Greg's Opinion Greg's big blog of whatnot



Belatedly, it's worth noting that I'm back in Austin for the legislative session. Theoretically, that would mean a slowdown in blogging. But given the pace over the past several months, who knows.

For the sake of making myself feel better, the work product over this span of time seems to be about 568 pages of notes from legislative and budget board hearings between the end of the last legislature and the beginning of this one. I'm sure it'll eventually be some fascinating reading for an archaeologist many years from now.

To make matters even more fascinating, the reading list has turned to matters more along the lines of a thrilling class on Public Administration. I've caved into the textbook pricing scheme and picked up some long-lost reading that includes some updates since my time in college. The Washington Monthly obliges with more recent spins on the subject. As fascinating as I find the subject matter, it's not exactly the most "blog-friendly" material.

With that, I expect a bit of a slowdown, but also some time to refocus on what it is that I'd like to spend more time blogging about (or updating the Almanac with). Until then, I'll be sure to update sparingly between now and the end of the legislative session.


2009-13 ACS Update

The end of the year means new Census data being released. I'm saving most of my work until the Citizen Voting Age data is out, but here are the top lines for total population in Harris County, with previous ACS updates included to show the gradual change over time:

          Tot. Pop. '10 (%) | Tot. Pop. '11 (%) | Tot. Pop. '12 (%) | Tot. Pop. '13 (%) 
TOTAL     4,092,459         | 4,025,409         | 4,101,752         | 4,182,285
Anglo     1,349,646 (33.0%) | 1,353,868 (33.6%) | 1,354,869 (33.0%) | 1,361,568 (32.6%)
Hispanic  1,671,540 (40.8%) | 1,621,065 (40.3%) | 1,671,262 (40.7%) | 1,717,940 (41.1%)
Afr.-Am.    754,258 (18.4%) |   747,398 (18.6%) |   775,085 (18.9%) |   774,120 (18.5%)
Asian       249,853  (6.1%) |   246,924  (6.1%) |   257,467  (6.3%) |   262,251  (6.3%)
Other        67,162  (1.6%) |    56,154  (1.4%) |    43,069  (1.1%) |    66,406  (1.6%)

On a technical note, this is all based on the 5-yr dataset, which is the only dataset that provides CVAP details at the block group level. The one-year data at the county level, however, shows a more current snapshot of the county:

         5-yr ACS             1-yr ACS
        4,182,285            4,336,853   
Anglo   1,361,568 (32.6%)    1,376,670 (31.7%)
Hisp    1,717,940 (41.1%)    1,803,547 (41.6%)
AfrAm     774,120 (18.5%)      798,658 (18.4%)
Asian     262,251 ( 6.3%)      276,803 ( 6.4%)
Other      66,406 ( 1.6%)      81,175  ( 1.9%)

I asked some people smarter than me about how the five year set was calculated. I figured they may weight more recent years or do something fancy. Turns out, there's no magic to it at all. Everything is weighted the same. What this means is that the "Pulaski Effect" leads to the 2012 and 2013 5-yr data sets closely tracking with the full 2010 Census data. That's due to the aggregate of yearly data in, say, the 2008-2012 data set averaging somewhere toward the middle of that range - which is precisely 2010. I trust that everyone in the world was as curious as I was about that.

Again, CVAP data and mapping to happen as soon as the Census puts numbers online.


In Session

Today, I'm off to settle into a new workspace and a temporary residence in order to work with my new State Representative, Gene Wu, in Austin. Before anyone thinks to call, comment, or text about how exciting any of that is, you should be reminded that I was raised to loathe all things Austin. While I'm always excited to work on behalf of my community in HD137, I'm a little leery of most things Austin-related. So the emphasis for me is on the word "temporary." My intent is to be back home in Houston for as many weekends as possible. So I think that'll stave off much of the homesickness.

My anxiety over changes in geography aside, it's the change that I think is warranted for my blogging habit that warrants more attention. I'm not yet convinced or told that the blog should be iced. I'm not overly concerned if that decision does come from either myself or someone else. But I don't see the topical coverage of the blog shifting toward the legislative session and probably not much on Texas politics. That's a fairly significant limitation to start with. But part of this limitation has to do with the fact that I'm much more eager to flesh out the Texas Political Almanac. That gives me an outlet for organizing my thought and archiving a lot more information that I come across.

Another outlet is that Gene wants to launch a blog to cover much of the work that comes up in session. Time will tell how well that comes to fruition, but we expect it to be something different than what Aaron Pena and Trey Martinez Fischer have tried in the past. We'll have more to announce on that front in a few days.

Given that the time demands of work and my desire to spend more time on the Almanac represent a significant time limitation to what ever else I'll blog about, I think it's fair to say that there will be less written here. Particularly while the legislative session is in gear. Whether that ultimately turns into a once-a-day, once-a-week or just plain irregular posting pattern, I don't know.

Topic-wise, there is always ample news coverage to feed the habit. But I'd probably like to narrow it down a bit and focus more on one or two topics. And I honestly don't know what that would be as I type. I might like to cover City Hall and local news a bit more. But I suspect the need for a less wonky focus might come in handy. So this remains an open question for me. I'm sure an answer will come to me soon.


2007-11 Citizen Voting Age Population Update

I missed out on commenting on the Chronicle's coverage of the recent update on Census data. This comes from the American Community Survey's annual rolling update to their population counts.

I've only scratched the surface and updated some of my counts on how the total population translates down to citizen voting age population. Here are the topline numbers now. I owe it to myself to double-check these for accuracy, but there are some interesting notes for what turns up here. These are all taken from the 5-year ACS summary.

          Tot. Pop. '10 (%)   Tot. Pop. '11 (%)
TOTAL     4,092,459           4,025,409
Anglo     1,349,646 (33.0%)   1,353,868 (33.6%)
Hispanic  1,671,540 (40.8%)   1,621,065 (40.3%)
Afr.-Am.    754,258 (18.4%)     747,398 (18.6%)
Asian       249,853  (6.1%)     246,924  (6.1%)
Other        67,162  (1.6%)      56,154  (1.4%)
            18+ Pop. '10 (%)    18+ Pop. '11 (%)
TOTAL      2,944,624            2,893,717
Anglo      1,085,630 (36.9%)    1,085,427 (37.5%)
Hispanic   1,082,570 (36.7%)    1,049,076 (36.3%)
Afr.-Am.     541,108 (18.4%)      540,203 (18.7%)
Asian        194,956  (6.6%)      193,555  (6.7%)
Other         40,360  (1.4%)       25,456  (0.9%)
           CVAP-09 (%)        CVAP-10 (%)         CVAP-11  (%)
TOTAL     2,195,535          2,230,550          2,276,903
Anglo     1,090,624 (49.7%)  1,051,265 (47.1%)  1,048,230 (46.0%)
Hispanic    494,695 (22.5%)    530,490 (23.8%)    560,416 (24.6%)
Afr.-Am.    481,492 (21.9%)    506,150 (22.7%)    519,122 (22.8%)
Asian       106,547  (4.9%)    120,660  (5.4%)    125,733  (5.5%)
Other        22,177  (1.0%)     21,985  (1.0%)     23,402  (1.0%)

Did you notice that the raw number and percentage of total and 18+ Hispanic population decreased from the '10 counts to the '11 counts? Keep in mind that the ACS data isn't the same as the Census. The methodology for counts isn't the same. But it's still interesting to see a drop in population share. Even more interesting is that they come as the Citizen Voting Age Population rose for Hispanics in both the overall estimate and the share of the county's population.

At some point during football games tomorrow, I'll get around to both mapping this out, double-checking my math and digging into more granular detail. My hunch for now is that much of the change seen here may be due to methodology changes as much as actual numerical growth patterns. I'll update as time and findings permit. Here's the full update I did from the 2010 results, if you're up for some comparison.


The Year Ahead

Up till now, I've generally subscribed to Jim Carville's maxim that "I wouldn't want to work for any government that would be willing to hire me." But this is Texas ... we seem to need a bit of help.

So, starting in January, I'll be working in the legislature for Gene Wu. He won the job title of State Representative fair and square, so I'll be settle for the role of "bill monkey", I guess. This'll be my first time to work the Lege and I'm looking forward to building on what Gene started the day he started campaigning. HD137 is my home and there's no other spec of dust on the globe that I care about more. Our hope is to have a blog for either Gene or the entire office staff once the session is in gear. As soon as there's anything to report on that, I'll pass it on.

What that means for this little blog is currently under review. To be honest, I'm more committed to find more time to build That project has been a stop-and-start effort for a number of years now yet I think it holds the most potential. Since building that site helps build the knowledgebase I'll need working with Gene, I'm eager to spend more time with that project on a day-to-day basis. Whether the 10+ year blogging project goes on hiatus, a change in focus, or gets put on ice permanently ... those are among the options under consideration.

One of the tangents on my mind is that I'm not sure that TXPA is the greatest repository for map-based information. Certainly, I can add a map to a page without any problem. But the type of map that goes on those pages tends to be of the more self-explanatory variety. A number of maps that I like to blog about, however, require a bit more explanation and tend to get more into weeds which I've not yet organized into TXPA.

A higher-order issue for the upcoming session is that the pace of work builds to a sprint toward the end of the session. So there's a challenge of starting off with the hope that I can maintain any kind of productive pace for writing, explaining, opinionating, pontificating and whatnot ... only to see the time for such endeavors dry up in the spring. We'll see what the future holds.

Tagged as: , 2 Comments

Election Season Math: In Review

Following from some earlier posts about number-crunching in election season, here's a starter for what I saw as the Early Votes came in ...

At the county level, there were multiple ways of looking at how the Early Vote crowd voted. The two different individualized scores showed the electorate at 45.45% Dem and 54.16% Dem at the pessimistic and optimistic ends of the spectrum. Accounting for the biases seen in the 2008 electorate, these two scores placed the outcome at about 48-49% Dem.

The DPI-by-Precinct scores showed the Obama results clocking in at 49.98% Dem - less than a half point behind the pace from 2008. And since that score is in a multi-candidate field, it would be enough to open up Election Night with a small lead.

The reality was an opening mark of 48.00% (behind Romney) for Obama and a closing mark of 49.38% (ahead of Romney) for Obama.

Below is what the results showed among some of the districts in Harris County with contested district races. Two obvious DPI scores I chose were ones that had an easy correlation to 2012: Barack Obama and Adrian Garcia. I didn't expect either of them to do as well as they did in 2008, so I added CO Bradford's results from his District Attorney race to see how a narrow loss in 2008 might translate. Later in the process, as the results started showing a very 50-50 county, I added two judicials: Ashish Mehendru and Josefina Rendon. Rendon was the narrowest win in 2008 and Mehendru's showing was the low end of the range from that year.

Finally, to compile a somewhat pessimistic average DPI of my own, I folded each of these into an average DPI score. Traditionally, the DPI scores created by others are just that - an average. And on a good day, they'll tell you what they're averaging.

                                    DPI SCORES
        | ACTUAL EV |   Avg   | Obama  Garcia  Bradford | Mehendru  Rendon
HD134   |   42.6%   |  43.0%  | 46.1%   49.8%   39.9%   | 39.7%    39.5%
HD135   |   37.7%   |  39.5%  | 39.3%   45.2%   38.3%   | 36.6%    38.0%
HD137   |   63.6%   |  63.2%  | 62.9%   67.5%   62.0%   | 61.2%    62.6%
HD139   |   79.2%   |  77.5%  | 77.0%   81.0%   76.7%   | 75.9%    77.1%
HD141   |   89.4%   |  86.5%  | 86.1%   88.2%   86.4%   | 85.6%    86.4%
HD143   |   71.0%   |  68.5%  | 64.4%   74.2%   68.0%   | 66.4%    69.7%
HD144   |   46.6%   |  51.7%  | 47.4%   58.6%   51.4%   | 48.8%    52.3%
HD149   |   58.9%   |  55.0%  | 54.6%   58.9%   54.1%   | 53.1%    54.2%
HD150   |   28.6%   |  31.7%  | 31.2%   37.7%   30.9%   | 28.8%    30.1%
CD2     |   30.6%   |  36.1%  | 36.6%   42.5%   34.9%   | 33.4%    33.4%
CD7     |   34.7%   |  38.5%  | 40.1%   44.5%   36.6%   | 35.7%    35.7%
CD18    |   76.5%   |  77.6%  | 77.3%   81.1%   76.9%   | 76.3%    76.3%
Const 1 |   59.0%   |  61.3%  | 61.8%   67.3%   59.4%   | 59.1%    59.1%
COUNTY  |   48.0%   |  50.3%  | 50.0%   55.5%   49.2%   | 47.9%    49.2%

I found each of these DPI scores instructive, so I let them stand on their own. For instance, Obama's DPI isn't necessarily instructive for many Hispanic districts - and I think this is visible in HD143. Obama tends to perform at the low end of the spectrum of scores and local Hispanic candidates typically perform better. For a district like that, looking at Adrian Garcia's DPI score is more helpful. And averaging down that peak score is even more helpful.

What surprised me upon seeing the first Early Vote numbers, was how close the numbers were in HD137 and HD147. In the case of 137, there's a small sample size of only 28 precincts. As mentioned before, I ignored the DPI number all through EV and chose to focus on the individual scores, which had our district between 57-59%. Hubert Vo's HD149 is instructive for how crossover votes aren't accounted for in a DPI model like this. That's always going to be an X-factor unless you have a very well-developed ID program with the data fed into the voter database and tracked closely. Most State Rep campaigns won't have the manpower for that, so you're really left to play that by ear in the real world.

All in all, I continue to be surprised by how accurate this approach is. If you take a look at what Robert Miller posted on E-Day, you'll see what has traditionally been offered. This is basically showing the count of "Hard/Soft Dems", "Hard/Soft Republicans" and a big "unknown" category mistakenly identified as "Independent". I've heard those counts offered year in and year out - usually as proof that Dems were going to win big right before they didn't.

I'd suggest to everyone that there is a significantly better way to measure Early Voting without getting that 20-30% unknown factor. This is the second cycle that I've used this approach and it's been fairly good. There's certainly some artistic license that goes with it to determine what kind of DPI measurement is best for a given race. In 2010, we were staring at wildly optimistic DPI's from vendors assuming that 2006 and 2008 were a good basis of comparison. The 2010 results I posted were based on the DPI scores provided by NCEC, which is the standard place a lot of campaigns go for that information. But I'd suggest that if you have a good enough research ability on your team, you can and should know your own turf better than a national political shop.

Anyways, that's what I saw going on during Early Voting. I'd hope that results like this could be posted publicly at some point in the future. But campaigns are fairly selfish entities, so I'm not sure how likely that is.


2006-10 Citizen Voting Age Population Update

Time for some new data from the Census Bureau. As stated a couple of dozen times before, the counts for citizen, voting age population (CVAP) are no rolled out on an annual basis as part of the Bureau's American Community Survey. It's been a little while since this came out, but I seem to be stuck in work mode for a couple of clients waiting on a district to be finalized and approved for running in. Priorities and whatnot.

Anyways, here's the Harris County view, with the 2005-09 CVAP counts left in and the 2006-10 CVAP counts tacked on for easy comparison.

          Total Pop. (%)     18+ Pop. (%)       CVAP-09 (%)        CVAP-10 (%)
TOTAL     4,092,459          2,944,624          2,195,535          2,230,550
Anglo     1,349,646 (33.0%)  1,085,630 (36.9%)  1,090,624 (49.7%)  1,051,265 (47.1%)
Hispanic  1,671,540 (40.8%)  1,082,570 (36.7%)    494,695 (22.5%)    530,490 (23.8%)
Afr.-Am.    754,258 (18.4%)    541,108 (18.4%)    481,492 (21.9%)    506,150 (22.7%)
Asian       249,853  (6.1%)    194,956  (6.6%)    106,547  (4.9%)    120,660  (5.4%)
Other        67,162  (1.6%)     40,360  (1.4%)     22,177  (1.0%)     21,985  (1.0%)

If those seem like huge changes for one year on the two CVAP counts, there are a few things worth keeping in mind:

1. Some of this is obviously affected by natural demographic changes from the five-year period initially calculated to the new five-year period. I'm not inclined to accept that the majority of the change from the 2009 to 2010 CVAP counts is a result of true population changes.

2. An issue noted from the 2005-09 data is still relevant to keep in mind: there are datapoints from as far back as 2006 incorporated into the estimates. This post gives a fair snapshot of it. Basically, what CVAP captures something of a midpoint of the change from the 2000 Census numbers to the 2010 Census numbers. I doubt this has a huge impact on the year-to-year changes, outside of losing the 2005 datapoints in the 2005-09 CVAP data. But it's definitely something worth checking before anyone gets too excited or too depressed over any particular change for any given column or row above.

3. The Census Bureau itself does a little bit of updating in how they calculate these estimates, so there's bound to be a little bit of correction built into these numbers. I'm not sure how much of an impact this has on year-to-year changes. But the Bureau's reporting of CVAP data has been an issue even beyond redistricting. I've not read any updates on specific changes, but I think it's worth chalking up a not-insignificant share of the change to changing methodology.

With that, here's the map of Harris County by Block Group, color-coded to reflect which demographic group has a majority within the block group. Standard coloring applies: red is for Anglo majority; black is for African-American majority; brown is for Hispanic majority; green is for Asian majority (this is actually a fairly new wrinkle for those keeping track at home). Yellow is for no majority, aka - multicultural.

Numbers and whatnot are included in the info window for those who want to poke, zoom, and click. Knock yourself out.

full pageGoogle Earth file for all of Southeast Texas

There's definitely some interesting finds here. One really nice change from last year is that the data is collected with 2010 block group boundary definitions instead of those from 2000. That might not mean much for those just using a visual overview of the map below. But the change makes it easy to stack this data up against 2010 Census data.

A cursory look at some CVAP Conversion ratios shows that 52.5% of adult Hispanics in Harris County are citizen. For Asians, the countywide ratio is 64.4%. Anglos and African-Americans are 96.6% and 96.2% respectively. That's taken straight from the ACS survey data's count of 18+ and CVAP. Interestingly, if you look at the combined Census Tract 4214 in Gulfton (bordered by Hillcroft, Gulfton, Renwick, and Bellaire), the 2010 Census counts 6,718 18+ Hispanics and the ACS counts 1,180 Citizen and Voting Age. That's a conversion rate of 17.6%. Welcome to Gulton, ya'll.

If you want to look around more of the data for Harris County and see some side-by-side comparisons, the combo map page is updated with the new map. As noted, the Google Earth file includes not just Harris County, but also Fort Bend, Brazoria, Galveston, Chambers, Jefferson, and Montgomery counties. It's fun for the entire family.


Denver’s Demography and the Case of the Anglo Dems

In my earliest iteration of local political map-making (2007 to be precise), I came up with 5-group definition of Harris County political groups. These are voter targets that everyone tends to acknowledge and agree on, but I sought to define the geographic boundaries of where the core of these voters were.

Traditional African-American neighborhoods are obviously the most solid pool of Democratic leaning voters. Traditional Hispanic neighborhood are another that lean toward Democratic candidates. And among Anglos, there tends to be a small carve-out of what I classify as "Anglo Dem", leaving the rest as "Anglo GOP". The reason I felt compelled to spend some time on the project in 2007, however, was because our shop's experience in southwest Houston led us to believe that the "no majority" or "multicultural" areas in the county were growing in importance. So I went through and defined a "Multicultural" core where there was generally no majority present and the votes tended to reflect that.

Now, I hadn't worked with any extensive GIS software prior to then and the Census data was barely relevant late in the decade. So the whole exercise took a bit of creativity, some heavy abuse of Photoshop to create the rough drafts of maps, and some generosity of a local engineering firm's GIS resources to help us add some more layers to the research. But, all in all, that exercise was the launch pad of a great deal of what you see in the way of the more recent cartography. The results, based on the 2006 election cycle and a ton of guesstimating based on HCAD queries looked like this ...

Briefly, the biggest things to jump out of this were the following:

- The notion that "Multicultural" parts of the county played a big role was certainly substantiated by seeing 20% of the vote come from areas where nobody held a majority.

- The share of vote to come out of Traditional Afr-Am and Traditional Hispanic areas was not sufficient enough to warrant the typical approach of doing field work in those areas and considering the job "done" for Hispanic or Afr-Am targeting and outreach.

- A closer look at the Hispanic numbers indicated that there is a similar amount of Hispanic vote to be had in the Traditional Hispanic precincts was there was in either the Multicultural OR the Anglo GOP area. While the first comparison might not have served as much of a surprise, the latter one was very attention-getting. It was also something that I think was borne out in some of the results for Adrian Garcia in 2008.

- In looking at historical election returns for each of these areas, the Anglo Dem cluster proved rather fickle, actually voting GOP in the most competitive contest we looked at in that year. The Multicultural cluster also voted slightly in favor of the GOP that year, though part of that was because what was "multicultural" or "transitional" in 2006/07 wasn't necessarily the case in 1994. But for a "base" area such as the Anglo Dems, the swing was more noticeable.

As a first order of business, I probably should go through and see how this same Anglo Dem cluster performed in 2010 with regard to Bill White and any other downballot races. It wouldn't surprise me at all to see a similar swing to the GOP below Bill White's showing. While I stand by my working theory that 2010 was a turnout phenomenon rather than a "mind-changing" one, I think it stands to reason that the canary in the proverbial coal mine is to see a re-coloration of the map in Anglo Dem areas.

While the rise of multicultural areas in Harris/Ft Bend counties has been through the roof (feel free to compare the 1980-2010 maps), the definition of Anglo Dem clusters is worth exploring in a bit more detail. This is basically another thread I'm pulling from the sweater of Friday's Lanier Public Policy Conference. Ron Brownstein's point was that successful Democratic constituencies of the future are going to have to be created from minority voters and those Anglos comfortable with diversity. Unfortunately, Houston's Anglo Dem area seems relatively small to draw too many conclusions with.

This is what sent me to Denver. While there are some notable differences between the two states that aren't good comparisons - namely Colorado's lack of racial conflict compared to southern states - the pool is big enough in Denver to start milling about and exploring how big the difference is once you account for some of the more traditional reasons that influence Democratic support among Anglo voters.

For starters, there's the Jewish community and the GLBT community that tend to be supportive. In most major urban areas, there should be a common parallel even if it's hard to derive good, solid numbers for each. A bit easier to get good data on is the presence of multi-degreed individuals. College towns have young students who may vote more liberally than the adults in town (ie - Austin). More the case outside of Houston, there are also the occasional union hotbeds. I'm not sure how much of a presence unions have in the Denver area, but I'm starting off with the assumption that it's on the low side compared to midwestern metros. Beyond those factors, it starts to get fairly thin. So the hope is that by spending some time in Denver, there's enough of a pool of Anglo Dems to do some more digging.

First up, the raw demographics of the area. I look at the Denver area two ways: Denver county (it's actually a city-county, so I'm just going to call it Denver and leave it at that) and the four-county area that includes the three other counties that surround Denver. In each case, they are all majority-Anglo, with Denver at a slim Anglo majority on it's own. The math for the aggregate and just Denver is as follows:

The Denver Four   
           TOTAL         ANGLO            HISPANIC         AFR-AM        ASIAN
Total  .. 2,148,307  1,336,889 (62.2%) 540,810 (25.2%) 131,253 ( 6.1%) 77,633 (3.6%)
18+  .... 1,627,004  1,091,306 (67.1%) 344,649 (21.2%)  95,461 ( 5.9%) 59,158 (3.6%)
CVAP .... 1,392,465  1,045,065 (75.1%) 204,347 (14.7%)  79,995 ( 5.7%) 34,983 (2.5%)

Denver County
           TOTAL         ANGLO            HISPANIC         AFR-AM        ASIAN
Total ..   600,158    313,012 (52.2%) 190,965 (31.8%)  58,388 ( 9.7%) 19,925 (3.3%)
18+ ....   471,392    274,874 (58.3%) 125,111 (26.5%)  43,709 ( 9.3%) 15,941 (3.4%)
CVAP ...   384,850    250,265 (65.0%)  78,092 (20.3%)  38,843 (10.1%)  9,096 (2.4%)

The pattern of Hispanic population shares going south as you go from Total Population to CVAP is certainly familiar. And, to me, it also raises doubt on whether there's really a "sleeping giant" there to be awoken.

What's interesting is that in each individual county, Michael Bennett carried the total vote for the 2010 US Senate contest and John Hickenlooper did the same in the 2010 Governor's race. Each won statewide as well. It was as if 2010 just didn't happen in those contests, although the GOP did perform better for downballot state contests. Hickenlooper's contest was an oddity in that Tom Tancredo ran as a third-party candidate, which had the effect of taking votes away from the GOP nominee and making the race more hyperpartisan. The impact in terms of which precincts went blue vs red (or whatever color for a third party) did not change terribly much on the scale that I'm used to working with. That's not to say that the percentages in each precinct were close to identical. But merely to point out that deep red areas didn't suddenly swing toward Hickenlooper. The map basically stayed the same. Going back to my point about wanting to see some amount of re-coloration before concluding that minds are changing on the ground, it just didn't look like it was on display here.

One of the post-election comments heard about both Colorado and Nevada was that it was Hispanic voters that "saved" Bennett and Harry Reid. In some preliminary views of the returns, what struck me was that while it was impressive that the share of vote among Hispanics did not tail off like a lot of other midterms do, the showing in Anglo-majority areas was impressive. To me, all that says collectively is that each campaign had impressive field teams. This is even more impressive in Reid's case considering how far behind he was at the start of the cycle. Likewise, that the field teams placed enough importance on multiple segments of voters is my starting point. And in an election where Anglo voters should have proven more fickle (and hence, more Republican), the showing among Anglo Dems strikes me as more impressive.

Furthermore, the sheer number of Hispanic voters in each state, based on some assumptions about how CVAP translates to SSVR, isn't substantive enough to suggest that a "sleeping giant" has been awaken. Those voters are certainly a key component to a successful Democratic coaltion, but the biggest question I see on the board is explaining how the Anglo Dem areas held up while they usually start breaking out some red in pendulum swing elections like 1994 or 2010.

And just to start coloring in the lines, here's an overview of the Denver area itself. First with demographic majorities shaded in. As always, click 'em to big 'em ...

Total Population

Voting Age Population

Citizen Voting Age Population

And then for the 2010 Senate election ...

2010 Senate Election Results

There should be some obvious and fairly large Anglo Dem areas that jump out when you go through these. If you want one particular highlight, there's this ...

SE Denver Anglo Dem Cluster
2010 Senate results in Majority-Anglo region

This area of Denver is majority-Anglo at each population count: total population, voting age population and CVAP. And the voting precincts are solidly in favor of Michael Bennett in the US Senate race. Over the remainder of the week - possibly more - I'll be diving into these areas of Denver to see what parallels there are to Houston and what makes Denver unique. In the spirit of crowdsourcing the research, if there's anything you know first-hand about Denver that might help along these lines, feel free to drop a comment or an email my way. Feel free to download the Google Earth file for all of this and flip through the layers of it for everything shown here, plus the 2010 Governor's election (with yellow for Third Party sum > Dem or Third Party sum > GOP results).


Three More Views of Harris County Demographics

Three visualizations of Harris County demographics that pivot from the broadest, overall view to one that reflects more of the political reality. The first is a repeat of the map I ran yesterday.

Total Population:

18+ Population:

Citizen Voting Age Population:

If you want to dig into the details, here's the link for a side-by-side map view that lets you choose which two maps you want to compare. The default is Total Population (left) and CVAP (right). For good measure, I've also got a version of the side-by-sides for Fort Bend County, also. By all means, poke around.

The grand total numbers, so you can see the aggregated total is as follows:

            Total Pop. (%)      18+ Pop. (%)            CVAP (%)
TOTAL       4,092,459            2,944,624             2,195,535
Anglo       1,349,646 (33.0%)    1,085,630 (36.9%)     1,090,624 (49.7%)
Hispanic    1,671,540 (40.8%)    1,082,570 (36.7%)       494,695 (22.5%)
Afr.-Am.      754,258 (18.4%)      541,108 (18.4%)       481,492 (21.9%)
Asian         249,853  (6.1%)      194,956  (6.6%)       106,547  (4.9%)
Other          67,162  (1.6%)       40,360  (1.4%)        22,177  (1.0%)


The CVAP numbers are going to get very interesting this time around. In the last Census, citizenship was asked on the regular Census form - the one that got asked of 100% of people (give or take) one in six Census surveys. This time around, the question was not on the Census form, but it was on the Census bureau's American Community Survey questionnaire. That is essentially a 2.5% sample that's used to get a number that also has a +/- margin of error. So that's how you get the statistical anomaly of Anglos in Harris County gaining nearly 5,000 people after backing out citizens from the 18+ universe. In areas that are undergoing significant demographic change, the measurement also uses datapoints from the middle of the decade. As we saw in the Pulaski County, Arkansas example ... it understates the current reflection of demographic change.

The bad news is that CVAP data is required as part of the Voting Rights Act. As an example of it's uses, it is designed to show that a majority-Hispanic district does not lose enough voting strength for Hispanics to elect a candidate of their choosing. That's the point that commenter Mainstream has been berating me for since delving into demographics and redistricting.

I'm not yet to the point of looking for more info on how the Census vs ACS matter might be resolved, but it's pretty clear that other people have spotted this glitch and I can't imagine that the Justice Dep't will be completely flat-footed on it. In September of last year, David Hanna of the Texas Legislative Council noted some of the issues that may come up as a result of it with the Senate Redistricting Committee. So it's at least on the radar, along with all of the other redistricting laws, rules, customs, and guidelines that sometimes conflict.

As a sidenote and interesting algebra crunch from the data above, here's what the Under-18 population looks like in Harris County:

            >18 Pop. (%)
TOTAL       1,147,835  
Anglo        264,016 (23.0%)
Hispanic     588,970 (51.3%)
Afr.-Am.     213,150 (18.6%)
Asian         54,897  (4.8%)
Other         26,802  (2.3%)


Meet your future, folks. Those younguns are more likely to be voter-eligible when they grow up.


Fatal error: Call to undefined function xwp_footerx() in /nfs/c01/h12/mnt/4748/domains/ on line 18