What to make of Jennifer Roberts’ new poll?

Democratic mayoral candidate Jennifer Roberts released a poll Thursday, but her opponents’ camps are crying foul.
Democratic mayoral candidate Jennifer Roberts released a poll Thursday, but her opponents’ camps are crying foul. dlaird@charlotteobserver.com

Jennifer Roberts came out with a new poll Thursday showing her in the lead in the race for Charlotte mayor, but other candidates and their supporters are focused on the pollster’s negative portrayals of them.

Roberts’ campaign said 30 percent of voters picked Roberts as their top choice, followed by Dan Clodfelter and Michael Barnes with 16 percent each and David Howard with 9 percent (with 29 percent undecided). The four are expected to face off in a Sept. 15 Democratic primary.

There has been buzz around town in recent days, though, that people were receiving telephone calls in which a pollster would insinuate negative things about Roberts’ opponents. Depending on who was polled, how questions were worded and the order of the questions, the horse-race results could be badly skewed. Anytime a political campaign commissions its own poll then releases only some of the results, take it with a grain of salt. Or a small pile.

Josh Ulibarri, with Lake Research Partners, said the campaign did in fact test negative messages about the other candidates. But he said that is common practice and, importantly, that the horse-race questions were asked first, before anything about the other candidates was said or asked.

Candidates are sometimes victims of what’s called “push polling.” The pollster conveys negative information about a rival and disguises it as a poll question. Sen. John McCain was singed by this tactic in the 2000 South Carolina presidential primary. Pollsters asked voters if they would be more or less likely to vote for McCain if they knew he had fathered an illegitimate black child.

Ulibarri said Roberts’ survey was not a push poll. He said it asked questions to hone strategies to use against opponents later in the campaign. The difference? Push polls are done days before Election Day, are asked of thousands of voters and are designed to change their votes and the election outcome. Roberts’ poll was done nearly three months before the primary and of 400 voters, probably not enough to change outcomes.

“The second half of the poll is strategic: How do we frame our candidate, how do we talk about the other candidates. That’s usually one-sided,” Ulibarri said. The strategic messaging questions are about “how to help Jennifer Roberts become mayor. It’s not designed to change their votes in an election. It’s to see what persuades people to move one way or another.”

That’s why the campaign won’t release the wording of the poll’s questions, Ulibarri said. “If we release the questions to the public and to you then the other campaigns would learn our strategy.”

He said he asked the horse-race questions first because the campaign wants an accurate reading, not a skewed one.

It’s impossible for the public to judge the poll’s integrity without more information. Still, the results are an early warning that is sure to get Clodfelter’s, Barnes’s and Howard’s attention. -- Taylor Batten