Official US 2010 Census Results

[muon2]

The state apportionments will be released before the end of this year. The redistricting data will be released during the first three months of 2011, completed by the end of March.

[muon2]

The census bureau will release the apportionment data on Dec 21. The press conference will be at 11 am EST.

[muon2]

I guess the county and city data will be released sometime next year ?

It will be released February 2010.

Redistricting data, ie data from the full census at the block level, must be released by Mar 31, 2011. Historically, the Census rolls out the data state-by-state over the spans of a few weeks before the deadline. Currently they show that as happening from Feb-Mar 2011. County and city data are built up from the block data.

[muon2]

LAST TEN SEATS:
Seat #426: Texas 35 (+3)              Priority: 728,933
Seat #427: Pennsylvania 18 (-1)   Priority: 726,147
Seat #428: California 52 (-1)         Priority: 723,412
Seat #429: Georgia 14 (+1)           Priority: 718,097
Seat #430: South Carolina 7 (+1)  Priority: 713,709
Seat #431: California 53 (nc)         Priority: 709,631
Seat #432: Florida 27 (+2)            Priority: 709,610
Seat #433: Washington 10 (+1)   Priority: 708,829
Seat #434: Minnesota 8 (nc)         Priority: 708,767
Seat #435: Texas 36 (+4)             Priority: 708,396

NEXT TEN SEATS
Seat #436: North Carolina 14 (+1)  Priority: 706,817
Seat #437: Missouri 9 (nc)               Priority: 705,802
Seat #438: New York 28 (-1)           Priority: 704,775
Seat #439: New Jersey 13 (nc)       Priority: 703,915
Seat #440: Montana 2 (+1)             Priority: 699,622
Seat #441: Louisiana 7 (nc)            Priority: 699,514
Seat #442: Ohio 17 (-1)                  Priority: 699,503
Seat #443: Oregon 6 (+1)              Priority: 699,455
Seat #444: Virginia 12 (+1)            Priority: 696,400
Seat #445: California 54 (+1)         Priority: 696,366

I had a slightly difference sequence from you. Did you use resident or apportionment population? Apportionment population is larger and includes overseas military and government personnel.

#431 FL 27 (713.4 K)
#432 WA 10 (711.9 K)
#433 TX 36 (711.9 K)
#434 CA 53 (711.3 K)
#435 MN 8 (710.2)

#436 NC 14 (709.1 K)
#437 MO 9 (708.5 K)
#438 NY 28 (706.3 K)
#439 NJ 13 (705.2 K)
#440 MT 2 (703.2 K)

[muon2]

Odd choice of states.

Not really, those are the states that have state legislative elections this year, so they need to redistrict ASAP.

Ah, gotcha.

The Census Bureau has tried to prioritize the data releases in terms of when states will need to have districts in place for the next election. Obviously the states with elections in Nov 2011 have the most pressing need. That's also why IL is likely to be early. The petitions for 2012 begin circulation in Sep 2011, and filing is in Dec. The Dems will want to have an approved map by May 31 in order to use their majority to get the map they want.

[muon2]

Illinois, Texas, Oklahoma, and South Dakota ship next week.

I'm ready and waiting.

[muon2]

The congressional districts that appear to have lost population are (within state ordered from most declining):

AL-7
AR-4
CA-31, 53, 38, 47
FL-10
IL-4, 1, 2, 17, 7, 9, 6
IN-7, 6
IA-5
KS-1
LA-2, 3, 5
MI-14, 13, 5, 12, 1, 9, 15, 11
MN-4, 7
MS-2
MO-1
NE-3
NJ-10, 8
NY-28, 27, 24, 25, 26, 29
NC-1
OH-11, 10, 6, 17, 5, 1, 4
PA-14, 12, 5, 4, 2, 3
RI-1
SC-6
TN-9
VA-2
WV-3, 1

Most of which are self-explanatory (inner cities, black belt, plains, rust belt), but I was somewhat surprised at IL-6, VA-2, and FL-10.

The suburbs that make up IL-6 are essentially built out. There was a lot of housing built a generation ago, and the children have moved lot reducing the population. The growth areas occur when old properties have been removed for higher-density housing, but that is not a large factor in the last decade.

[muon2]

 So I take a look at the data for Illinois released today:

  Big shocker:  Non Hispanic population -.8%
                       Hispanic population +32.5%

  If it weren't for Hispanics, our state's population wuold have been sunk.

Ill Ind

Yes.  Illinois' non-Hispanic white (-3%) and black (-1.3%) population both fell.  Asians and Hispanics grew.

Chicago shrunk by almost 7%, at just under 2.7 million.  That's about 150,000 less than  2.85 million 2009 census estimate.  If Houston's population has been grossly understimated, the Second City might be our fourth largest... but I doubt it.

Far west suburban Aurora (+38%) is now the second largest city in the state, passing Rockford (+2%).  Far southwest suburban Joliet (+39%) rocketed from seventh to fourth.   Naperville's population was up by 10.5% - but fell from fourth to fifth.

Like Chicago, Cook County lost population (-3.4%).  DuPage tread water (+1.4%).  The far-out Chicago suburban collar counties grew the most, particularly Kendall (+110%), Will (+35%), Kane (+27%) and McHenry (+19%).

Without Chicago, Cook gained about 18K in population.

It looks like the Hispanic growth was enough to have IL not lose a 2nd CD this decade.

[muon2]

Non Hispanic whites went from 52.4% to 45.3% while Hispanics went from 32% to 37.6%, if these trends continue then by the next census Texas will be Hispanic plurality.

Is this going to make Republican gerrymandering harder or it won't have much of an effect overall?

Are you talking about now, or in the future?  It might actually make things easier now, but you need to see how the numbers hit.  As for the future, well, you're guess is as good as mine.  The numbers kind of speak for themselves.

For now of course, nobody knows what happens in ten years.

Also, not that Texas has become a minority-majority state, does that mean that the DOJ can ask that almost half of its districts be VRA?

Well, first consider that VAP (not total population) is 49.6% white, 33.6% Hispanic.  Then you've got to rebalance using actual citizen VAP, which will be even less Hispanic.  Under LULAC, it's either going to be VAP or citizen VAP (though I suspect it's citizen VAP) for permissible minority-majority CDs.

If it's citizen VAP (and probably even if it's VAP), then consider that you're going to need at least 60%-65% of baseline Hispanics (not VAP or citizen VAP) in a CD to get Hispanic minority-majority VAP or citizen VAP.  There's only so many districts I can draw that will reach that number because the Hispanic population is pretty spread out (too many 20%-30% Hispanic voting districts, especially in the 'burbs) outside of the inner city core, which will be required to maintain certain CDs - for example Gene Green's CD is tough to get that much higher than 70% - inclusion of even a trivial amount of whiter suburbs creates problems.  Also, keep in mind LULAC can be used as a sword too to prevent requiring ridiculous looking strip districts, as the Austin to border CD was no good to create a Hispanic CD.

You draw the districts under 60% Hispanic - the Supreme Court will yell at you for impermissible dilution and you'll get bad results, the Republicans will win far more often than you want them to, especially if there's no blacks or its not an inner city core or Austin, they'll almost always win.  Another structural problem with creating good Hispanic Democratic districts, which connects to this, is that too many blacks are locked up in east Texas where they're f-cked and you can't get to them.

The fact is that 3 more Hispanic CDs may well be required than 2000, but I suspect it won't be any more, unless patterns are changed from what I was seeing.  And apart from the Dallas CD, the GOP will attempt to use Doggett and Green for the other two, if need be.  I need to see what the voting districts look like and draw a few maps.  The voting districts won't be changed much except to split or combine, so it's very useful.

The standards for Hispanic districts will probably make up some of the key redistricting cases in this decade. One part is the VAP versus CVAP problem you alluded to. The 5th circuit used CVAP in LULAC, but the SCOTUS avoided the question and decided the case on other grounds. In Bartlett the SCOTUS noted the issue of CVAP, but again avoided the question since it was less relevant for a Black population and they could decide the case without going into CVAP. CVAP will be further complicated since the 2000-cycle of cases had citizenship on the census long form, but it is absent on the short-form only 2010 census. The statistics from the ACS are all that's available and they are much weaker statistically than the long-form numbers from 2000.

In the 7th circuit there was a rejection of plaintiff's claims for a 65% standard for Hispanic districts to account for citizenship in Gonzales v Aurora. They didn't have Bartlett yet and the case wasn't taken to the SCOTUS. There is also the question of whether it is required to create 50%+ VAP Hispanic districts where a CVAP majority isn't possible.

A second issue that will arise is the conflict between Black and Hispanic districts. SCOTUS opinions have been about the minority group's voting rights compared to the White majority. Bartlett makes clear that the only protected section 2 groups are single race majorities. There is no guidance to resolve a situation where either a Black-majority or Hispanic-majority district can be drawn, but not both.

This will be an interesting cycle for redistricting cases.

[muon2]

muon2, in the suburban/urban areas of Houston, DFW and San Antonio, in order to draw more 50% Hispanic VAP CDs, you'll have to dilute the inner Hispanic urban core, as the suburbs are filled with far too many 10%-30% Hispanic VTD.  Even then, I'll be lucky to get 2, maybe 3 more CDs.  And no one's gonna like that, as you'll marginalize the present Dem CDs and the new ones.

Sam, I think that the best plan is first to look at how districts perform at electing candidates of choice, probably by regression analysis of voting patterns. However, once suitably performing districts are drawn then, if there is not a roughly proportional number of districts for the minority, additional majority minority districts would be drawn where possible. That would eliminate the dilution problem in the urban core, yet still provide some level of VRA protection for additional population so that the remainder is not easily cracked.

[muon2]

muon2, in the suburban/urban areas of Houston, DFW and San Antonio, in order to draw more 50% Hispanic VAP CDs, you'll have to dilute the inner Hispanic urban core, as the suburbs are filled with far too many 10%-30% Hispanic VTD.  Even then, I'll be lucky to get 2, maybe 3 more CDs.  And no one's gonna like that, as you'll marginalize the present Dem CDs and the new ones.

Sam, I think that the best plan is first to look at how districts perform at electing candidates of choice, probably by regression analysis of voting patterns. However, once suitably performing districts are drawn then, if there is not a roughly proportional number of districts for the minority, additional majority minority districts would be drawn where possible. That would eliminate the dilution problem in the urban core, yet still provide some level of VRA protection for additional population so that the remainder is not easily cracked.

And I'm just telling you that you're going to end up coming to my conclusion.  You almost don't need regression analysis when you know most of these areas personally.    But let me extrapolate.

By my maths, by VAP numbers, it says that we should aim for 12/36 Hispanic majority-minority (50%) CDs under the VRA (33.6% VAP/36 CDs).  Now, the CDs in El Paso and along the border/South Texas are just going to have more than 50% VAP by their nature, and, furthermore, you're going to require more like 55% VAP (and maybe more) to ensure in these areas that they don't vote the "wrong way" all the time (they still may anyway, if present trends continue, which is, of course, a giant if), so there's going to be some slippage based on this factor alone which is probably worth at least one CD, maybe two.

Meanwhile, in the cities I mentioned, as I'm sure you realize - in Houston, it's going to be hard enough to draw two majority-minority Hispanic CDs.  But maybe the DOJ forces it on them - Al Green won't be happy, but that's his problem.  I don't really see how you do more, because all of the rest of the big Hispanic numbers (i.e. over 40%) are locked within blacks, or impossible to get without diluting blacks (I know the f-ing geography too well) or are, surprise, surprise, voting Republican, less than the other areas, but enough to cause dilution problems.

In DFW, one Hispanic district will be drawn, and can be, but I don't see how you do another.  Just look and see how many 20%-30% Hispanic VTDs there are in the DFW area and how impossible they are to unlock without screwing up other places.

In Austin, no one's ever figured out how to unlock the Hispanics, and there's really only one answer - combine them with the other half of San Antonio Hispanics, but I already proposed that as something that may well be done.  The liberal whites will get screwed then, but they already are.  I would try to combine the other half of San Antonio Hispanics with somewhere, but I then just get another marginal Hispanic majority CD.  Ugh.

So that leaves us with 9 majority-minority CDs, one lost because of normal population inequalities and the Hispanic undervote, and at least two others lost because of the damn spread out nature of the Texas Hispanic population (do I need to mention how many Hispanics are lost in west Texas - they don't really vote the right way, so no one cares that much - and I tried unlocking them but Lewis told me it was illegal...  ).  

Maybe you can figure out a way to create an Austin Hispanic majority-minority CD without attaching it to San Antonio - that's about the only other way I can think of to get another minority-majority CD, but the surrounding areas are simply not that favorable to the task - I'll reexamine Williamson and Hays when the new numbers come out.

I think we aren't so far apart. I have no doubt that reaching 12 Hispanic CDs in TX is nigh impossible. That's the target, however, and one should make a good effort to reach that without diluting other districts so much that they cease to be opportunities.

For instance, since one is still under the goal one looks at areas like Harris Co for 2 districts rather than one. For the Latinos there it's a matter of having two opportunities, rather than zero or one. I posted that one on the TX thread back in Jan, and your comments suggest that the map may well go that way, especially as it doesn't impact the GOP districts there.

In DFW I agree that there is only one district, precisely because you can't draw two at over 50% VAP. And I too will be curious to see the actual numbers in central TX.

[muon2]

Lastly, I can tell you that the Hispanics outside the Houston, DFW and Austin metros are going to demand 50% CVAP - otherwise it's too easy to design districts that will vote Republican.

In the last round CVAP could be easily surmised from the long form data. That leaves only the ACS which has much less statistical reliability than the long form data. Also, the Census Bureau acknowledges that the 2009 ACS citizenship data doesn't always match up well with the new 2010 census block groups, since it used 2000 geography. The 2010 ACS with the new geography will be out probably late this year, so where does that leave states like TX and IL that have to create maps rather early?

[muon2]

That majority black MO-01 may be possible after all. The 2005-2009 ACS estimate for Bellefontaine Neighbors was 50.3% black, 48.4% white, so I guess the ACS did not pick up on white flight in North STL County very well. Had Riverview Village as 59.5% black, too (although was much closer for Spanish Lake, estimating 73.9% black).

It's still going to be very difficult to do.  By my math, the average Missouri CD will have about 748,000 residents.  A majority would be about 374,000.  There are about 390,000 African Americans in St. Louis city and county combined - before taking into account Hispanic status.   I doubt you'd be able to draw a district that captures 95% of St. Louis City and County's African-Americans - and that's before deciding whether black Hispanics should qualify under whatever race-based standard you are applying.

I'm not sure why creating such a district is necessary, since any district centered on St. Louis city will likely vote for the Democrat, anyway, similar to how the African-American population would vote.

It's not only about the party preferred by the black population, but by the candidates preferred by the minority group. If there is a clear difference in voting preference between blacks and whites, even in the primary, and it is possible to create a >50% VAP black district, then failure to do so can be the basis for a federal VRA challenge.

[muon2]

113th CD demographics have been released, though doesn't appear to list VAP or white non-hispanics.

http://www.census.gov/newsroom/releases/archives/american_community_survey_acs/cb13-tps07.html

Those are both cross tabs from race, hispanic and age. The easy stats don't list them, just the main categories. This is also only from the 1-year sample so the statistics get weaker as one tries to cross tabulate different questions.

[muon2]

City and town estimates for July 1, 2012 were just released by the Census. The press release highlights the growth in TX.