Another week, another press release touting poll results that can’t be truly judged by the Fourth Estate because a campaign won’t release any of the data that goes along with the poll other than what they want reported.
The irony of the latest poll is that it was conducted by Moore Information, whose owner, Bob Moore, was one of my sources in an item last week about the information reporters need to be more vigilant in gathering before writing about polls.
Today’s poll was released by Yes on 100, the campaign committee that is stumping for the temporary one-cent sales tax increase lawmakers referred to the ballot for a May 18 special election. The press release said the poll, which was conducted in late February – though the release didn’t note the days, Moore told me it was done Feb. 24 and 25 – shows 59 percent of likely voters support the tax increase.
To its credit, the campaign released the poll question asked of the 506 voters who self-identified as certain or likely to vote in the May election. That poll question is the exact ballot language that voters will see in a couple of months:
A ‘yes’ vote shall have the effect of temporarily increasing the state transaction privilege sales and use tax by one cent per dollar for three years to fund primary and secondary education, health and human services and public safety.
A ‘no’ vote shall have the effect of keeping the state transaction privilege sales and use taxes at their current levels.”
But, as with the polls I examined in last week’s blog post, the campaign refused to provide a copy of the script of the entire survey or cross-tabs. Moore gave me some of the demographic information, including a breakdown of respondents by age and party affiliation, but the data can still be summed up with one word: incomplete.
There were other questions asked to voters, but Yes on 100 spokesman David Leibowitz wouldn’t release them to one of my colleagues who spoke with him today. We don’t know how many questions they asked, what they asked or even why. It’s likely the questions were done to help shape the campaign’s strategy in the coming weeks by testing possible messages, but we don’t know if it was limited to that. We have to take Moore and the campaign at their words when they say none of the other questions asked would have an impact on how the respondents view the ballot measure.
Frankly, that’s not out of the ordinary. Campaigns often do polling to gather data that will be used to guide strategy decisions in the run up to the election. But the information gleaned from those polls, and the results themselves, are used internally and not made public. And that’s well and good.
But by offering a portion of the survey for public consumption, they want the best of both worlds. The campaign expects reporters to take that bit of the poll it wants to tout publicly and cover it, but they refuse to release the complete poll. They can’t have it both ways.
Or maybe they can, if the media is willing to be complicit. Polls are like candy to political junkies, and it’s tough not to report on them. It’s no secret why reporters and those who follow politics gravitate to them.
Already, my Capitol press corps colleague Dennis Welch, of The Arizona Guardian, has written about the poll without questioning the lack of supporting information that should come with a poll. Others haven’t yet written about the poll results, but they may. And it will be a journalistic sin to not call attention to two things in the stories: The campaign paid for the poll, and they refuse to release the data needed to fairly judge the validity of the results.
It’s possible the poll is entirely accurate. Moore has a sterling reputation nationally as a pollster and a poll he did on the sales tax increase last summer showed what I believe were the most unbiased results of any of the polls conducted last year. That poll, which was done in June 2009, showed 64 percent of likely voters supported the sales tax increase for education and public safety.
Why was it the best poll on the subject last year? Because it was an independent poll. There were also polls done by backers and opponents of the tax increase, but both were suspect because of who paid for them. When I asked Moore about the poll last year, he said he did it to try to drum up business in Arizona, a state where he hadn’t worked before. At that time, he readily produced full cross-tabs and information on the poll.
But this is a different situation. This poll is paid for by the campaign and, as such, is their property. Whether Moore wanted to or not, he couldn’t release the data to me or any other reporter, as long as the campaign doesn’t twist the data or publicly mischaracterize what it said.
The campaigns, though, need to be confronted with a skeptical Arizona media that refuses to lap up whatever numbers they put out as poll results if they aren’t willing to give us the information we need to properly evaluate the results. If that happens, Arizona will be all the better for it.
- By Jim Small