The Texas Politics Project Poll
The Texas Politics project is the work of the UT government department. Its polls, directed by professors Jim Henson and Daron Shaw, will be a quarterly fixture on the Texas political scene. (Evan Smith commented on the poll Thursday in his “State of Mine” blog.) This is an Internet poll; Henson believes that telephone polling is no longer reliable due to caller ID, which allows targeted households to refuse calls, and the preference of young voters for cell phones. I will publish the methodology at the end of this discussion. Here is what I found most interesting in the poll: 1a. Party identification (unscreened) Strong D 25% Not very strong D 14% Independent leaning D 7% Strong R 20% Not very strong R 13% Independent leaning R 9% D 46% R 42% I 8% Other 4% This is eye-opening. The Democrats’ edge is accounted for by their advantage in “strong” adherents. While it is hard to conceive of Texas as a blue-trending state, there is evidence to support it: the tarnishing of the Republican brand in the suburbs (see More Numbers: R’s Lose but D’s Can’t Gain,), and the Democrats’ 2 to 1 edge in primary turnout. Whether this will translate into success at the ballot box is yet to be seen. As we will see from another question in this poll, the ideological makeup of the Democratic party is majority conservative. 1b. Party identification (registered voters) Strong Democrat 24% Not very strong Democrat 12% Independent leaning Democrat 7% Strong Republican 21% Not very strong Republican 14% Independent leaning Republican 10% D 43% R 45% I 7% Other 7% Among all respondents, D’s held a +4 advantage. The restriction to registered voters gives Republicans a small edge. R +2 strikes me as a very credible estimate of where the state is right now. The poll shows Republicans have a five-point advantage (24-19) over Democrats among voters whose party ties are weak (“not very strong” or “independent leaning”). These categories apply to the disillusioned Republicans who are gravitating away from the GOP but not toward the Democrats. This is the most volatile segment of the electorate. 2. Ideology (by party and by the state as a whole) Among strong Democrats, 42% consider themselves liberal. Among strong Republicans, 88% consider themselves conservative. Only 42% of strong Democrats are liberals! This is an amazingly small number. Liberal 20% Conservative 39% Moderate 41% 3. Major Statewide Races President McCain 42.5% Obama 33.3% Barr 4.9% Nader 2.2% Don’t know 17% U.S. Senate Cornyn 44.3% Noriega 30.9% Don’t know 24.9% These numbers are in line with previous polling that showed Cornyn with a double-digit lead. 4. Presidential Race (party affiliation) For all the unhappiness with McCain among Texas Republicans, strong R’s are staying hitched: 90% support McCain, and you need a magifying glass to see the thin strip of blue at the bottom of the bar graph for Obama. Only 82% of strong D’s support Obama. McCain gets around 5% of this group and 17% of not very strong D’s. But McCain gets virtually no support from independent leaning D’s, which suggests that this group may be dominated by young voters who are loyal to Obama. 5. Better off/Worse off than last year Better off 17%, must be the oil and gas industry. Worse off 48% About the same 34% These are pretty gloomy numbers for a state that is leading the nation in creating jobs. 6. Most Important Issue Economy 27% Immigration 25% Energy 17%, there’s the oil and gas industry again 7. Politicians (approve/disapprove) Rick Perry Strongly approve 14% Somewhat approve 30% Somewhat disapprove 15% Strongly disapprove 18% Don’t Know 22% Tom Craddick Strongly approve 2% Somewhat approve 15% Somewhat disapprove 10% Strongly disapprove 10% Don’t know 63% David Dewhurst Strongly approve 7% Somewhat approve 19% Somewhat disapprove 13% Strongly disapprove 4% Don’t know 60% Texas Legislature Strongly approve 6% Somewhat approve 25% Somewhat disapprove 24% Strongly disapprove 14% Don’t know 31% The middle of an interim is not the best time to poll about state officials. These numbers don’t tell us much, except that the public isn’t paying attention at all. More than one in five have no opinion about the longest-serving governor in Texas history. Democrats are going to have a hard time making Craddick an election issue this fall, and Dewhurst’s efforts to get media coverage haven’t been rewarded in name ID. Perry’s positive numbers will surprise some. If you look at the “Most Important Issue” category above, you will notice that the top three issues–economy, immigration, energy–are topics on which Perry has been very active. Think what you like about him, but he is a seasoned professional politician. POLL METHODOLOGY The UT-Austin Texas Politics Poll is designed by researchers at UT-Austin and conducted by YouGovPolimetrix, a firm with demonstrated success in internet polling. YouGovPolimetrix accomplishes internet polling through a unique sampling procedure known as “matched random sampling.” The firm begins with two lists: (1) a list of all consumers in Texas (covering approximately 95 percent of the adult population), and (2) a list of people who have agreed to take YouGovPolimetrix’s surveys. For each list, Polimetrix has an extensive set of demographics. The sampling procedure then progresses in two stages. First, a random sample of consumers is drawn. For each person drawn from this sample a list of key demographics is recorded. In essence, each individual drawn is represented as a cluster of demographic characteristics, including age, income, education, race, gender, longitude and latitude, etc. Second, YouGovPolimetrix uses a matching algorithm to find the PollingPoint panelist who is the closest match to the person drawn off the consumer file. In this way an entire matched random sample is constructed for all people in the sample. The current poll of 800 adult Texans has a margin of error of +/- 3.46 percentage points at the 95% confidence level. The poll includes interviews with 677 registered voters, with an attendant margin of error of +/- 3.77 percentage points. Response rates are almost 100% given the matching methodology. The YouGovPolimetrix pool includes people who are much less likely to have access to the Internet or a personal computer. They have been especially assiduous at enlisting people with lower incomes and ethnic and racial minorities, part of an attempt to bolster the representativeness of their samples. * * * * I hope that readers who have expertise in the mathematics of polling will share their thoughts about this methodology. While I agree with the founders of the Texas Politics project that traditional telephone polling faces a crisis of reliability, I have some qualms about polls that allow people to self-select to be poll participants. It seems to me that this compromises the randomness of the sample. Notwithstanding this concern, this particular poll seems to be in the ball park for Texas politics today.