If one looks at the latest poll out of Texas, among registered voters the outcome is in line with what previous pollsters have found during this cycle: a lead for Mitt Romney that is quite a bit less than the margins enjoyed by John McCain in 2008 (7-8 points).
However, if a "likely voter" screen is added, the results get dramatically different. What makes these polls intriguing (and, some would argue, suspect) is how it became different. The Texas poll (done by the University of Texas for the Texas Tribune) rockets Romney out to a 20-point advantage. In other words, there is a double-digit gap between what the poll found among "registered" voters and what they found among "likely voters."
Based on previous Texas polling this cycle, and 2008 electoral performance, the "likely voter" poll seems infinitely less plausible than the survey of registered voters. Which raises again the debate of what constitutes a likely voter.
But, first, here are the numbers to kick off the week:
PRESIDENTIAL GENERAL ELECTION TRIAL HEATS:
NATIONAL (Gallup Tracking): Obama tied with Romney (46-46)
NATIONAL (Rasmussen Tracking): Obama d. Romney (47-44)
OKLAHOMA (Sooner Poll): Romney d. Obama (62-27)
TENNESSEE (Vanderbilt University): Romney d. Obama (47-40)
TEXAS (Univ. of Texas/Texas Tribune): Romney d. Obama (46-38)
DOWNBALLOT POLLING:
TX-SEN (Univ. of Texas/Texas Tribune): Republican Candidate 45, Democratic Candidate 34
TX-SEN--D (Univ. of Texas/Texas Tribune): Paul Sadler 35, Addie Dainell Allen 22, Sean Hubbard 22, Grady Yarbrough 12
TX-SEN--R (Burnt Orange Report): David Dewhurst 43, Ted Cruz 30, Tom Leppert 14, Craig James 5, Lela Pittenger 4, Others 3
TX-SEN--R (Univ. of Texas/Texas Tribune): David Dewhurst 40, Ted Cruz 28, Tom Leppert 15, Craig James 5, Others 5, Lela Pittenger 3
A few thoughts, as always, await you just past the jump ...
- A quick glance at the Texas Tribune write-up on their own presidential and Senate poll gives a quick tip off to how their likely voter screen so dramatically altered the results of their survey:
And "likely voters" were defined as those who indicated they were "somewhat" or "extremely" interested in politics and who voted in "every" or "almost every" election in recent years.
And which elections were they asking about? From the survey instrument itself:
There are many elections in the state of Texas. Furthermore, many people intend to vote in a given election, but sometimes personal and professional circumstances keep them from the polls. Thinking back over the past two or three years, would you say that you voted in all elections, almost all, about half, one or two, or none at all?
"Every" or "almost every" election? But not dating back to the last presidential election? Really?!
So, by the self-identification of the respondent, if I miss most city council or county commissioner races (which would be, if turnout stats are accurate, most voters), I would be automatically disqualified from being a "likely voter"? For a presidential election? That has to be among the most odd sets of criteria I have seen for determining a likely voter. How about asking about participation in particular elections, like 2004 and 2008? Or basing it solely off of interest level (which is a common question to determine voter intensity)? Having to participate in "almost every election" without acknowledging that voters hold different elections at different priority levels seems completely bizarre. And, I suspect, it'd go a long way toward explaining how a race that was just a single-digit race among registered voters ballooned to a 20-point race among so-called likely voters. Worth noting: Because of my issues with that screen, I used the registered voter numbers for the polling data above the fold, which would explain any discrepancies you see between these numbers and those posted in other media outlets.
- In other polling today, we see that the prospect of runoffs in both ends of the U.S. Senate primary in Texas becomes likely, given that both the Burnt Orange Report and the UT/Texas Tribune polls show Lt. Gov. David Dewhurst several notches below the 50 percent threshold he will need to avoid a runoff. Meanwhile, the prospect of a runoff on the Democratic side seems even more likely, with no candidate over 30 percent of the vote.
- That Tennessee poll is getting a great deal of attention, meanwhile. However, it isn't necessarily a shocker: There have been multiple polls in the Volunteer State showing the president within single digits of Mitt Romney during this cycle. This one, however, is the first one to suggest that some developing animosity between the newly minted GOP state government in Tennessee and the electorate might be contributing to that movement away from the GOP standard bearer in a state the party carried with ease in 2008. Meanwhile, as the primary results there made clear at the time, the president is unloved in Oklahoma. But seeing how that was among his two worst states in the country in 2008, that should come as a surprise to precisely no one.