There is finally some polling data on the upcoming 2015 Houston Mayoral Election, conducted by KHOU and Houston Public Media.
Exclusive Poll: Turner, Garcia lead pack in race for Mayor. Doug Miller, KHOU
2015 Mayoral Race. Houston Public Media
The poll was conducted under the direction of Bob Stein who was, unsurprisingly, "not surprised" by the results. If you've followed local politics for any length of time you know that Mr. Stein is rarely surprised by the results of any poll that he releases.
Without viewing the cross-tabs or the entire poll itself, which has not (as far as I can tell) been released to the public, it's difficult to do too much of a deep-dive into what, if anything, this poll tells us.
Absent any context it's pretty much just a quick identifier of name ID early in the election cycle.
Here's how the candidates stack up: (I'm just using the 'likely voters' results although I caveat that by pointing out that I'm unsure of what screen they used)
Sylvester Turner: 16%
Adrian Garcia: 12%
Chris Bell: 8%
Ben Hall: 3%
Bill King: 3%
Stephen Costello: 2%
Marty McVey: 0%
Undecided: 50%
There are plenty of unanswered questions which could skew these results greatly when taken into consideration.
1. What is the geographical dispersion of the persons polled? If, for instance, the responses were clustered in certain geographical areas you could have responses that were weighted toward candidates to whom the respondents were familiar. I'm not saying this IS the case, I'm just saying we don't know.
2. Why is the Republican percentage so low? Only 15% of all respondents identified as "Republican". Given what we know about Houston demographics, that is skewed very low. Either there was an error in the sampling (the calls made during business hours when people were at work) or there's a problem in the statistical model that's being used. Given voting trends we know that over 30% of the population of Houston tends to vote Republican. This survey doesn't reflect that.
3. In what order were the candidates listed as options? Was it random, per call, or was there a fixed order with Turner and Garcia first, and then the rest? Ordering matters as many people will default toward answering the top choices offered. Unfortunately, again, we just don't know.
I could go on and on but I won't. You can see where over-analysis of an incomplete poll can create issues of identification. I do find it odd that NO likely voters chose Marty McVey. While I don't think he has a chance of winning the fact that zero respondents in that bucket selected him raises huge red flags to me.
What we do know is that a lot of people are not yet ready to reveal their choice. While many are saying they are surprised that the undecided vote is so "low" I would suggest that, given we're supposedly dealing with active municipal voters, I'm slightly surprised that it's so high.
It probably understates reality however. I would imagine that, in most cases, people are not even paying attention to this race much yet and that the undecided votes are really those who just haven't taken the time to look at it yet.
The poll also doesn't tell us where the conservative vote is going to coalesce, and I think that's a big weakness because I think that many of them have decided on who their candidate will be. Part of this may be because we don't know how much of this poll was conducted after news of the ReBuild Houston Supreme Court decision broke. Part of it is probably that Mr. Stein doesn't take that reality all that seriously in his analysis and modeling.
One last caveat: It is unclear whether or not the release of this poll is raw answers, or statistically sampled results. That is a pretty big item. If the report is just producing the raw answers then we really don't have a scientific poll of where things stand, we just have the opinions of 500 people. While we do know some basic demographic information there's no direction in the report given as to how this correlates to expected voting patterns. For example, I would surmise that, given recent voter turn-out, the number of Hispanic "likely voters" may be skewed.
As far as the other questions asked I think we're left with a confusing mixed bag.
A plurality of poll respondents claim to want increases in public transportation (41%) yet a majority do not support tax increases to fund them (60%). While a majority rate Mayor Parker's performance as either "excellent" or "good" (55%) majorities disagree with many of her policy selections (Term limit changes ran 53% against for example)
One thing that is clear, a large majority of the 500 people surveyed (68%) do not want the public to finance projects renovating the Astrodome.
Unfortunately, we're getting what we typically get in local elections, a partial release of a possibly flawed survey whose methodology is unclear, and whose interpretations are swallowed whole by a local media who are happy with the results because it melds with what the so-called smart set presupposes is happening.
This leads me to predict that the only thing we know right now is that Sylvester Turner has the best name recognition, that there are some people who are not shaken by Adrian Garcia's negative press, and there remains a disconnect between Houston's ruling class and the voters who vote them in. The conservative vote in Houston will coalesce around a single candidate, this has always happened and it always will. Right now I think that candidate will be Bill King. Although I have no data to back that up, intuitively it makes the most sense.
Perhaps, in the coming weeks, we'll get some more data that fleshes this out further. Right now we're not much better off than we were before this poll dropped. We're still grasping at straws.
