A University of Texas/Texas Tribune Poll, released October 25, shows these results for Governor: Republican Rick Perry 50%, Democrat Bill White 40%, Libertarian Kathie Glass 8%, Green Party nominee Deb Shafto 2%. There are no undecideds, because if a respondent first said “Undecided”, the pollster then encouraged the respondent to say which way he or she is leaning.
The poll has the results of all the partisan executive statewide posts, showing Libertarians in some statewide races at 12%, even though all these races have a Democrat and a Republican in the race as well. In the one executive statewide race with no Democrat running, the Green Party nominee is at 9%. Thanks to Taegan Goddard’s Political Wire for the link.
This is a poll of registered voters, and 7% of respondents said they “did not vote” in the 2008 presidential election. In fact, 40% of Texas registered voters did not vote in that election.
I guess this poll has a very non-representative sample, and/or respondents are lying.
I hope it was the Perry and White respondents who were lying about not voting. Hooray for Kathie Glass!
What’s amazing here, and predictably so, un-reported, is that the Libertarian and Republican (coalition) vote is a whopping 58%!
Perry, who once touted seccession, and is an extreme fiscal conservative, leans wholly libertarian. Kathie Glass is a conservative-leaning Libertarian. The two of them are not that far apart.
And they are stomping the socialist Bill White. That’s the much bigger story here, but of course, it’s not in the interests of Libertarian Party partisans, nor the liberal media to report it that way, so it gets ignored.
The UT Poll in an internet poll conducted for them by Yougov/Polimetrix. Yougov uses a panel of internet users who have volunteered to be polled on political and consumer issues. Yougov then does a demographic match to the population at large, to choose their sample for a particular poll.
There is likely to be rather extreme selection bias when your sample is restricted to those with internet access – and use it enough to volunteer to be polled, who have an affirmative interest in being polled, and who are interested enough in a particular poll to complete it after they are selected. This is likely to over-represent Libertarian voters.