The latest Zogby/Wall Street Journal/Battleground States poll–and I use the word with considerable misgivings, given its methodology of using a nonrandom sample gleaned from people who have signed online up to be interviewed–shows Perry with a double-digit lead over Bell:
Perry 33.0%
Bell 22.3%
Friedman 18.9%
Strayhorn 15.5%
This is consistent with most recent polls–but not with Zogby’s poll earlier this month, which showed Perry leading Bell by a mere five points, 30.7% to 25.3%. No other poll that I am familiar with shows the race so close. Nor is it consistent with a private poll I mentioned a couple of days ago that showed Perry at 42%.
I continue to have discussions with people in politics about polls, and whether it makes sense to believe some polls and scoff at others. Why, for example, should I disbelieve the Zogby poll showing Bell within 5 points of Perry? Two reasons: (1) My answer is this: Because, if Bell were truly within 5 points of Perry in Perry’s in-house poll, Perry would be going negative against Bell. Instead, he continues to run positive ads. He is running like someone sitting on a lead. His proposals are aimed at reinforcing his base, as with the spending restraints he proposed yesterday, not at trying to broaden his appeal. (2) If private polls showed Bell within five points of Perry, Bell would be touting them. And money would be flowing his way. Neither of those things are happening.
Opinions differ about private polls. I happen to trust them. The market provides discipline. When Texans for Insurance Reform (a trial lawyers group) hires Opinion Associates to poll the governor’s race, they want true information. Jeff Smith, the pollster, is on the Democratic side, but his loyalty is to his client. They need the most accurate numbers Smith can give them so they can decide where to put their money. The same is true for the private poll by a Republican group that I cited a couple of days ago (without identifying the party affiliation). Their client wants accuracy, because those numbers will determine where they put their chips. By the way, the two polls I just mentioned are the only two recent polls that have Perry in the forties. If I had to choose between them and Zogby as to credibility, I’d choose them. Bell and Strayhorn have awful TV spots. They aren’t catching on with voters. I have a feeling that voters were willing to look for someone to support other than Perry (who can blame them?) and they looked, and they didn’t like what they saw, and so they are drifting back. Is that a permanent drift? Not necessarily. The negative-ad phase of the campaign has yet to begin, and the race may get more volatile when that occurs.
I am more likely to doubt internal polls of campaigns than external polls, private or public. always suspect that there are two sets of books–one for what the candidate is told, one for what the media is told. In fact, internal pollsters rarely reveal their numbers. Sometimes I’ll get a call from a Perry-friendly or Strayhorn friendly source who says, “That poll is/is not close to what our numbers are showing.” I believe in the inate goodness of mankind–but not that segment who are connected with political campaigns in the heat of an election. I am more likely to give credence to the report of a state senator who tells me that a lot of Republicans are voting for Kinky.
I will confess to one unfortunate weakness: I don’t understand statistics. I don’t have a handle on when the difference between two polls is small enough to be immaterial or glaring enough to suggest error. But I do know that polls with large swings from polling period to polling period, like Zogby, are not trustworthy, especially when their methodology is suspect.
However, I am endeavoring to learn. This morning I signed up with Zogby to be part of the poll’s internet data base. Mainly, I was curious about how it worked. Here’s how Zogby gets its respondents and what it wants to know from them:
“Zogby Interactive Survey (ZIS) will allow you to voice your opinions from anywhere within the United States, online. ZIS combines the latest trends in technology with Zogby International’s record of success and technical expertise in the field of public opinion research.
“With a simple click of a button, you too can join the millions of respondents who have taken Zogby International surveys.
“For years we’ve heard from you asking ‘Why don’t you ever call me?’ – Well, now, with the advancement of the Internet, you have the opportunity to let your opinion be known with the click of a button.
“We request that you fill out this registration form as honestly as possible. In order to safeguard the scientific accuracy of our online polls, you may be randomly chosen and telephoned as a quality control back-up and asked to repeat a couple of your answers – this maintains the accuracy of our polls.”
“How it works:
1. Complete the online registration page and submit it to us.
2. Zogby Interactive will e-mail you and inform you when an online poll is ready for you to take.
3. Complete the survey. (Easy? We thought so.)
4. Zogby International will e-mail you and inform you when results are available to view.
5. See how your opinion stacks up.”
Here is what the registration form wants to know:
*Are you registered to vote in the United States?
Not Sure
Yes
No
*In which party are you either registered to vote or do you consider yourself to be a member of ?
Not Sure
Democrat
Republican
Independent
Libertarian
Constitution
Green
Reform
Natural Law
Other
*Which of the following best describes your ideology?
Not Sure
Very liberal/progressive
Liberal
Moderate
Conservative
Very Conservative
Libertarian
*How likely are you to vote in national elections?
Not likely
Somewhat likely
Very likely
*Overall, how would you rate the performance of George W. Bush as president
Not Sure
Excellent
Good
Fair
Poor
*What is your age?
*Which of the following best describes your educational level achieved?
Not Sure
Less than high school graduate
High school graduate
Some college
College graduate
*Which of the following best represents your race or ethnic group?
Not Sure
White, non-Hispanic
Hispanic
African American
Asian
Pacific
Arab American
Other/mixed
*Which of the following best represents your religious affiliation?
Not Sure
Roman Catholic
Protestant
Jewish
Muslim
Other/no affiliation
*Are you or is anyone in your household a member of a union?
Not Sure
Yes
No
*Which of the following best represents your household income last year before taxes?
Not Sure
Less than $15,000
$15,000-24,999
$25,000-34,999
$35,000-49,999
$50,000-74,999
$75,000 or more
*Gender
Male
Female
I’m waiting for the phone to ring.
