5 takeaways from the new UT/TT Poll

Photo by Marjorie Kamys Cotera

The latest University of Texas/Texas Tribune Poll, released last week, offers a look at the latest clues about the contours of the electorate going into the final days of the election. Data collection was completed Oct. 19, before early voting started and a little more than two weeks before Election Day. Here are five initial observations based on a first reading of the data.

1. Men and women seem to be moving in opposite directions. Following the release of a number of gubernatorial polls in recent weeks, much attention has been paid to Democrat Wendy Davis' performance among women. The October 2014 UT/TT results show signs of progress for Democrats compared with 2010: Attorney General Greg Abbott, the Republican nominee, leads Davis among women by only 2 percentage points. Some might wonder: only? It's important to note that women aren’t a homogenous bloc of voters, and while they tend to be more Democratic than men, this is still Texas, where women tend to vote Republican in significant numbers. Case in point, and good context for that 2-point gap: In 2010, Republican Rick Perry beat Democrat Bill While among women by 8 points, according to exit polling.

Texas men appear to be moving in the other direction. In 2010, Perry bested White among men by 17 percentage points. Among likely male voters in our October 2014 survey, Abbott is leading Davis by 29 points — a large gap for a very large portion of the electorate. The overall attitudes of men and women toward the candidates also reflect theses patterns: While 49 percent of likely female voters view Davis favorably (44 percent view her unfavorably), 58 percent of likely male voters view her negatively (only 31 percent have a favorable view). 

2. The largest racial group in the state is still wedded to the Republican Party. Abbott leads Davis among white voters by 31 points. This gap persists in both the lieutenant governor's race as well as in the attorney general's race. As much as Davis should be happy with her seeming improvement among women compared with Democrats’ performance in 2010, white voters will likely make up between 60 and 70 percent of the 2014 electorate. Hispanic growth in Texas will translate into a less Anglo electorate, but gradually. In the short or medium term, it's hard to imagine a competitive Democratic Party in Texas that’s this far behind Republicans among Anglos.

3. The most compelling but hardest-to-answer questions still center on the Hispanic vote. A number of issues may explain how close the race is between Abbott and Davis among Hispanics in the October poll (48 percent for Davis, 46 percent for Abbott). We’ll explore that in a later piece. But in short, there have been indications throughout the campaign — like Davis’ poor performance in the Democratic primary in several South Texas counties and Abbott’s investment in Latino outreach — that Abbott may prove more competitive among Hispanic voters than expected. It's unlikely that the Texas GOP, with its own heated internal disagreements about both policy and rhetoric on immigration, is poised to become the party of the majority of Hispanics. But it's also unlikely that Texas Democrats can assume a default 60 percent-plus share of the Hispanic vote just by showing up.

It's worth noting that the UT/TT results, as well as the variation in the data on Hispanics in other public polling, have large margins of error because they’re sub-samples of the overall sample. Depending on what assumptions are made about the size of the Hispanic electorate on Election Day, the margin of error for a Hispanic vote estimate in the UT/TT poll is about 8.5 percentage points — largely comparable to other statewide polls. In other words, Abbott may have 46 percent of the Hispanic vote, but he also could have 38 percent — the exact same share of the Hispanic vote that Perry received in 2010.

4. For all the fascination about it among insiders, the lieutenant governor's race hasn't exactly captured the public's imagination. Texans’ views of state Sen. Dan Patrick, the Republican nominee for lieutenant governor, have been static for months, as have views of Leticia Van de Putte, his Democratic opponent. For Patrick, this means that he's still highly regarded among Republicans and that his relatively low profile since winning the incendiary GOP primary has served him well. For Van de Putte, enthusiasm among Democratic loyalists but low name identification among the general electorate has remained a defining feature of her candidacy, with roughly half of registered voters in Texas still unable to offer an opinion about her.

5. Voters flirt with Libertarians in polls, but they don't marry them on Election Day. The Libertarian candidates did pretty well in the statewide races on our trial ballot, polling in the range of 6 to 11 percent. Does that portend a Libertarian earthquake in Texas on Election Day? Probably not. Respondents on surveys, especially when the elections don't appear to be terribly competitive (for example, when one party is leading the other in every statewide race by 10 or more points), may feel the freedom to make a statement by selecting a third-party candidate. But by Election Day, a large share of the Republicans who say they’ll vote for a Libertarian end up voting Republican. This happens for a variety of reasons: People don't like to feel like they're "throwing away" their vote; they want to vote for the "winning" candidate; it's just so easy to vote straight ticket after waiting in line for a while.

The current results also look like past elections. In the October 2010 UT/TT Poll, Libertarian gubernatorial nominee Kathie Glass got 8 percent of the vote — but ended up with about 2 percent on Election Day. Glass, who’s running again this year, received 6 percent in the October 2014 poll, but it's unlikely that she’ll perform that well on Nov. 4. 

Disclosure: The University of Texas at Austin is a corporate sponsor of The Texas Tribune. A complete list of Tribune donors and sponsors can be viewed here.

James Henson

Texas Tribune pollster and director of the Texas Politics Project at the University of Texas at Austin


Joshua Blank

Research director, Texas Politics Project, University of Texas at Austin