Another Election Cycle, Another Couple of Polls to Consider
By Stuart Rothenberg
For some reason, campaign managers and press secretaries never tire of distributing press releases and campaign polls that they apparently assume will be taken at face value and regurgitated by political observers.
We’re six months into the 2010 election cycle, and I have already received a pair of press releases about polls that set off alarms, though for very different reasons.
One of the press releases that I received came from the campaign of Texas Senate hopeful Florence Shapiro, a Republican state Senator.
Shapiro is running for the seat held by Sen. Kay Bailey Hutchison (R), who may resign her Senate seat in order to focus on her quest for the GOP gubernatorial nomination next year.
The mid-April press release crows that “Shapiro has highest name ID among announced GOP candidates,” which is a little like saying that Shapiro has the highest favorability ratings when her immediate family was surveyed.
The poll was conducted from March 30 to April 1 by Global Strategy Group, a Democratic firm not in any way connected with Shapiro’s Senate bid. The sample included 603 voters likely to participate in a special election.
The Shapiro campaign’s release cited the Democratic survey and proclaimed that the state Senator’s name identification is more than twice the ID of the “next best known Republican currently running or exploring a campaign for U.S. Senate” and that she “enjoys a +8% net favorability rating,” which is “higher than any other Republican candidate who has taken public steps toward a Senate run.”
The only problem is that none of the other three Republicans who “have taken steps toward a Senate run” — former Texas Secretary of State Roger Williams and state Railroad Commissioners Elizabeth Ames Jones and Michael Williams — is regarded as among the frontrunners in the race, once the field eventually forms.
Most Texas observers expect either Lt. Gov. David Dewhurst or Attorney General Greg Abbott to enter the Senate race if and when Hutchison’s seat becomes vacant.
The Global Strategy Group poll did include Dewhurst in its questionnaire, and the results showed more than twice as many voters familiar with Dewhurst’s name than with Shapiro’s. Dewhurst’s net favorable rating was three times Shapiro’s. Abbott apparently was not tested in the survey.
Shapiro’s press release conveniently left out the Dewhurst numbers, since they weren’t flattering to the state Senator.
The Shapiro press release’s focus on the candidate’s “net favorability rating” advantage over the other announced candidates is also misleading, since it is a net +8 while the largely unknown Jones’ is a statistically identical +5, even though her total name ID is just 8 percent.
The Shapiro release, and the claims contained in it, makes the entire campaign look amateurish.
The other poll and press release comes from the campaign of Congressional hopeful John Garamendi (D), who is running in the expected special election in California’s 10th district, assuming Rep. Ellen Tauscher (D) is confirmed to a State Department post.
As far as I can tell, there is nothing wrong with the survey of 400 likely special election voters conducted in early May by JMM Research for Garamendi, or with the polling memo written by Jim Moore. It’s the breathless press release that accompanied the polling memo and the implication that Garamendi has a huge advantage in the contest.
The reason that Garamendi, the state’s lieutenant governor, leads the race now is that he has been on the ballot frequently for the past 30 years. In fact, Garamendi was on the ballot in California in every even-numbered year between 1974 and 1994, with the exception of 1978 and 1992 (when his wife ran unsuccessfully for a seat in Congress).
He has run for state Assembly, state Senate, state insurance commissioner, state controller, lieutenant governor and governor. Shortly before he released his Congressional poll, he was running in a different race — for governor.
Of course, name ID is important, and Garamendi’s name identification advantage certainly is worth noting. But name ID at the end of a campaign is much more important than before the race has even begun.
And Garamendi’s 11-point advantage over the second-place finisher in the hypothetical ballot test isn’t all that substantial given his name identification advantage.
The sitting lieutenant governor has a name ID of 80 percent but is drawing only 24 percent on the hypothetical Congressional ballot. That means that a large chunk of voters who say they have heard of Garamendi aren’t automatically drawn to him.
The Garamendi folks might also take a moment to consider what happened to Jim Tedisco, a well-known New York Republican state legislator who had been in politics for years. He held an early special election lead over an unknown Democratic opponent — who got known quickly once he went on TV and who overtook the longtime legislator when all the votes were counted.
Garamendi certainly is a very serious contender for this seat. It’s fair to call him the frontrunner. He’s been in politics far longer than his opponents, and that has given him name ID and connections. But the early poll is of limited predictive value, and the lieutenant governor’s campaign staffers might want to note that political longevity isn’t what it was once cracked up to be. Just ask President Barack Obama.
This column first appeared in Roll Call on May 14, 2009. 2009 © Roll Call Inc. All rights reserved. Reprinted with permission.
