Another good poll result for Peter Brown.
According to [an 11 News / KUHF Houston Public Radio] poll, 24 percent of likely voters in the Houston mayoral race plan to vote for Brown. This is up sharply from the five percent of likely voters who said they would vote for him in an 11 News poll conducted in August, before Brown had launched a massive advertising campaign and spent more than $2.4 million of his own money to promote his candidacy.
Brown’s nearest challenger is comptroller Annise Parker, who holds the support of 16 percent of likely voters in the current poll. Former city attorney Gene Locke has 14 percent support, and Harris County Department of Education trustee Roy Morales has 5 percent support. Forty-one percent of likely voters told pollsters that they have not yet settled on a candidate. The margin of error is plus or minus 4.2 percent, which means it is difficult to determine whether Locke or Parker has more support than the other.
“It seems very likely that if this trend holds up, that Peter Brown will be in the runoff,” said 11 News political expert Bob Stein, who conducted the poll with the Center for Civic Engagement at Rice University. “The most likely opponent will be Annise Parker, but that is still very much up in the air,” he said.
Stein said he did not expect to see such a high percentage of likely voters who said they still do not know who to vote for. When Bill White was first elected on 2003, Stein said, a similar poll conducted at the same time showed an “undecided” figure of 12 percent.
I don’t see a link to crosstabs, or any discussion of what “likely” means here. Is it the case that the screen was done by asking people if they were likely to vote, as appears to be the case with the Chron poll, or was there a pre-screen based on recent voter history, as was the case with the Parker internal poll? If it’s the former, and I kind of think that it is, then I have the same issues I did with the Chron poll. I mean, through the first four days of early voting, 75% of all early voters had voted in all three of the last three city elections, according to the analysis done by Kyle Johnston. These are the hardest of hardcore voters. Only 2% of early voters had not voted in any of the last three city elections. I’m sorry, but unless someone has a demonstrated history of actually voting in odd-numbered years, I’m not taking their word for it when they say they’re “likely” to vote.
I also think that the high number of undecideds in this poll, which compares to a 24% “undecided” rate in the Parker poll, is evidence that this poll isn’t really sampling “likely” voters. I believe that among those who really are going to vote, there aren’t that many who don’t have a good idea of whom they prefer. To be sure, there are still folks who are wavering, but I don’t believe it’s that high.
Having said all that, this is now two good polls for Brown, and they have definitely changed the perception of the race. He’s run a good race, he’s got a good message about which he’s been very consistent, and he’s been able to get that message out. That’s got to be having an effect. I’m just still not sure how big that effect has been.
UPDATE: Here’s the KUHF story. The key bits:
Rice University Professor and Political Scientist Bob Stein conducted the survey of 545 registered voters over the last week and a half. He asked people how strong their support was for each candidate.
“Peter Brown’s support is interestingly weak. Tepid would be the right word. Gene Locke and Annise Parker, 3-1 voters who tell us they’re voting for Gene Locke and Annise Parker are strong supporters — 75 percent. In Peter Brown’s case it barely breaks 58-59 percent. I think what Peter Brown has got is a broad base of support. He’s got support in every community. The problem for Peter will be probably that support is not very deep and may not take very much to, how shall I say, peel it away.”
Again, this strongly suggests what we have here is a self-screen for likelihood of voting. I am not surprised.