up of a number of districts rounded up to the next whole number. The remaining counties would be grouped into single district regions. Clusters that are proximate to one another can be placed into a single larger region. Regions can also be merged to facilitate chops designed to reduce erosity. When a single region includes two adjacent clusters, districts may split the clusters to reduce chops and erosity.

In OH, there are no small counties that can bring the Akron cluster up to the level of a one district quota, and the population can't be made up by microchops alone. Therefore Akron will have to combine with either the Cleveland or Youngstown cluster, or single counties including Stark. One possibility that occurs with a Cleveland merger is that it may may more sense to split Summit and Portage to maintain other district criteria. For example:



The Cleveland and Akron clusters can combine with Geauga and Ashtabula to form a compact region with 3.997 CDs, so no microchops with any other region are necessary. By combining Summit and Medina only a microchop using Oakwood brings it to within quota, this also provides for a much less erose CD 14 which otherwise would have to wrap around the south edge of Cuyahoga to get enough population within this region. The districts are all well within the Cleveland CSA, so the split of Summit from Portage at the district level works for me.

"Anticipates" might be too strong of a word, since it may cause a map drawer to miss the opportunity, for example, to place the Dayton and Cincinatti clusters into a single region, that only has to encompass 500,000 persons outside the cluster, rather than 1.2 million persons outside the two areas combined.  Rounding up does set some sort of outer limit, but could produce fewer apportion regions than competing plans.

Akron is within 5%.  So I would go ahead and add Geauga and Ashtabula to bring Cleveland up to 3.022.  You have more regions, and the correction is quite localized, since the excess in Cuyahoga complements the deficit in  Moving 15,000 persons from Medina or Geauga is not grossly disrespecting counties given the density of population.

Note that under my rules, it is also possible to create an apportionment region entirely within a cluster.  This might work for Montgomery+Greene, and then Miami would be place in another region.   Since the region will eventually have to be split, it is OK to split it now, if we can at this stage.

[jimrtex]

As I understand the concept of the urban clusters applied to OH, one could start with a regional plan like this one. It's certainly not the only one, but it's the one I used to produce the NE OH picture I posted earlier.



Each of the seven clusters identified by jimrtex is embedded in a single region. The deviations are small enough that only microchops are needed to balance population between regions. Akron is merged with the Cleveland cluster and the Dayton and Cinci clusters are together. If one doesn't like the shape of the Youngstown region it can be combined with the Canton region using a chop in Stark to improve the shape. That's the type of tradeoff that should be permitted, one less region for less erosity.

What I envision happening is that some agency (eg secretary of state, legislative council, state statistical agency) would prepare the data.   They could also include some simple interactive tool for drawing maps.  There would be some data standard so that people can use their own software to submit maps.

In particular, they would produce a file of border lengths and connectivity.

In Ohio, I would not permit a direct connection between Clermont and Clinton, nor Mercer and Shelby (Auglaize and Darke are not contiguous).  I would include the length of these border connections in determining the length of the border.

Stark-Holmes, Ross-Jackson, Montgomery-Clark do meet my suggested standard of border gap greater than 10% of the square root of the area of the smaller county.

I could take your map and separate the Akron cluster as its own region.  Though this would have greater error, it would have more regions.   I suspect a similar split would be possible in the southwest.   And how close is Franklin+Licking or Frankilin+Delaware to 2.00?  In the case of Franklin+Licking, I would also require the other region to include both Delaware and Fairfield.

[jimrtex]

As I understand the concept of the urban clusters applied to OH, one could start with a regional plan like this one. It's certainly not the only one, but it's the one I used to produce the NE OH picture I posted earlier.



Each of the seven clusters identified by jimrtex is embedded in a single region. The deviations are small enough that only microchops are needed to balance population between regions. Akron is merged with the Cleveland cluster and the Dayton and Cinci clusters are together. If one doesn't like the shape of the Youngstown region it can be combined with the Canton region using a chop in Stark to improve the shape. That's the type of tradeoff that should be permitted, one less region for less erosity.

I could take your map and separate the Akron cluster as its own region.  Though this would have greater error, it would have more regions. 

Ideally there should be no population shifts between regions while providing districts within 0.5% of the quota. In practice up to one microchop may be used between any two regions to get districts within that limit. If a regular chop is needed, then those two regions are a single larger apportionment region. Chops used within a region should be kept to a minimum.

Maximizing the number of valid apportionment regions is equivalent to minimizing the number of regular chops. That chop count has to be balanced against erosity and the court will expect that for a given set of criteria (ie chop count and erosity) the population range between the most and least populous district will be minimized.

In OH, there are no small counties that can bring the Akron cluster up to the level of a one district quota, and the population can't be made up by microchops alone. Therefore Akron will have to combine with either the Cleveland or Youngstown cluster, or single counties including Stark. One possibility that occurs with a Cleveland merger is that it may may more sense to split Summit and Portage to maintain other district criteria. For example:



The Cleveland and Akron clusters can combine with Geauga and Ashtabula to form a compact region with 3.997 CDs, so no microchops with any other region are necessary. By combining Summit and Medina only a microchop using Oakwood brings it to within quota, this also provides for a much less erose CD 14 which otherwise would have to wrap around the south edge of Cuyahoga to get enough population within this region. The districts are all well within the Cleveland CSA, so the split of Summit from Portage at the district level works for me.

Akron is within 5%.  So I would go ahead and add Geauga and Ashtabula to bring Cleveland up to 3.022.  You have more regions, and the correction is quite localized, since the excess in Cuyahoga complements the deficit in  Moving 15,000 persons from Medina or Geauga is not grossly disrespecting counties given the density of population.

So are you saying that you would disallow the map above? How would you draw the CDs in Cuyahoga using those regions?

I have quite a difficulty with the 5% rule. In order to be a useful constraint the permitted deviation has to be based on the permitted range or deviation allowed for the type of district in the particular state. In OH a 5% deviation is fine for a legislative map since districts are permitted a 5% deviation (except for certain single county districts). In IL, 5% deviation isn't helpful since the court has ruled that legislative districts must be within 0.5%. Using a 5% standard there wouldn't help since once one shifts population in excess of the deviation, the court will inquire as to why the shift didn't get to equality. Once one requires exact equality it gets much harder to constrain the map against gerrymandering, since almost any shaped chop can be justified in the name of strict population equality.

For congressional districts it's clear to me that had WV used a 5% standard to form three regions in 2011, they would not have survived in court. They did prevail because WV provided its criteria and then since the range was under 1% the court (federal courts seem to prefer range to deviation) put the burden on the plaintiff to show that they could get a smaller range with the same criteria. Like my example above, a 5% deviation for congressional districts will force one to use exact population equality and with that most constraints are weakened as they must take a second place to strict equality. I want constraints that are on a par with population equality as shown in the WV case.

This is the initial step, to establish the number of regions.  It is like pulling a suit off the rack and checking for the basic fit.  Other suits may fit better, and require less customization.

After the number of regions is established, there would be additional rounds of submissions.  Plans that had substantially greater deviation or greater erosity would be excluded.

You should read Tennant again.  West Virginia largely made up its rules on fly, and rationalized them on the basis of past practice.

The 3 stated interests of West Virginia were:

(1) Not pairing incumbents.   This is totally a made up interest, and of dubious legitimacy,
(2) Not splitting counties.  A reasonable criteria, though justified in part based on the state constitution's requirement to not split counties to form senate districts (which is flatly unconstitutional given the number of senate districts.
(3) Not making too many changes.  Possibly reasonable, but not if the districts were flawed in the first place.

You are probably correct about the 5% vs 0.5%, but that is because of sticker shock on the part of the SCOTUS.

Ohio has a specific standard in its constitution, and constrains how house districts are formed.  Illinois has no standard in its constitution, and no rules other than compactness,

There are better more practicable ways of measuring equality.  Use of range permits greater inequality, and grossly disregards the totality of a plan.  It can not be used to determine whether there is “a good-faith effort to achieve absolute equality.”

[jimrtex]

I think this is exactly what I have in mind, and the public should be able to participate by adding to that pile as well as the computer. Where I perceive our greatest disagreement is that I see a huge advantage to single districts that can be constructed from single counties like in IA and WV. The most frequent comment in IL was "why can't we do it like Iowa?" and part of that meant the type of district that was produced from whole counties. Look at the congressional map of IA-01 and its western boundary with IA-04. The public is OK with that, but I fear that it's too erose for you. In states with larger population centers the next best thing to IA is to get a reasonable number of IA-like districts that aren't too erose, then look at fair ways to chop the large urban cores.

Iowa has 99 counties and 4 congressional districts, or roughly 25 per district.  Further the counties have relatively modest populations.  None grossly huge, none very small.  The current map has districts of 39, 24, 20, and 16 counties, for a standard deviation of 8.7 counties.

Now imagine you have an state like Iowa where counties line up in neat rows, and pretty neat columns.   You have a n x 2 area where the boundary between districts should go.   The border roughly goes between two columns, n counties high.   If we permit the boundary to advance one county to the west, remain the same, or retract one county to the east, in each row, how many variations are there?

Answer: 3ⁿ.

If n is 4, that is 81.  If n is 5 it is 243.   With some variation between the counties, we get an almost continuous distribution of population.  And even the more toothier combinations aren't that horrible looking when they are made out of bricks.

If n is 3 or 2, then the number of combinations shrinks.  And if some of the counties are large, the distribution of population becomes decidedly chunkier.

It is this exponential explosion of options when the district are made up of a large number of counties that makes Iowa easy.  Iowa is a degenerate case.  And the problem is trying to extrapolate from Iowa.  It is better to have a general method when applied to Iowa drops out a nice map.

Thought exercise.  Divide Iowa on an east-west line that is approximately equal in population.  Shift the fewest numbers of counties at the east or west end that produces better equality.  So our north-south boundary is a line with at most a single jog.   Then divide the north and south halves on separate north-south boundaries.

Since Iowa is roughly 11 counties wide by 9 counties high, this gives 20 county pairs along the boundaries.  If we adjusted those boundaries one county to the east or west or north or south, we would be able to make 3²⁰ or 3.48 billion variations.  We could get quite nice districts, and eliminate the infamous Minnesota Finger.  There are some constraints near the junction of the districts, so that we can't shift a county both east and south, for example, but the number is still close to one million.

[jimrtex]

I just followed Jimtex's map, and I see now that Union and Madison are not included in the Columbus urban cluster, so I withdraw the comment. I have marked the areas of the map that I would simply not do, and if OH-08 on your map jutted much more to the south, while having that jut to the north, that would be a fail for me, but the jut to the south is not that much, so it's OK. OH-05 does have two juts, although the one to the east is totally understandable, but the one to the west on top of it is a bridge too far - it becomes a two jut CD. If you are going to take the county to the east, you need to take the one below first, rather than go west.  I assume Canton and Akron are not in the same urban cluster, because you  split them. If they are, such a total bifurcation, as opposed to a nip,  bothers me. The worst aspect is OH-04 doing that extra rectangle to the east.




If I get up the energy, I might take your map, and make it less erose, while adding a couple of chops presumably, to see what happens, for purposes of comparison. The issue is not so much intolerably erose, but that a chop or two more is worth it, if it makes the map discernibly less erose. I would not characterize your map as "intolerably erose," but rather "uncomfortably erose."  I guess what I do is go for the least erosity, and then where possible add a bit or erosity to try to eliminate chops. I guess I must plead guilty to putting erosity first.

And I oppose the 5% variation, as opposed to the 0.5% variation. Did SCOTUS really sign off on the 5% variation in some recent case? If not, where did that come from?  Heck if it's 5%, some of my macro-chops probably go away.  The 5% will never sell anyway in most places. It's DOA.

As I noted, the 5% is acceptable only for legislative districts when there is no stricter state standards. jimrtex favors a more relaxed standard for apportionment regions, but that is ineffective at reducing regular chops. For CDs the range can be up to 1% if the state can show consistently applied principles, and there is no smaller range using those same principles. A 0.5% deviation is an easy way to insure a range within 1%.

You're reading Tennant as narrowly as Karcher had been read.

Good faith effort can not be so narrowly defined as:

GOOD_FAITH = 0.005.

While avoiding a mini-chop which would have moved 15,000 persons.   You sliced between Portage and Summit.  You did not make a good faith effort to preserve that urban county cluster.

And range criteria is the wrong one to apply.  If you'd like, I'll show the MALC proposal for Texas House seats that cranked every Republican/Anglo seat up to 4.99%.  If the redistricting application displayed another digit, they would have gone to 4.999%.  Good faith is attempting to hit the bulls eye, and scoring 10 if you are within or even touching the ring.  Bad faith is to carry the red arrows down to the target, and insert them so that they are touching the ring.

[jimrtex]

While avoiding a mini-chop which would have moved 15,000 persons.   You sliced between Portage and Summit.  You did not make a good faith effort to preserve that urban county cluster.

I posed this map of NE OH precisely to understand how well one must preserve a cluster beyond the first step of its inclusion wholly within one region. I noted that there were two factors that should be weighed against preservation of the cluster within the region. One was the ability to have a nearly whole county district. The other factor was the reduction of erosity that was provided by the split. Without that split CD 14 needs a wrap around to Parma and Strongsville which is far less compact than the map I drew. You are suggesting that the weight of those two factors cannot outweigh the chop of the cluster.

That implies that the chop of the cluster within a region is not merely a scoring factor but is a rule in and of itself. Perhaps something of this form:

An urban county cluster within a region cannot be represented by more districts than the nearest whole number of districts in excess of the number that would be apportioned to the cluster alone.

With that rule Summit and Portage would always be wholly together plus about 18K from one of the neighboring counties. Without that rule one has to consider what set of factors is enough to balance the chopped cluster.

While eliminating a shift of 18,000 persons across county lines, you have swapped 330,000 persons between the core part of the two metropolitan areas.  While there is certainly a connection between Medina and Akron, and Portage and Cleveland, the connections, as measured by census data is much stronger the other way:

Population in Urbanized Area:

Medina: Cleveland UA 88K, Akron UA 24K
Portage: Akron UA 73K, Cleveland UA 28K

Commuting to Main County of MSA:

Medina: Cuyahoga 29K, Summit 10K
Portage: Summit 19K, Cuyahoga 13K

The supposed increase in erosity is because of the odd shape of Cleveland, and the VRA.

[jimrtex]

As I noted, the 5% is acceptable only for legislative districts when there is no stricter state standards. jimrtex favors a more relaxed standard for apportionment regions, but that is ineffective at reducing regular chops. For CDs the range can be up to 1% if the state can show consistently applied principles, and there is no smaller range using those same principles. A 0.5% deviation is an easy way to insure a range within 1%.

You're reading Tennant as narrowly as Karcher had been read.

Good faith effort can not be so narrowly defined as:

GOOD_FAITH = 0.005.

While avoiding a mini-chop which would have moved 15,000 persons.   You sliced between Portage and Summit.  You did not make a good faith effort to preserve that urban county cluster.

And range criteria is the wrong one to apply.  If you'd like, I'll show the MALC proposal for Texas House seats that cranked every Republican/Anglo seat up to 4.99%.  If the redistricting application displayed another digit, they would have gone to 4.999%.  Good faith is attempting to hit the bulls eye, and scoring 10 if you are within or even touching the ring.  Bad faith is to carry the red arrows down to the target, and insert them so that they are touching the ring.

I don't find any comfort in Tennant that a 5% variation would be permitted in congressional districts. The opinion said that a range of 0.79% was considered a minor deviation under Karcher and is still minor today, despite the ability of computers to create plans with less deviation. Since it is a minor deviation the state needed to show that the deviation was necessary to meet legitimate state objectives, not merely a good faith effort to meet those objectives. A 5% deviation for congressional districts has not been seen as minor to SCOTUS in the past, and there is nothing in the opinion to suggest that would change. I also reiterate that SCOTUS has consistently used population range, not percent deviation, to measure inequality between the largest and smallest district as they did in Tennant and Karcher.

If it was possible to create a map with whole counties, and it was necessary to have a district with a greater deviation than 0.5%, it might well be be approved by the SCOTUS.

What has happened in Kirkpatrick v Preisler and Karcher v Daggett, is that the Missouri and New Jersey legislatures did not make a good faith effort to achieve equality.

In Kirkpatrick, the legislature claimed that they were making adjustments for non-resident population, and future growth.  But these were transparent guises.  When a plan was eventually drawn, it had the same deviation as that drawn by the legislature.

At the time of Karcher, 12 congressional plans had greater deviation than New Jersey, and 4 were of comparable magnitude.  But the New Jersey sloppily drew a partisan gerrymander, and tried to justify it on VRA grounds.

Where the Supreme Court has used the word "justify", the legislatures have tended to use "rationalization".

The SCOTUS has stated that there is no de minimis standard.  That means that it is impossible to determine whether larger ranges might be acceptable based on actual cases.

If I am going to justify a somewhat larger deviation,  I am going to have to show that my goal was to use whole counties, and that I had made a good faith effort in other portions of the state, where it was possible to achieve equality.   I can not demonstrate that with range.  I can with standard deviation.

I don't see any real evidence that the SCOTUS has expressed a preference for range over better and more appropriate measures of equality among multiple districts - it was just simpler for them to express their opinion.  In Karcher, they did make a comparison between two other districts.  They have also used average deviation - which is not a good idea either, since one can transfer from one overpopulated district to another even more overpopulated district with no penalty for the increased inequality.  Standard deviation penalizes collective disparity.

White v Regester has the actual house district populations in Texas.  They form a very nice normal distribution.   District populations since then have had worse and worse since then, approaching a uniform distribution.  

But the plan had an asymmetry in range.  It was possible to create one district with two counties that had a deviation of 5.8%, and there were no alternatives in compliance with the Texas Constitution.  At the same time there was a single county with a deviation of -4.7%.  Rather than violating the Texas Constitution by cutting a county.  They kept the 5.8% district, and violated the Texas Constitution by adding part of another county to the county that was naturally within the 5% limit, in a less egregious violation of the Texas Constitution, such that the overall range was 9.9%.  A good faith effort to achieve equality could have accepted the deviations of two districts at 5.8% and -4.7%.   Or the cut should have been applied to the 5.8% district (once a cut was made, it would have been possible to get much closer to 0.0% for that district.

And then they compounded their mistake by creating a multi-county district with a deviation of +5.7%, in an area where it was quite possible to have a district with near zero deviation.  That is, using total range permitted them to create another district with a larger deviation when it was absolutely not required.

Since then, range has been used to create or propose plans with greater overall inequality.  It is the wrong standard to use for measuring inequality.  There is no reason to dumb this down for the SCOTUS.

Note, I am not attempting to justify a 2.5% deviation for a Summit-Portage district.  I am permitting a larger deviation at this stage as part of a comprehensive method, devised in advance, as opposed to a post hoc rationalization of a legislature-drawn plan, that would:

(1) Be gerrymander resistant;
(2) Would largely adhere to political boundaries;
(3) Recognize larger metropolitan areas as significant communities of interest;
(4) Be reasonably compact; and
(5) Permit significant and real public participation.

Note that geographic compactness, may be simply one way to express conceptual integrity or concreteness.  Why are compact districts desirable?  It doesn't have anything to do with rubber bands.  It is because they are more likely to place neighbors with similar interests together.  An urban concentration is conceptually compact, as is a less urban area, even if it somewhat wraps around urban regions.

[jimrtex]

If I am going to justify a somewhat larger deviation,  I am going to have to show that my goal was to use whole counties, and that I had made a good faith effort in other portions of the state, where it was possible to achieve equality.   I can not demonstrate that with range.  I can with standard deviation.

I don't see any real evidence that the SCOTUS has expressed a preference for range over better and more appropriate measures of equality among multiple districts - it was just simpler for them to express their opinion.  In Karcher, they did make a comparison between two other districts.  They have also used average deviation - which is not a good idea either, since one can transfer from one overpopulated district to another even more overpopulated district with no penalty for the increased inequality.  Standard deviation penalizes collective disparity.

I can easily accept a statistical measure for district inequality, but the issue the court has considered most important is the disparity between the largest and smallest districts. The relative size from the largest to smallest district was the issue in Reynolds v. Sims, not the average or standard deviation. The fact that there is a disparity in the value of a person's vote drives the decisions related to one man one vote, and the greatest disparity occurs when viewing the extreme cases among districts.

Wesberry v Vandiver which was the lower court decision that was overturned by Wesberry v Sanders focused on relative deviation, which it erroneously referred to as "variance".   The plaintiffs in that case had suggested that Georgia should be limited to 15% relative deviation, and included in their evidence the number of districts across the country that exceeded 15% deviation, as well as 10% deviation.

In Wesberry v Sanders the SCOTUS noted both that the Atlanta district would have more than twice the average population (ie a relative deviation of more than 100%, and three times that of the smallest.   Georgia happened to be an egregious case where a district comprised of Fulton, DeKalb, and Rockdale had a population of twice the ideal, while the others were just sloppily unequal.

The Wesberry v Vandiver opinion said, "[w]e hasten to add that we neither expressly nor impliedly adopt any mathematical standard.  We know of no basis for an exact standard. ... Sanders v. Gray, supra, where sufficient basis existed. We use plaintiffs' suggested standard here in amplification of their contentions.  It is clear by any standard however that the population of the Fifth District is grossly out of balance with that of the other nine congressional districts of Georgia ..."

The SCOTUS opinion quoted "It is clear by any standard ...", and Justice Harlan in his dissent noted, "The Court's 'as nearly as is practicable' formula sweeps a host of questions under the rug. How great a difference between the populations of various districts within a State is tolerable? Is the standard an absolute or relative one, and, if the latter, to what is the difference in population to be related? Does the number of districts within the State have any relevance? Is the number of voters or the number of inhabitants controlling? Is the relevant statistic the greatest disparity between any two districts in the State, or the average departure from the average population per district, or a little of both? May the State consider factors such as area or natural boundaries (rivers, mountain ranges) which are plainly relevant to the practicability of effective representation?

There is an obvious lack of criteria for answering questions such as these, which points up the impropriety of the Court's wholehearted but heavy-footed entrance into the political arena."

The 1911 apportionment law stated that representatives "be elected by districts comprised of contiguous territory, and containing as nearly as practicable an equal number of inhabitants."  The "as nearly as practicable [equality]" language has found its way into the current legal standard, even though Congress appears to have deliberately removed it from the 1929 apportionment, and subsequent laws.  As Justice Harlan noted, it was as if the SCOTUS was saying that the 1911 standard would be applicable even if Congress had never passed a law at all.

Might Congress pass a law stating how equality was to be measured?  Of course.  And in the absence of congressional legislation, might a State legislature do the same?  Yes.

Use of range leads to mischievous inequality.

Consider the following hypothetical set of 20 districts with an ideal population of 100,000 persons in each. Two plans are submitted that comply with all state objectives. In plan A there are 10 districts with a population of 99,900 and 10 with a population of 100,100; giving a range of 200 (0.2%) and a standard deviation of 100 (0.100%). In plan B there are 19 districts with a population of 99,980 and one with a population of 100,380; giving a range of 400 (0.4%) and a standard deviation of 87.2 (0.087%). Plan B has the smaller standard deviation, but I find it unlikely that the court would ignore the fact that plan B has twice the maximum disparity present in plan A. So though I prefer standard deviation as a statistical measure, I think the law of one man one vote requires a range measure.

If these were the results of a physics experiment, you'd expect that the results had been fudged in either case, perhaps measuring instruments had less resolution than the displayed accuracy, or maybe some sort of quantization effect.

Let's try a different hypothetical.

Three plans are submitted: (Plan A) 10 districts with a population of 99,500 and 10 with a population of 100,500; giving a range of 1000 (1%) and a standard deviation of 500 (0.5%).  Plan B there are 19 districts with a population of 99,950 and one with a population of 100,950; giving a range of 1000 (1%) and a standard deviation of 300 (0.3%).   Plan C had districts of   100,139; 100,327; 100,103; 99,773; 99,948; 100,086; 100,500; 100,069; 100,109; 100,211;  100,335; 99,560; 99,738; 99,980; 100,002; 99,997; 99,752; 100,400; 99,558; and 99,413 giving a range of 1088 (1.088%) and a standard deviation of 287 (0.287%).

Plan A looks like that used in 1990s in Texas, where the Democratic seats in Harris County were uniformly underpopulated and the Republican seats were uniformly overpopulated; Plan B looks like that proposed by a rural legislator who didn't add Archer to Wichita county, as part of a plan to underpopulate West Texas districts, so that Harris County could be denied a representative.  In exchange, the Harris County Democrats were permitted to draw the districts in Harris County.

Plan C looks like someone used a spreadsheet to generate a normal distribution with a mean of 100,000, standard deviation of 250 (0.25%), knowing that there was a 64% chance of a district having a deviation greater than 0.5% (1 - 0.95²⁰), and an almost certainty that with enough district the range would be greater than 1%.

Which plan was a good faith effort to achieve equality, and which were the result of politicians with sticky fingers relying on the advice of lawyers?

[jimrtex]

We agree that one of the goals is significant and real public participation. I'd love to see statistical education get to the point where the standard deviation is as obvious to the public as the mean, but that day is far away. Standard deviation is just not a measure that is set up for public participation.

There's a reason that the legislatures and courts tend to use maximum deviation, average deviation or range, and that's because these are values they understand. Even if the Congress were to enact a standard for population inequality, and that standard were upheld by SCOTUS, I would be surprised if the choice is standard deviation.

It's true that the more accessible measures I mention allow for more obvious games to be played with district populations. A skilled mathematician can tilt the field with standard deviation, too. The point is that population inequality should be measured to show that after considering other criteria such as chops and erosity the plan has the least inequality. I'll rely on those other criteria to control gerrymandering. Thereafter I'd like the simplest measure to show the plan has addressed inequality. The courts have used range in many decisions, and it's simple to understand. Mathematically I found that it correlates slightly better than average deviation to the number of geographic units available to make a map.

I would use standard deviation and erosity as measures to (dis)qualify proposed plans, from which the general public, or a body representative of the general public may choose.

Here is a general outline of my process:

1) State agency (legislative council, secretary of state, state statistical agency) prepares data.  This may require coordination with the census bureau, and local governments well in advance of the census.  For example, if neighborhoods are used for forming districts in larger cities, they should be defined well in advance, so that they represent real communities of interest.

2) Census data release and massaged by state agency.

3a) Ordinary persons invited to prepare plans with apportionment regions.  The plan with the most regions, and the best equality and erosity becomes the target.

3b) Ordinary persons invited to prepare new or improved plans, with the target number of regions.

3c) Qualifying plans are identified.

4a) Large panels of voters vote on which plan they prefer.   The panels would be large enough to be statistically representative of the electorate.  Each panel member would be presented maps of the proposed apportionment regions that contained their residence.  Their ranking of a plan would be based solely on its effect on them personally.

4b) Best overall plan would be identified using Condorcet.   Within each apportionment region of the best overall plan, results would be re-evaluated to determine if voters in that region favor that plan.   That is, is their local concurrence with the statewide plan.

5) Iterate for areas of two or more adjacent regions, to determine whether there is a more amenable local plan that will fit withing the best overall plan.

6) Divide multi-district regions into single districts.

[jimrtex]

Returning to the issue of adjacent clusters I looked at MN.

Plan A puts the St Cloud cluster in with Mpls. In doing so I only need to add Rice to bring the Twin Cities region to within 0.16% of 5 CDs. The region range is 0.33% (st dev 0.12%). Within the TC there are the minimum 4 chops (2 Hennepin, Ramsey, Dakota) and CD 6 was kept more compact by going south into Hennepin rather than stretching east to the WI line.

What happens if you take 5 out to Eden Prairie, 3 down to Brooklyn Center and Robbindale, and 6 more of NW Hennepin.  That is, a little more treating it as a 3-way split of Hennepin, as opposed to starting from Minneapolis and working outward.

Another possibility would be to pull 6 out of Hennepin entirely, and go into Anoka.  Or maybe Minneapolis-St. Paul.  If that happened you would probably need to put S.St.Paul with the rest of Dakota, and make a small chop into Carver out west.

I see an advantage of two stage process, is that simplifies the first stage.  A liability is that it could place you into a trap in the second stage.

An alternative would be to be too choose the "best" plans at the first stage, and then develop all of them at the second stage.  But how do you know what to prune at the first stage.  In a sense, you might not want the "best" plans, but the best of a diverse set.

Plan B keeps the St Cloud cluster separate from Mpls. The erosity is lower than in plan A but the region range inequality rises to 0.98% (st dev 0.42%). It also raises a question about subregions of a cluster. The Dakota county CD 2 could exist as a region unto itself, but requires counties outside the cluster yet within the CSA. Keeping it to counties within the CSA might be a desirable extension from the pure cluster subregion.




Both these seem like plausible plans using the hard rule, so I don't see a problem here.

I'm kind of wary using the CSA.  For example, how do you justify leaving out McLeod and Sibley?  Or what if the numbers worked out better to include Sibley and Goodhue, but not Rice and Le Sueur?

This map illustrates a pitfall of a 2-stage process.  It produces an arguably better Metro map, with each of the big 4 counties having their own district (2 for Hennepin), but could be a tougher sell for the rest of the state.  

[jimrtex]

Thinking out loud...

We can think of traditional districts comprised of whole counties as being like building a stone wall, where we fit different sized and shaped stones together to build a wall of roughly equal height.  Had the legislatures not been so careless in maintaining the walls, such that some were 9-feet high, and others tumbled over, we might still be using them.

Modern block-based districting is more like using a 3D printer, where we can create phantasmagorical creatures with infinite plasticity.

Iowa still does it the old way.   Why can they do it?  The stones are of modest size, none so large that by itself it is too large.  And even when they are moderately large, they are dispersed in a way  that when the wall is build we can use them in different parts of the wall.  Iowa also uses limestone, so that the stones look almost like bricks, giving the surfaces of the wall a flat appearance, akin to might be achieved with bricks or sawn logs.

But can we pre-cut the larger stones, boulders really?  Perhaps at least score them so that they are easily cut?

So in Minnesota, we would have to precut Hennepin.  But it probably wouldn't hurt to cut Ramsey, Anoka, Washington, and Dakota.   What are the rules?  Counties of a certain size?  Or counties of a certain size within proximity of others?  Polk County, IA has more population than either Dakota or Anoka, and nearly as much as Ramsey.  Perhaps there could be a local option, the folks in Sherburne or western Stearns might prefer to be split, while those in Olmsted rigidly insist in being in the same district.

Or do we provide cuts of Polk County, IA, and add some rules that discourage their use?  Perhaps cutting of a county would have more of a penalty.  We can limit county cuts to more than is necessary, ... or reasonable, which might permit a double cut of Hennepin County.

A rule might be: max (2, ceil (pop/quota + 0.5))

Instead of apportionment regions based on whole counties, they could be based on whole building blocks covering an urbanized area.   This would provide a dividing line between the St.Cloud and Twin Cities sphere of influence in Sherburne County.

We don't want equal sized stones.  It is easier to fit them together when there is variation.  So splitting on townships, and maybe gathering smaller sized cities together works.

As an alternative, what if if we first created districts of more equal size, where the number is an integer multiple of the number of congressional districts.  These initial districts would be closer to the size of legislative districts, and we could apportion them to whole counties.   We would then assemble them to form a congressional district.  Perhaps some adjustment would be needed to get better equality, but maybe not.  While some of these preliminary districts might have a deviation of 5%, it is unlikely that all those assembled would have a collective deviation so large.  I suspect we would be down below 2%.

Some possible integers:

8 - a power of two which would permit pairwise merging.  Simple, but might result in elongated districts where they are joined end to end.

9 - a power of three, which might permit trio-wise merging, and more circular or square districts.

10 - easy to count on our fingers.

11 - maybe the reasons for a factorable number are bogus.

12 - factorable by 2 and 3, and 6, does this lead to hexagonal districts?

All 5?  Create 5 alternatives and vote on them?

This latter scheme might also be useful with respect to the VRA.  The initial smaller districts need not be drawn as obsessively based on race, and it may be possible to combine them to form congressional districts that are not strictly contiguous.

[jimrtex]

This map shows the metropolitan statistical areas in the southeastern USA.  A metropolitan statistical area is based on a core of an urbanized area (ie urban area of 50,000 or more persons).  The MSA includes the counties that contain the core urbanized area (central counties), as well as outlying counties with significant commuting to or from the central counties. 

The outlying counties may themselves contain urbanized areas or urban clusters which would have qualified them to be the central counties of a metropolitan or micropolitan statistical area, but were captured by a stronger core.  These may be thought to be more like satellite cities and larger towns, rather than suburbs.  Nonetheless they represent an urban setting with strong economic ties to the central counties.

Outlying counties may also contain smaller urban clusters (less than 10,000 persons) or have no urban population (totally rural).  Essentially, these areas provide relatively few jobs, but are within commuting distance of the central counties.  These counties typically have few residents, and their exclusion from the urban county cluster has little effect on the metropolitan population.   They might be included in a congressional district based on a metropolitan area, based on their proximity, but it is not essential.   And the identification with the "big city" for these more rural areas may be relatively weak.

In the tables that follow, counties are classified based on the urban area with the largest share of the county's population.

NUA: No Urban Area.  The county is totally rural.

NQUA: Non-qualifying urban area.  The urban area with the largest share of the county population has fewer than 10,000 persons total, or more a small isolated town in a largely rural setting.

NQC: Non-qualifying core.   The urban area with the largest share of the county population has more than 10,000 persons total, but fewer than 5,000 in the county (or less than 50% for counties of 10,000 or less persons).  These counties do not qualify as a central county, but represent more of a small overlap of a core area into the county, and are not that dissimilar from the previous classifications.

The above 3 classifications are shown in a lighter shade on the map, and shown in red in the tables.  They represent the least urban, more peripheral counties of a metropolitan area.

UC: Core area is an urban cluster with a population of greater than 10,000.  It represents a larger town, that could have served as the core of a micropolitan statistical area, had not the county been captured by the metropolitan area.

UA: Core area for county is an urbanized area, which always have a population greater than 50,000.  The urbanized area need not be the core for the metropolitan area.  Census Bureau delineation procedures protect urbanized areas that have grown together, so such an urbanized area can not truly be regarded as independent or separate, particularly given the commuting ties.

The following tables represent the portion of metropolitan areas within each state.  Metropolitan statistical areas are delineated without regard to state boundaries, but for redistricting purposes, only the area within each state is of interest.  The first number is the population of the metropolitan area within the state (in thousands), while the second number is the population excluding the more peripheral counties.  The number in parentheses is the number of counties, within the state, that are in the metropolitan area.

For each county, the county population (in thousands), and classification code are shown.

Alabama

Birmingham-Hoover, AL 1128  1004  (7): Jefferson County 658 UA; Shelby County 195 UA; St. Clair County 84 UC; Walker County 67 UC; Blount County 57 NQUA; Chilton County 44 NQUA; and Bibb County 23 NQUA.

Huntsville, AL 418  418  (2): Madison County 335 UA; and Limestone County 83 UC.

Mobile, AL 413  413  (1): Mobile County 413 UA.

Montgomery, AL 375  363  (4): Montgomery County 229 UA; Elmore County 79 UA; Autauga County 55 UA; and Lowndes County 11 NUA.

Tuscaloosa, AL 230  195  (3): Tuscaloosa County 195 UA; Pickens County 20 NUA; and Hale County 16 NQUA.

Daphne-Fairhope-Foley, AL 182  182  (1): Baldwin County 182 UA.

Decatur, AL 154  119  (2): Morgan County 119 UA; and Lawrence County 34 NQUA.

Florence-Muscle Shoals, AL 147  147  (2): Lauderdale County 93 UA; and Colbert County 54 UA.

Dothan, AL 146  102  (3): Houston County 102 UA; Geneva County 27 NQUA; and Henry County 17 NQUA.

Auburn-Opelika, AL 140  140  (1): Lee County 140 UA.

Anniston-Oxford-Jacksonville, AL 119  119  (1): Calhoun County 119 UA.

Gadsden, AL 104  104  (1): Etowah County 104 UA.

Columbus, GA-AL 53  53  (1): Russell County 53 UA.


Arkansas

Little Rock-North Little Rock-Conway, AR 700  671  (6): Pulaski County 383 UA; Faulkner County 113 UA; Saline County 107 UA; Lonoke County 68 UA; Grant County 18 NQUA; and Perry County 10 NUA.

Fayetteville-Springdale-Rogers, AR-MO 440  424  (3): Benton County 221 UA; Washington County 203 UA; and Madison County 16 NUA.

Fort Smith, AR-OK 188  188  (2): Sebastian County 126 UA; and Crawford County 62 UA.

Jonesboro, AR 121  96  (2): Craighead County 96 UA; and Poinsett County 25 NQUA.

Pine Bluff, AR 100  77  (3): Jefferson County 77 UA; Lincoln County 14 NUA; and Cleveland County 9 NUA.

Hot Springs, AR 96  96  (1): Garland County 96 UA.

Texarkana, TX-AR 57  43  (2): Miller County 43 UA; and Little River County 13 NQUA.

Memphis, TN-MS-AR 51  51  (1): Crittenden County 51 UA.


Florida

Miami-Fort Lauderdale-West Palm Beach, FL 5565  5565  (3): Miami-Dade County 2496 UA; Broward County 1748 UA; and Palm Beach County 1320 UA.

Tampa-St. Petersburg-Clearwater, FL 2783  2783  (4): Hillsborough County 1229 UA; Pinellas County 917 UA; Pasco County 465 UA; and Hernando County 173 UA.

Orlando-Kissimmee-Sanford, FL 2134  2134  (4): Orange County 1146 UA; Seminole County 423 UA; Lake County 297 UA; and Osceola County 269 UA.

Jacksonville, FL 1346  1346  (5): Duval County 864 UA; Clay County 191 UA; St. Johns County 190 UA; Nassau County 73 UC; and Baker County 27 UC.

North Port-Sarasota-Bradenton, FL 702  702  (2): Sarasota County 379 UA; and Manatee County 323 UA.

Cape Coral-Fort Myers, FL 619  619  (1): Lee County 619 UA.

Lakeland-Winter Haven, FL 602  602  (1): Polk County 602 UA.

Deltona-Daytona Beach-Ormond Beach, FL 590  590  (2): Volusia County 495 UA; and Flagler County 96 UA.

Palm Bay-Melbourne-Titusville, FL 543  543  (1): Brevard County 543 UA.

Pensacola-Ferry Pass-Brent, FL 449  449  (2): Escambia County 298 UA; and Santa Rosa County 151 UA.

Port St. Lucie, FL 424  424  (2): St. Lucie County 278 UA; and Martin County 146 UA.

Tallahassee, FL 367  275  (4): Leon County 275 UA; Gadsden County 46 NQUA; Wakulla County 31 NQUA; and Jefferson County 15 NUA.

Ocala, FL 331  331  (1): Marion County 331 UA.

Naples-Immokalee-Marco Island, FL 322  322  (1): Collier County 322 UA.

Gainesville, FL 264  247  (2): Alachua County 247 UA; and Gilchrist County 17 NQUA.

Crestview-Fort Walton Beach-Destin, FL 236  236  (2): Okaloosa County 181 UA; and Walton County 55 UA.

Panama City, FL 185  169  (2): Bay County 169 UA; and Gulf County 16 NQUA.

Punta Gorda, FL 160  160  (1): Charlotte County 160 UA.

Homosassa Springs, FL 141  141  (1): Citrus County 141 UA.

Sebastian-Vero Beach, FL 138  138  (1): Indian River County 138 UA.

Sebring, FL 99  99  (1): Highlands County 99 UA.

The Villages, FL 93  93  (1): Sumter County 93 UA.

[jimrtex]

Georgia

Atlanta-Sandy Springs-Roswell, GA 5287  5081  (29): Fulton County 921 UA; Gwinnett County 805 UA; DeKalb County 692 UA; Cobb County 688 UA; Clayton County 259 UA; Cherokee County 214 UA; Henry County 204 UA; Forsyth County 176 UA; Paulding County 142 UA; Douglas County 132 UA; Coweta County 127 UA; Carroll County 111 UC; Fayette County 107 UA; Bartow County 100 UA; Newton County 100 UA; Rockdale County 85 UA; Walton County 84 UA; Barrow County 69 UC; Spalding County 64 UA; Pickens County 29 NQUA; Haralson County 29 NQUA; Butts County 24 NQUA; Dawson County 22 NQC; Meriwether County 22 NQUA; Lamar County 18 NQUA; Pike County 18 NQC; Morgan County 18 NQUA; Jasper County 14 NQUA; and Heard County 12 NUA.

Augusta-Richmond County, GA-SC 378  325  (5): Richmond County 201 UA; Columbia County 124 UA; Burke County 23 NQUA; McDuffie County 22 NQUA; and Lincoln County 8 NUA.

Savannah, GA 348  348  (3): Chatham County 265 UA; Effingham County 52 UC; and Bryan County 30 UA.

Columbus, GA-AL 242  201  (4): Muscogee County 190 UA; Harris County 32 NQC; Chattahoochee County 11 UA; and Marion County 9 NUA.

Macon, GA 232  156  (5): Bibb County 156 UA; Jones County 29 NQC; Monroe County 26 NQUA; Crawford County 13 NUA; and Twiggs County 9 NUA.

Athens-Clarke County, GA 193  150  (4): Clarke County 117 UA; Oconee County 33 UA; Madison County 28 NQC; and Oglethorpe County 15 NQC.

Gainesville, GA 180  180  (1): Hall County 180 UA.

Warner Robins, GA 180  168  (3): Houston County 140 UA; Peach County 28 UC; and Pulaski County 12 NQUA.

Albany, GA 157  123  (5): Dougherty County 95 UA; Lee County 28 UA; Worth County 22 NQUA; Terrell County 9 NQUA; and Baker County 3 NUA.

Chattanooga, TN-GA 149  133  (3): Walker County 69 UA; Catoosa County 64 UA; and Dade County 17 NQUA.

Dalton, GA 142  142  (2): Whitfield County 103 UA; and Murray County 40 UA.

Valdosta, GA 140  109  (4): Lowndes County 109 UA; Brooks County 16 NQUA; Lanier County 10 NQUA; and Echols County 4 NUA.

Brunswick, GA 112  80  (3): Glynn County 80 UA; Brantley County 18 NQC; and McIntosh County 14 NQUA.

Rome, GA 96  96  (1): Floyd County 96 UA.

Hinesville, GA 78  63  (2): Liberty County 63 UA; and Long County 14 NQC.


Louisiana

New Orleans-Metairie, LA 1190  1190  (8): Jefferson Parish 433 UA; Orleans Parish 344 UA; St. Tammany Parish 234 UA; St. Charles Parish 53 UA; St. John the Baptist Parish 46 UA; St. Bernard Parish 36 UA; Plaquemines Parish 23 UA; and St. James Parish 22 UC.

Baton Rouge, LA 802  733  (9): East Baton Rouge Parish 440 UA; Livingston Parish 128 UA; Ascension Parish 107 UA; Iberville Parish 33 UA; West Baton Rouge Parish 24 UA; Pointe Coupee Parish 23 NQUA; East Feliciana Parish 20 NUA; West Feliciana Parish 16 NUA; and St. Helena Parish 11 NUA.

Lafayette, LA 467  467  (5): Lafayette Parish 222 UA; Iberia Parish 73 UA; Acadia Parish 62 UC; Vermilion Parish 58 UC; and St. Martin Parish 52 UC.

Shreveport-Bossier City, LA 440  413  (4): Caddo Parish 255 UA; Bossier Parish 117 UA; Webster Parish 41 UC; and De Soto Parish 27 NQUA.

Houma-Thibodaux, LA 208  208  (2): Terrebonne Parish 112 UA; and Lafourche Parish 96 UA.

Lake Charles, LA 200  193  (2): Calcasieu Parish 193 UA; and Cameron Parish 7 NUA.

Monroe, LA 176  154  (2): Ouachita Parish 154 UA; and Union Parish 23 NQUA.

Alexandria, LA 154  132  (2): Rapides Parish 132 UA; and Grant Parish 22 NQC.

Hammond, LA 121  121  (1): Tangipahoa Parish 121 UA.


Mississippi

Jackson, MS 567  510  (6): Hinds County 245 UA; Rankin County 142 UA; Madison County 95 UA; Copiah County 29 NQUA; Yazoo County 28 UC; and Simpson County 28 NQUA.

Gulfport-Biloxi-Pascagoula, MS 371  371  (3): Harrison County 187 UA; Jackson County 140 UA; and Hancock County 44 UA.

Memphis, TN-MS-AR 247  161  (5): DeSoto County 161 UA; Marshall County 37 NQUA; Tate County 29 NQUA; Tunica County 11 NQUA; and Benton County 9 NUA.

Hattiesburg, MS 143  131  (3): Forrest County 75 UA; Lamar County 56 UA; and Perry County 12 NUA.


North Carolina

Charlotte-Concord-Gastonia, NC-SC 1881  1881  (7): Mecklenburg County 920 UA; Gaston County 206 UA; Union County 201 UA; Cabarrus County 178 UA; Iredell County 159 UA; Rowan County 138 UA; and Lincoln County 78 UC.

Raleigh, NC 1130  1070  (3): Wake County 901 UA; Johnston County 169 UA; and Franklin County 61 NQC.

Greensboro-High Point, NC 724  724  (3): Guilford County 488 UA; Randolph County 142 UC; and Rockingham County 94 UC.

Winston-Salem, NC 641  602  (5): Forsyth County 351 UA; Davidson County 163 UA; Stokes County 47 UA; Davie County 41 UA; and Yadkin County 38 NQUA.

Durham-Chapel Hill, NC 504  401  (4): Durham County 268 UA; Orange County 134 UA; Chatham County 64 NQUA; and Person County 39 NQUA.

Asheville, NC 425  404  (4): Buncombe County 238 UA; Henderson County 107 UA; Haywood County 59 UA; and Madison County 21 NQC.

Fayetteville, NC 366  366  (2): Cumberland County 319 UA; and Hoke County 47 UA.

Hickory-Lenoir-Morganton, NC 365  328  (4): Catawba County 154 UA; Burke County 91 UA; Caldwell County 83 UA; and Alexander County 37 NQUA.

Wilmington, NC 255  255  (2): New Hanover County 203 UA; and Pender County 52 UC.

Jacksonville, NC 178  178  (1): Onslow County 178 UA.

Greenville, NC 168  168  (1): Pitt County 168 UA.

Rocky Mount, NC 152  152  (2): Nash County 96 UA; and Edgecombe County 57 UA.

Burlington, NC 151  151  (1): Alamance County 151 UA.

New Bern, NC 127  104  (3): Craven County 104 UA; Pamlico County 13 NUA; and Jones County 10 NUA.

Goldsboro, NC 123  123  (1): Wayne County 123 UA.

Myrtle Beach-Conway-North Myrtle Beach, SC-NC 107  107  (1): Brunswick County 107 UA.

Virginia Beach-Norfolk-Newport News, VA-NC 36  0  (2): Currituck County 24 NQC; and Gates County 12 NUA.

[jimrtex]

The next one I'd like to look at is LA. It has some clusters in close proximity and VRA concerns as well. Here's what I understand are the clusters on the map.

New Orleans (1.546) Jefferson, Orleans, Plaquemines, St Bernard, St Charles, St John the Baptist, St Tammany
Baton Rouge (0.970) Ascension, East Baton Rouge, Iberville, Livingston, West Baton Rouge (Iberville is on the map, but it should be off with the revised definition since the urban population is under 50% bringing the cluster to 0.925)
Shreveport (0.492) Bossier, Caddo
Lafayette (0.390) Iberia, Lafayette (but should this be disallowed due to a lack of connectivity?)
Houma (0.276) Lafourche, Terrebonne

The key is that one of the regions will have to contain a VRA district with >50% BVAP.

Yes Iberville is off under revised criteria.

The Iberia-Lafayette border is about 16% of the square root of the smaller county (Lafayette) so qualifies.  Moreover it is right on the direct route from Lafayette to New Iberia on the high not-so-low ground to the west of the Atchafalaya.

Maybe it demonstrates that Louisiana fails the Gingles test.

[jimrtex]

Oklahoma

Oklahoma City, OK 1253  1184  (7): Oklahoma County 719 UA; Cleveland County 256 UA; Canadian County 116 UA; Grady County 52 UC; Logan County 42 UA; McClain County 35 NQUA; and Lincoln County 34 NQUA.

Tulsa, OK 937  921  (7): Tulsa County 603 UA; Rogers County 87 UC; Wagoner County 73 UA; Creek County 70 UA; Osage County 47 UA; Okmulgee County 40 UC; and Pawnee County 17 NQUA.

Lawton, OK 130  124  (2): Comanche County 124 UA; and Cotton County 6 NQUA.

Fort Smith, AR-OK 93  0  (2): Le Flore County 50 NQUA; and Sequoyah County 42 NQUA.


South Carolina

Greenville-Anderson-Mauldin, SC 824  824  (4): Greenville County 451 UA; Anderson County 187 UA; Pickens County 119 UA; and Laurens County 67 UC.

Columbia, SC 768  709  (6): Richland County 385 UA; Lexington County 262 UA; Kershaw County 62 UC; Fairfield County 24 NQUA; Saluda County 20 NQUA; and Calhoun County 15 NUA.

Charleston-North Charleston, SC 665  665  (3): Charleston County 350 UA; Berkeley County 178 UA; and Dorchester County 137 UA.

Charlotte-Concord-Gastonia, NC-SC 336  303  (3): York County 226 UA; Lancaster County 77 UC; and Chester County 33 NQUA.

Spartanburg, SC 313  313  (2): Spartanburg County 284 UA; and Union County 29 UC.

Myrtle Beach-Conway-North Myrtle Beach, SC-NC 269  269  (1): Horry County 269 UA.

Florence, SC 206  206  (2): Florence County 137 UA; and Darlington County 69 UC.

Augusta-Richmond County, GA-SC 187  160  (2): Aiken County 160 UA; and Edgefield County 27 NQUA.

Hilton Head Island-Bluffton-Beaufort, SC 187  162  (2): Beaufort County 162 UA; and Jasper County 25 NQUA.

Sumter, SC 107  107  (1): Sumter County 107 UA.


Tennessee

Nashville-Davidson--Murfreesboro--Franklin, TN 1671  1544  (14): Davidson County 627 UA; Rutherford County 263 UA; Williamson County 183 UA; Sumner County 161 UA; Wilson County 114 UA; Maury County 81 UC; Robertson County 66 UC; Dickson County 50 UC; Cheatham County 39 NQUA; Hickman County 25 NUA; Macon County 22 NQUA; Smith County 19 NQUA; Cannon County 14 NQUA; and Trousdale County 8 NUA.

Memphis, TN-MS-AR 1027  989  (3): Shelby County 928 UA; Tipton County 61 UC; and Fayette County 38 NQUA.

Knoxville, TN 838  774  (9): Knox County 432 UA; Blount County 123 UA; Anderson County 75 UA; Roane County 54 UC; Loudon County 49 UA; Campbell County 41 UC; Grainger County 23 NUA; Morgan County 22 NQC; and Union County 19 NUA.

Chattanooga, TN-GA 379  336  (3): Hamilton County 336 UA; Marion County 28 NQUA; and Sequatchie County 14 NQUA.

Kingsport-Bristol-Bristol, TN-VA 214  214  (2): Sullivan County 157 UA; and Hawkins County 57 UA.

Johnson City, TN 199  180  (3): Washington County 123 UA; Carter County 57 UA; and Unicoi County 18 NQUA.

Clarksville, TN-KY 172  172  (1): Montgomery County 172 UA.

Jackson, TN 130  98  (3): Madison County 98 UA; Chester County 17 NQUA; and Crockett County 15 NQUA.

Cleveland, TN 116  99  (2): Bradley County 99 UA; and Polk County 17 NUA.

Morristown, TN 114  114  (2): Hamblen County 63 UA; and Jefferson County 51 UA.


Texas

Dallas-Fort Worth-Arlington, TX 6426  6359  (13): Dallas County 2368 UA; Tarrant County 1809 UA; Collin County 782 UA; Denton County 663 UA; Johnson County 151 UA; Ellis County 150 UA; Parker County 117 UC; Kaufman County 103 UC; Hunt County 86 UC; Rockwall County 78 UA; Wise County 59 NQUA; Hood County 51 UC; and Somervell County 8 NUA.

Houston-The Woodlands-Sugar Land, TX 5920  5849  (9): Harris County 4092 UA; Fort Bend County 585 UA; Montgomery County 456 UA; Brazoria County 313 UA; Galveston County 291 UA; Liberty County 76 UC; Waller County 43 NQUA; Chambers County 35 UA; and Austin County 28 NQUA.

San Antonio-New Braunfels, TX 2143  2033  (8): Bexar County 1715 UA; Guadalupe County 132 UA; Comal County 108 UA; Medina County 46 NQUA; Atascosa County 45 UC; Wilson County 43 NQUA; Kendall County 33 UC; and Bandera County 20 NUA.

Austin-Round Rock, TX 1716  1716  (5): Travis County 1024 UA; Williamson County 423 UA; Hays County 157 UA; Bastrop County 74 UC; and Caldwell County 38 UC.

El Paso, TX 804  801  (2): El Paso County 801 UA; and Hudspeth County 3 NUA.

McAllen-Edinburg-Mission, TX 775  775  (1): Hidalgo County 775 UA.

Corpus Christi, TX 428  428  (3): Nueces County 340 UA; San Patricio County 65 UA; and Aransas County 23 UC.

Brownsville-Harlingen, TX 406  406  (1): Cameron County 406 UA.

Killeen-Temple, TX 405  386  (3): Bell County 310 UA; Coryell County 75 UA; and Lampasas County 20 NQUA.

Beaumont-Port Arthur, TX 403  389  (4): Jefferson County 252 UA; Orange County 82 UA; Hardin County 55 UA; and Newton County 14 NUA.

Lubbock, TX 291  279  (3): Lubbock County 279 UA; Crosby County 6 NUA; and Lynn County 6 NQUA.

Waco, TX 253  235  (2): McLennan County 235 UA; and Falls County 18 NQUA.

Amarillo, TX 252  242  (5): Potter County 121 UA; Randall County 121 UA; Carson County 6 NQUA; Oldham County 2 NUA; and Armstrong County 2 NUA.

Laredo, TX 250  250  (1): Webb County 250 UA.

College Station-Bryan, TX 229  195  (3): Brazos County 195 UA; Burleson County 17 NQUA; and Robertson County 17 NQUA.

Longview, TX 214  175  (3): Gregg County 122 UA; Rusk County 53 UC; and Upshur County 39 NQUA.

Tyler, TX 210  210  (1): Smith County 210 UA.

Abilene, TX 165  132  (3): Taylor County 132 UA; Jones County 20 NQUA; and Callahan County 14 NQUA.

Wichita Falls, TX 151  132  (3): Wichita County 132 UA; Clay County 11 NQUA; and Archer County 9 NQC.

Midland, TX 142  137  (2): Midland County 137 UA; and Martin County 5 NUA.

Odessa, TX 137  137  (1): Ector County 137 UA.

Sherman-Denison, TX 121  121  (1): Grayson County 121 UA.

San Angelo, TX 112  110  (2): Tom Green County 110 UA; and Irion County 2 NUA.

Victoria, TX 94  87  (2): Victoria County 87 UA; and Goliad County 7 NUA.

Texarkana, TX-AR 93  93  (1): Bowie County 93 UA.

[jimrtex]

Does this reflect another set of criteria for consideration as urban county clusters? It would be helpful to concentrate on one version to see if it has any obvious weaknesses. I'm OK with your earlier definition for now, though a map that just has those qualifying multicounty clusters would be useful.

No.  I am showing the development from the beginning, starting with Metropolitan Statistical Areas.  My earlier maps had simply ignored the more rural counties and showed what was left.  Torie had said he wanted to look at all multi-county MSA.

So starting with all counties in each MSA:

(1) Exclude counties with no urban population.
(2) Exclude counties whose urban area with the largest population in the county has less than 10,000 total, and thus does not qualify as a core.
(3) Exclude counties whose urban area with the largest population in the county have less than 5,000 in the county (less than 50% for smaller counties), and also do not qualify to be a central county.

These counties are included in a MSA solely based on commuting.  In many cases this occurs precisely because they don't have much urban population, and relatively few local jobs.  Small towns at the same distance from the city will have more local jobs, some retail, perhaps doctors, etc., perhaps a small manufacturing plant, and relatively less commuting (commuting is based on percentage of a residents who work in the central counties).  If you had a map with arrows showing absolute magnitude of commuting flow, these areas would have skinny arrows, compared to adjoining "non-metropolitan" counties.  If there were a thematic map showing commuting percentages, there would only small variation.  But the definition reduces this information to a True/False value.

These excluded counties are shown as a lighter shade on the map.

[jimrtex]

The urban cluster definition has changed exactly how?  And the overlay is designed to solve exactly what problem? Jimtex's musings are just incomprehensible to the layperson.

You had written:

You must be logged in to read this quote.

The latest maps show all Metropolitan Statistical Areas.   It does not show Micropolitan Statistical Areas (both are CBSA).   It does include single-county MSA.  You should be able to ignore those without too much problem.   I think it is helpful to show them, as part of the process of understanding the derivation of urban county clusters.

Please note: "urban cluster" is a specific Census Bureau term for smaller "urban area", those with a population less than 50,000.   Before 2000, the census bureau had defined "urbanized area", which had a population greater than 50,000, and were intended to represent the suburban fringe around cities.

In 2000, they extended the concept to smaller populations, along with changing the rules for how they were delineated.  They retained the term "urbanized area" for larger areas, and to maintain comparability, added an over-arching term of "urban area" to include all areas, and "urban clusters" for smaller areas.