Though I didn’t have one originally, I now have a canvass of the GOP primary in Harris County from this year. Since it’s impossible for me to have data and not do something with it, I’ve taken a look at the Presidential and Senate results for each of the State Rep districts. Here are the Presidential numbers:
Dist Romney Paul Santorum Other Romney %
=============================================
126 7,448 1,189 805 993 71.38%
127 9,292 1,576 1,037 1,398 69.85%
128 6,817 1,445 823 1,144 66.64%
129 8,654 1,904 946 1,397 67.08%
130 8,542 1,741 1,149 1,268 67.26%
131 1,130 226 122 210 66.94%
132 4,557 1,129 716 745 63.76%
133 15,217 1,973 947 1,715 76.65%
134 10,802 1,868 606 1,303 74.09%
135 4,757 1,045 656 758 65.92%
137 1,728 358 147 254 69.48%
138 5,898 1,153 676 978 67.75%
139 2,006 459 252 322 66.01%
140 806 178 103 124 66.56%
141 657 153 75 115 65.70%
142 1,110 234 154 183 66.03%
143 1,227 285 158 216 65.06%
144 1,811 495 214 319 63.79%
145 1,718 428 202 324 64.30%
146 2,160 474 187 373 67.63%
147 1,334 457 141 270 60.58%
148 2,389 733 257 413 63.00%
149 3,153 576 335 498 69.11%
150 8,167 1,453 1,085 1,225 68.46%
Romney got 69.07% overall in Harris County, so you can see where he exceeded and lagged that total. I’ll leave it to those who are better versed in GOP geography to speculate why he did better in some places than in others. Romney’s 16 point range is as wide as Obama’s was on the Democratic side, and a close mirror of the bulk of his countywide numbers. In other words, nothing too unusual here.
Here’s the Senate data:
Dist Dewhurst Cruz Leppert James Other Dew % Cruz %
==============================================================
126 4,539 4,713 624 214 247 43.91% 45.59%
127 5,998 5,917 686 251 258 45.75% 45.13%
128 5,099 3,858 583 275 239 50.72% 38.37%
129 5,840 5,381 745 263 395 46.26% 42.63%
130 5,133 6,030 657 270 453 40.92% 48.07%
131 779 664 94 45 73 47.07% 40.12%
132 2,952 3,313 423 158 206 41.86% 46.98%
133 9,817 8,105 1,068 381 289 49.93% 41.23%
134 7,107 5,885 899 201 73 50.17% 41.55%
135 2,879 3,531 412 153 176 40.26% 49.38%
137 1,233 927 147 38 108 50.26% 37.79%
138 3,846 3,813 553 201 206 44.62% 44.24%
139 1,327 1,406 160 48 68 44.10% 46.73%
140 623 427 74 33 36 52.22% 35.79%
141 458 407 71 26 31 46.12% 40.99%
142 691 793 86 42 46 41.68% 47.83%
143 947 716 86 44 71 50.80% 38.41%
144 1,460 1,009 173 80 93 51.87% 35.84%
145 1,228 1,038 201 64 97 46.73% 39.50%
146 1,512 1,248 214 63 92 48.32% 39.88%
147 916 954 137 49 71 43.07% 44.85%
148 1,607 1,699 212 102 105 43.14% 45.61%
149 2,000 1,944 307 114 126 44.53% 43.29%
150 5,008 5,543 667 229 331 42.52% 47.06%
Dewhurst carried the county by a bit less than 2.5 percentage points, and won 15 of the 24 State Rep districts, including a clear majority in five. Cruz had no majorities, but after being creamed in absentee ballots and nipped in early in-person voting, he cruised to a ten-point win on Election Day, falling just short of 50% on May 29 voting. Draw your own conclusions about what that may mean for the runoff. The other point of interest, as I brought up when I looked at county results, was whether Cruz would have any advantage in Latino-majority areas. For Harris County, the answer is a clear No, though I suspect there weren’t that many actual Latino voters casting those ballots. Again, I’ll defer to Mainstream or any other actual Republicans who may have a theory to propose.
I thought about looking at the countywide races – DA, Tax Assessor, County Attorney – but didn’t get around to it. Honestly, when a race is won by 20 or more points, precinct data is unlikely to tell you anything interesting. I may come back to these races later, but for now you can presume that Mike Anderson, Mike Sullivan, and Robert Talton did pretty well everywhere. I’ll have some Democratic countywide results next. Let me know if you have any questions about this.
First, I note that Romney, like Obama, seems to do a little worse in blue collar, working class districts. Bubba is up for grabs in November, or perhaps he is just turned off of politics entirely this year, casting protest votes.
I am not sure whether Cruz is getting a slight bump from his Hispanic surname or not in the GOP primary. I looked at precinct level data, but like you, am unsure how many GOP primary participants are Hispanic, and whether their distribution mirrors general demographics of the precinct or not. It seems plausible to me that there might be plenty of Hispanic surnamed folks in River Oaks voting GOP, while in Denver Harbor the typical GOP voter may be an elderly Anglo. I also could imagine that for some Mexican-American GOP voters, Cruz as a Cuban, Ivy League attorney lacks authenticity.
I noticed for example that Cruz carried several precincts in Montrose, Midtown, Heights, Woodland Heights, Binz, (all fairly small turnouts) while Dewhurst carried wealthier areas, like the Hermann Park high rises, River Oaks, Tanglewood. Cruz carried Timbergrove, while Dewhurst carried Lazybrook. Dewhurst won in Pasadena, Cruz in Cypress. If I get some time, I will try to analyze these data further. I am wondering if there is a generational divide, with younger voters siding with Cruz.
I will be surprised if Dewhurst has the same advantage in mail in ballots in the runoff.
11 of the 206 Republican election day voters in Timbergrove near the Heights were Hispanic surnamed. This 5% compares to maybe 12% Spanish surnamed voter registration for the entire list and maybe 30% Hispanic total population for the whole district.