« Jack And The Boss | Main | Svengali Speaks »
January 22, 2006
On A Certain Poll
This is Keir Murray writing.
I have watched with interest vitriolic response from Congressman Tom DeLay's campaign and his supporters to the Houston Chronicle's recent poll of the 22nd Congressional District. I want to take a moment to clarify a few things.
In today's edition, James Campbell (I think) pretty effectively defends the poll's findings against charges of shoddy methodology or bias.
In the piece, Campbell admits the Chronicle should have disclosed my involvement in Mr. DeLay's opponent's campaign -- I agree. The operative word should be "worked" rather than "works" for the Lampson campaign -- I helped Lampson from May to July 2005 until he was able to hire a campaign manager, and was paid a lump sum for those three months of help (a separate $1,000 payment was strictly reimbursement for buying campaign cell phones, Blackberry, etc.) -- but the overall point is taken.
The idea, however, that any involvement of mine in the Lampson campaign somehow biases the recent poll is simply nonsense. Richard Murray (full disclosure: he's my dad) has been polling in Houston for 30 years. Whatever his own political views, he's been getting it right for a long time, and has built respect for his work from across the political spectrum.
The only two pollsters I've seen criticizing the recent poll are Chris Wilson and Dr. David Hill both of whom are respected partisan Republican pollsters -- since so much has been made about full disclosure in all this, let's disclose fully, shall we? Their criticisms are debatable, but focus in on a couple of very specific details, rather than the overall methodology of the poll. (Note: All polls are open to interpretation and reasonable criticism, and I don't dismiss Mssrs. Wilson and Hill's ability to objectively analyze the results simply because they are partisan, just as I believe two Democrats in this case have produced a valid, legitimate survey.)
But step back and take a deep breath folks, and look at the basics here. Since the Chronicle has made its internal numbers open and available to all, (hardly commonplace -- for example, we are never privy to internal numbers for the Scripps Texas Poll, we must accept its numbers on faith), we can see what's what.
Look at who was polled:
42% of those surveyed identified themselves as Republicans, 27% as Democrats and 23% as Independents; certainly not too many Democrats polled.
The sample was 74% Anglo, 8% African American, 8% Hispanic, 10% Asian/Other; certainly not too many minorities polled.
13% of those surveyed identify themselves as Liberals, 36% as Moderates, and 43% as Conservatives; hardly too many Liberals.
The recalled vote of those sampled in the 2004 22nd Congressional district race was 52% for DeLay, 39% for Morrison; 9% Other; the actual vote ratio was 55% for DeLay, 41% for Morrison; 4% Other; spot on, folks.
But the kicker for me is, if this is a sample so biased against Tom DeLay, then why does President Bush garner a solid 55% approval rating from those surveyed? The President received 63% of the 22nd District's vote in 2004, and his modest erosion of support here absolutely tracks with what's happened in all the national polls.
Now, with regard to the 215 or so certain or likely Republican primary voters asked about who they will support in the upcoming GOP primary, much has been made about the large margin of error due to a small sample size. Only about 40% of this group said they plan to vote for Tom DeLay on March 7th. So, with say, an 8% MOE, his support number could be as high as 48% or as low as 32%. Wherever the number within that range, this is an incumbent Congressman with 100% name ID, who has earned 80% of the vote in recent primaries. Margin of error or not, there is huge erosion here, no matter how you slice it. I suspect Tom DeLay will win his primary without a runoff, but not by much, and that, in and of itself is a big deal.
Putting on my political operative hat, I fully understand why the DeLay folks have yelped so loudly over this poll -- politically, what choice do they have? The numbers are simply devastating, and their only option is to try and discredit the results. Were I in their shoes, I would likely do the same. In such a situation, from a political standpoint, the facts are not nearly so important as perception. So, the campaign generates as much smoke as they can to cloud the results, which hopefully keeps the fundraising dollars coming in, and keeps supporters from jumping ship. That's all you can do.
As for the facts, only time will tell.
Posted by houtopia at January 22, 2006 10:51 AM