Greg's Opinion Greg's big blog of whatnot


Harris County Redistricting: The Demographics [UPDATE]

I'm still refining the Google version of the Harris County Commissioner Precinct map. For now, though, here are the demographics, reformatted from the PDF map of the draft plan ...

         Anglo            Hispanic         Afr-Am            Asian
1 Total 187,320 (18.0%)  392,565 (37.7%)  400,005 (38.4%)   55,840 ( 5.4%)
  18+   161,903 (21.4%)  251,869 (33.4%)  291,693 (38.6%)   45,263 ( 6.0%)

         Anglo            Hispanic         Afr-Am            Asian
2 Total 310,908 (31.4%)  566,738 (57.2%)   81,577 ( 8.2%)   26,037 ( 2.6%)
  18+   248,936 (35.7%)  366,570 (52.5%)   58,219 ( 8.3%)   19,642 ( 2.8%)
  CVAP  260,519 (50.4%)  194,774 (37.7%)   48,851 ( 9.4%)   12,349 ( 2.4%)

         Anglo            Hispanic         Afr-Am            Asian
3 Total 416,970 (40.0%)  356,206 (34.2%)  149,009 (14.3%)  113,576 (10.9%)
  18+   327,376 (43.2%)  235,551 (31.1%)  103,277 (13.6%)   86,772 (11.4%)

         Anglo            Hispanic         Afr-Am            Asian
4 Total 434,628 (42.8%)  356,031 (35.0%)  144,754 (14.2%)   73,095 ( 7.2%)
  18+   347,415 (47.4%)  228,580 (31.2%)   97,792 (13.3%)   54,280 ( 7.4%)

I added the CVAP numbers for Pct. 2 since it is obviously the one of most interest for Hispanic numbers. I'm hoping to do the same for Pct. 1, but I believe the numbers for Total Population and 18+ Population are demonstrative enough to conclude that it meets the criteria for a VRA district. Expect to hear about this from people with a greater interest than I have in it.

The biggest reason? It's retrogressive. Some context missing from the PDF of the plan is what the district looks like in its current format, as of the 2010 Census data. In order to shed some light on that, I ran those numbers from the Lege Council's data and got the following.

           2001     2010 BASELINE       2011 PROPOSED
          -------   ---------------     ---------------
Total               888,572             991,395
Anglo     (36.9%)   248,481 (28.0%)     310,908 (31.4%)
Hispanic  (52.1%)   533,812 (60.0%)     566,738 (57.2%)
Afr-Am    ( 7.8%)    83,886 ( 9.4%)      81,577 ( 8.2%)   

VAP                 626,274             697,829
Anglo      (41.5%)  203,286 (32.5%)     248,936 (35.7%)
Hisp       (47.6%)  345,683 (55.2%)     366,570 (52.5%)
Afr-Am     ( 7.6%)   58,788 ( 9.4%)      58,219 ( 8.3%)

CVAP                462,610             522,280
Anglo               219,504 (47.4%)     260,519 (50.4%)    
Hispanic            183,729 (39.7%)     194,774 (37.7%)
Afr. Am.             43,701 ( 9.4%)      48,851 ( 9.4%)

First things first, since it's been a while ... if you're wondering how the CVAP count for Anglos can be higher than the 18+ counts for Anglos, read this. It has as much to do with the counts coming from a non-Census survey (the Census Bureau's American Community Survey, to be precise) as it does the fact that datapoints from as far back as 2005 are included in the counts. While it's an obvious curiosity for Harris County in particular, the percentage shares that the ACS gets for Total Pop and 18+ Pop are typically within a couple of points of being in line with the apples-to-apples Census percentages.

The 2001 basis for Commissioner Precincts (taken from here) is how the district looked when it was created last decade. Comparing it to that, it looks like Pct. 2 is improved on Hispanic numbers. But the baseline for comparison should be the 2010 numbers for the existing district. On that basis, this map looks problematic.

At first glance, it seems possible to me that the issue of retrogression could be fixed without necessarily altering the partisan desires of those on Commissioners Court in drawing the district. But if it goes through Commissioners Court as-is, I wouldn't expect it to come out of the other side of the legal process the same way.

ADD-ON: The County Attorney's site now has a link for redistricting info.

SIDENOTE: I know I've written this out somewhere else before, but just a methodological note for CVAP in particular ... the counts are generated by me. My method is to visually check all Census block groups that are fully contained within the district and to make judgment calls on those that are split. In the case of Pct. 2, I feel that the Hispanic numbers are at the high end due to including a higher degree of split block groups that were majority Hispanic than I might otherwise feel comfortable including. From there, I run the list against the Census Bureau's counts for each block group. It's not perfect, but checking it against how the Texas Legislative Council does it for State House districts, it's proven to be a reliably accurate method against that yardstick.

UPDATE: A quick election comparison for the existing Pct. 2 vs the proposed draft Pct. 2 ...

  • Existing Pct, 2 - 2010 results: Bill White won 49.4% - 48.7%
  • Proposed Pct. 2 - 2010 results: Bill White lost 45.6% - 52.6%

Related Posts:

Comments (3) Trackbacks (0)
  1. I am not sure this is retrogressive at all. The first dispute is whether the district is protected by the VRA at all, since no minority group formed a majority of likely voters. Influence districts are good politics and political correctness sometimes, but not required. Second, after looking at the incredible contortions the mapdrawers went to, sharply using race as the defining feature of the boundaries of the new district, carving black pockets away from Moorman’s district, I have a hard time imagining how to make the district more Hispanic without violating more communities of interest, neighborhoods, compactness and the like. A Hispanic candidate (though not necessarily a candidate of choice of Hispanics) might be competitive to replace Eversole at some point, based on these proposed districts.

  2. There are some very obvious places that were left out of the northern extension of Pct 2 that are CVAP majority Hispanic, as well as some areas around Ellington Field. A quick check on my part suggests that they can add those areas, remove a portion of the Kingwood additions and avoid reducing Hispanic numbers while still making it more GOP friendly than the current district.

  3. Fair enough. But if the commissioners have other interests in keeping certain neighborhoods together, spreading facilities among the commissioners, maintaining so much as possible of the current districts, etc., they are permitted to make that choice instead.

Leave a comment

Trackbacks are disabled.

Fatal error: Call to undefined function xwp_footerx() in /nfs/c01/h02/mnt/4748/domains/ on line 18