The city of Charlotte recently touted a poll showing that residents of Mecklenburg’s six suburban towns support a public transportation plan that’s expected to cost between $8 and $12 billion.
To pay for it, the city wants to raise the sales tax by a penny, to 8.25%.
“Sixty-two percent — really a 2-to-1 margin — favor increasing funding for local public transportation infrastructure in Mecklenburg County,” city planning director Taiwo Jaiyeoba told the City Council in early February.
The polling details are important because Jaiyeoba has said state legislators want proof of support for the city’s transit plan before they agree to allow a referendum on the ballot.
Jaiyeoba told reporters this month: “We believe that once we secure support from our county and towns and we have this type of information, it gives us a very good place to go back to the General Assembly and say that we have answers to some of the questions that they had.”
But his presentation to City Council only contained two PowerPoint slides. They listed some of the topics pollsters asked – but not the actual questions.
So WFAE made a public records request for the entire script of the poll.
The city of Charlotte said it doesn’t have it. It also said it wouldn’t ask the pollster for a copy.
City spokesman Cory Burkarth said “it wouldn’t be appropriate to use a city employee and taxpayer resources to reach out to a third party to obtain documents and/or data on behalf of a reporter.”
That’s surprised some political observers. How a poll is conducted and the questions asked can make a big difference in the answers that are received.
“If an organization claims a poll says something, but doesn’t release the entire script of the poll, the cross-tabs or the full list of responses, those polls are always questionable,” said Larry Shaheen, a former political consultant.
Shaheen is a Republican. Tom Jensen with Public Policy Polling – a pollster that often works with Democratic clients – said the city of Charlotte’s handling of the poll is unusual.
“I definitely think that if you want to use a poll as sort of a tool to show support for something, the more transparent you are about what you asked, the more credence people can put in it,” Jensen said.
So what do we know about the poll?
It was conducted by two groups: one is the National Resources Defense Council Action Fund, an environmental group that’s pro-transit; the other is ALG Research, the top pollster for Joe Biden and Roy Cooper in last year’s election.
Jaiyeoba told council members that it was paid for by a grant the city received from former New York City Mayor Mike Bloomberg.
The city then said this week that the NRDCAF did the poll on behalf of the city and that the city never spent any money on it. The NRDCAF said it did the poll for the city as an in-kind contribution.
The PowerPoint presentation showed the city had some information about the poll, such as the number of people surveyed and the breakdown by political party.
One slide gave a general overview of the type of questions asked, but not the actual questions. It said that said 62% of people polled favor “increasing funding for local public transportation infrastructure in Mecklenburg County.”
It’s unclear if the survey told people what that infrastructure would be — and that 70% of the money would go to trains and buses, with only a small portion for roads.
Fifty percent of people surveyed said they would support raising the sales tax to pay for public transportation.
“Subtle things in how you ask things can lead to big differences in terms of how voters respond to them,” Jensen said.
He said the best example of that was in 2012 before the vote on Amendment One, which at the time prohibited North Carolina from recognizing or performing same-sex marriages.
Leading up to the vote, Jensen said one poll showed Amendment One failing by a 2-to-1 margin. Another poll showed it passing by a 2-to-1 margin.
“And the poll that said it was going to fail 2-to-1 asked, ‘Do you support banning gay marriage in North Carolina?’” Jensen said. “And when you are talking about banning something it sounds like you are taking away people’s rights and being punitive.”
He said the other poll asked: Do you support or oppose defining marriage in North Carolina as between one man and one woman?
“And when you put it that way and say you are defining it — people in that version said 2-to-1 they support that,” Jensen said “And that was what was on the ballot.”
Huntersville Mayor John Aneralla said he urged the city to include more detail in the questions, such as how much each family would spend on the plan over 30years. It appears that didn’t happen, he said.
“It turns out they didn’t get granular,” he said.
Republican City Council member Ed Driggs has questioned Jaiyeoba on whether the poll is legitimate.
“Sometimes I think some people try harder than others to avoid bias,” he said. “It wouldn’t surprise me if this poll was done in such a way to be sympathetic to these plans.”
It’s unclear when the city will ask state legislators to OK the sales tax referendum. The vote was planned for November, but will likely be pushed back due to election delays caused by a late data release from the U.S. Census Bureau.
