Folks are seizing upon this poll from the University of Houston's Hobby School of Public Affairs as an indicator that the presidential race in Texas will be much closer than anticipated. Some even believe that Clinton will make up the difference in the next three weeks. The numbers may be correct, but if you've been following other polls, there's plenty of reason to doubt it.
In fact, there are reasons to doubt whether many of the poll results reported are accurate at all.
In fact, there are reasons to doubt whether many of the poll results reported are accurate at all.
The poll shows Donald Trump leading Hillary Clinton by just 3 percentage points in a four-way race that includes Gary Johnson and Jill Stein. It comes on the heels of the WFAA (Fort Worth) poll last week putting Trump's lead at 4 points among likely voters in Texas. Since 2000, no Democratic nominee has come within 11 points of winning Texas, and no Democrat has won the state since Jimmy Carter in 1976. Bill Clinton came closer in his two elections mostly because Ross Perot was in the race.
However, a more interesting story lurks deeper in the results.
Undecided, Unfavorable, and Unusual
Let's begin with the figures for undecided voters. It's understandable that, this close to Election Day, 29% of self-identified Independents are saying "IDK." But it's bizarre that 11% of the Democrats polled are undecided at this late date, and 8% of Republicans. The equivalent numbers in the latest Quinnipiac University poll are 6%, 2%, and 4%, respectively, which are not far off from other national surveys. Even the WFAA poll found about 5% undecided.
The percentages of respondents with very unfavorable views of both Trump and Clinton are pretty astounding: 48% and 47%, respectively. This likely indicates that about half of those choosing either of those two candidates are doing so primarily because of their dislike for the other candidate.
Then look at who respondents think will be elected: A majority say Clinton will. Fewer than half the Republicans believe Trump will win. That's quite unusual.
Unidentified Voting Objects
The crosstab that has me seething with impotent rage has little to do with presidential preferences. It seems that a sizable majority of voters don't know the correct answer to the question of whether a photographic ID is required in order to vote. De facto, the correct answer is, "It depends." It depends on how well your precinct judges have been trained, and whether they want to make life difficult for young voters or voters of color. According to the Harris County Clerk's Office and the Texas Secretary of State's Office, the answer is Yes, at least for this election.
The folks at the Hobby School answered their own question wrong. One can hardly blame them for that, however, because various media outlets and self-appointed experts have been propagating waves of misinformation since the Fifth Circuit's decision on Texas's Voter ID law.
If you have one of the valid identifications, take it with you when you vote. If you forget, do not accept a provisional ballot, because you'll be lucky if that ever gets counted. And, as tempting as it may be, don't try to vote naked à la Katy Perry.
However, a more interesting story lurks deeper in the results.
Undecided, Unfavorable, and Unusual
Let's begin with the figures for undecided voters. It's understandable that, this close to Election Day, 29% of self-identified Independents are saying "IDK." But it's bizarre that 11% of the Democrats polled are undecided at this late date, and 8% of Republicans. The equivalent numbers in the latest Quinnipiac University poll are 6%, 2%, and 4%, respectively, which are not far off from other national surveys. Even the WFAA poll found about 5% undecided.
The percentages of respondents with very unfavorable views of both Trump and Clinton are pretty astounding: 48% and 47%, respectively. This likely indicates that about half of those choosing either of those two candidates are doing so primarily because of their dislike for the other candidate.
Then look at who respondents think will be elected: A majority say Clinton will. Fewer than half the Republicans believe Trump will win. That's quite unusual.
Unidentified Voting Objects
The crosstab that has me seething with impotent rage has little to do with presidential preferences. It seems that a sizable majority of voters don't know the correct answer to the question of whether a photographic ID is required in order to vote. De facto, the correct answer is, "It depends." It depends on how well your precinct judges have been trained, and whether they want to make life difficult for young voters or voters of color. According to the Harris County Clerk's Office and the Texas Secretary of State's Office, the answer is Yes, at least for this election.
The folks at the Hobby School answered their own question wrong. One can hardly blame them for that, however, because various media outlets and self-appointed experts have been propagating waves of misinformation since the Fifth Circuit's decision on Texas's Voter ID law.
If you have one of the valid identifications, take it with you when you vote. If you forget, do not accept a provisional ballot, because you'll be lucky if that ever gets counted. And, as tempting as it may be, don't try to vote naked à la Katy Perry.
RSS Feed