A word about polls: It all comes down to the methodology and sample size. The larger the sample, the more accurate the result, provided you’re asking question of the right voters. (Inveterate voters, the ones who not only say they’re going to vote on Election Day but who actually have voted on past election days, are the best.)
Having said that, we at Various Things & Stuff can say that Marvin Longabaugh of Magellan Research has always struck us as a good, honest pollster. He’s been doing it for awhile, and his record is pretty darn good. He understands the business and how it’s supposed to work. So let’s address his current and month-old numbers, and see what we can discern.
(The numbers were reported in John L. Smith’s column in the Review-Journal today.)
Here’s the matchup, as reported by Smith:
Jim Gibson: 39.5 percent
Dina Titus: 30.7 percent
Difference: 8.8 percent
Undecided: 29.8 percent
Now, let’s take a look at Magellan’s numbers from last month, also reported by Smith:
Jim Gibson: 38.1 percent
Dina Titus: 28 percent
Difference: 10.1 percent
Undecided: 33.9 percent
So, what conclusions can we draw?
First, it appears the numbers are solid, in that we don’t see big swings from month to month. About 4 percent made up their minds between the current poll and the last one, and those undecideds broke both ways.
Second, although early polls showed Titus ahead of Gibson, those were based on the most ephemeral of factors — name recognition. Gibson’s been on the air with ads, positive and negative, for a couple months, which could account for his surge in these Magellan polls. (In fact, the Gibson education spot in which he laments of schoolchildren, “If we’re last, where does that leave them?” is probably the best ad of the season thus far, even though it lacks specifics.)
Moreover, Gibson’s first-strike assault on Titus in the ad wars would naturally have the effect of lowering her support while raising his. Gibson has reminded voters up north that Titus hasn’t always had kind things to say about them in the past.
Third, Titus hasn’t just sat still. She’s fired back, starting at the end of June. This might explain her cutting into Gibson’s lead, from 10.1 percent in June to 8.8 percent this month. (If only the election were in February 2007, assuming a consistent rate of decline, they’d be even!)
Fourth, the truth of the matter is this: Ads move polls. Campaign appearances move polls. Stories in newspapers and on TV move polls. Debates move polls. And those poll numbers can shift between now and Aug. 15, when, as the politicians like to say, the only poll that matters is taken. (Usually, they say that when they’re behind in the other polls.)
Do the numbers reflect reality? Our unscientific view is that this is closer than an 8.8-point race, but we’ve nothing to base that on, other than gut feelings developed while covering umpteen campaigns. But there is no denying a few things: Gibson is ahead; Gibson’s TV ads have helped him; Gibson’s numbers are much better than when the race started.
But with 11 days until early voting begins and exactly one month until Election Day, our view is that either candidate can still win.
Up next: The Review-Journal’s latest poll, which should hit right before early voting, according to Editor Thomas Mitchell, who told a Las Vegas Chamber of Commerce audience recently the next poll would be available then.
We’ll be eagerly awaiting those results.
This entry was posted
on Tuesday, July 18th, 2006 at 1:45 pm and is filed under
Various Things & Stuff.
You can follow any responses to this entry through the
RSS 2.0 feed.
Both comments and pings are currently closed.
Multimedia
The following comments are provided by readers and are the sole responsiblity of the authors. By publishing a comment here you agree to the comment policy. If you see a comment that violates the policy, please notify the Online staff.