Margolis: When push comes to shove

Chittenden County State's Attorney TJ Donovan gives a speech at the launch of his campaign for Attorney General. VTD Photo/Taylor Dobbs

Chittenden County State's Attorney TJ Donovan gives a speech at the launch of his campaign for attorney general. VTD file photo/Taylor Dobbs

Jon Margolis is VTDigger.org’s political analyst.

Here’s the bad news about this poll story.

It isn’t that the poll taken by TJ Donovan’s campaign to help it figure out how to beat incumbent Attorney General Bill Sorrell in this month’s Democratic primary means “push polls” are coming to Vermont.

Push polls have been here before.

Besides, this was not a push poll, even if a few Sorrell supporters suggested it was. One of the Sorrell backers who complained about the poll said his interview lasted about 15 minutes. Push polls last a minute or two, often for one quick question such as, “Would you vote for Joe Smith if you knew he was a child molester?” This was a multi-question survey.

“If you’re trying to smear another candidate, you’re not going to take 15 minutes to do it,” said Scott Keeter, a former president of the American Association of Public Opinion Research (AAPOR). “You do it fast and move on.”

Besides, to be effective, push polls have to reach thousands of voters. This one reached a randomly selected sample of 400 likely Democratic primary voters.

Nor is the bad news that Donovan took the poll to detect what Sorrell’s vulnerabilities might be. Sorrell campaign manager Mike Pieciak said the Donovan poll was trying “to determine what was the best negative attack to use against Bill Sorrell, and that’s not the way we do politics in Vermont.”

As evidence, Pieciak noted the insult-free five-way Democratic primary for governor in 2010. That was an unusually civil campaign, but the key word there is “unusually.” Candidates in Vermont, like candidates elsewhere, have been exploiting their opponents’ weaknesses roughly forever.

The bad news illustrated by both the poll and the coverage of it is more subtle. There is no scandal here. There is, though, evidence that some of the less attractive facets of American political and journalistic cultures are seeping into Vermont. That there was also a small episode of comic relief in the proceedings does little to make those proceedings less disturbing.

To begin with, there is the simple fact that this story made the biggest splash of the Sorrell-Donovan race so far, even though it was arguably not a story at all. Nothing happened. At the urging of the Sorrell campaign, a few of the incumbent’s supporters called reporters to complain that some of the questions displayed an anti-Sorrell bias.

Perhaps they did. But this poll was not an objective effort undertaken by a news organization or academic researchers to find out who was ahead and why. It was a tool of the Donovan campaign, a mechanism for honing its message. That would require exploring Sorrell’s potential vulnerabilities.

Nor is it likely that Donovan is the first Vermont candidate to take such a poll. He may just be the first to get “caught,” because someone complained.

What seems to be happening is that, even in Vermont, a campaign is becoming about … itself. It’s post-modernist politics dominated by what Washingtonians call “inside-the-beltway” concerns. Neither Burlington nor Montpelier has anything resembling a beltway, so Vermonters will have to come up with a new metaphor. But it’s unlikely that the average voter cares very much about the questions asked in a poll.

It isn’t as though reporters have to wait for these two candidates to complain about each other to find good stories to write. Just check the two campaign websites. Though they are both center-left Democrats who agree on most public policy questions, these two candidates display very dissimilar approaches and attitudes toward the office they seek, and some differences on specific issues. But none of this has attracted as much ink or air time, or as many pixels, as the non-scandal over the poll.

Now comes the comic – or perhaps farcical – relief, with some not-so-funny consequences. In an effort to disprove that its survey was a push poll, the Donovan campaign invited reporters to read the questionnaire.

As long as said reporters did not tell their readers/viewers/listeners what the questions were.

In what campaign manager Ryan Emerson called an “unprecedented show of transparency on the part of this campaign,” reporters were given a half hour to view a copy of the poll questions. But they would “not be able to retain any questions either digitally or in hard copy, nor will the information be available for reporting as it is proprietary and protected.”

Unprecedented transparency but the information can’t be reported?

George Orwell, thou should’st be living at this hour.

Emerson said he’d “never before heard of a campaign in Vermont allowing the media to come in and view their questions.”

Actually, campaigns leak information – including the wording of questions – quite frequently when it is in their interest. In the view of Paul J. Lavrakas, the president of the AAPOR, the campaign “cannot credibly claim to be transparent and then expect journalists to restrict the reporting on what they have been told … journalists should shun covering the story if the polling group won’t disclosure basic info about the poll such as the actual question wording.”

Let’s be fair to the reporters, who probably felt they had no choice. At least one of them, the Burlington Free Press’ Terri Hallenbeck, saw enough absurdity in the situation to do a small parody in her “Vermont Buzz” blog, in which she imagined writing a story saying, “Democratic challenger for attorney general T.J. Donovan showed reporters some pieces of paper Friday afternoon. Then he took the papers back. Eventually, everyone left the room.”

But the reporters did have a choice. They didn’t have to agree to those conditions, because if the dispute was whether the Donovan survey was a push poll, they didn’t need to read the poll questions to solve the dispute. By the accepted definition of “push poll” this one was not.

This doesn’t mean that the Donovan poll was not a good political story, and one that might not reflect favorably on the challenger. Did he really have to take a poll to determine whether voters want an attorney general “who leads the way,” and who “will bring change to the office?” (These were among the “types” of questions, information reporters were allowed to communicate). If his own analysis of Sorrell’s tenure couldn’t give him enough reason to make the challenge, maybe he shouldn’t be running. But that still doesn’t make his survey a push poll.

It isn’t that reporters should never agree to withhold some information. Tom Rosenstiel, the director of the Project for Excellence in Journalism, said, “Whenever a newsmaker tries to set up unusual conditions, (reporters should ask whether) accepting these conditions adds to public knowledge or not. If you are not allowed actually to communicate anything about what you have learned, it becomes hard to say you’re adding to the public knowledge.”

In this case, accepting the conditions did not add to the public knowledge because the acceptance was not necessary. Again, the point here is not to berate these particular reporters, all of whom are dedicated and capable, and none of whom had a lot of time before deciding what to do. It’s the (sadly) dominant journalistic culture seeping in, what somebody once called the journalism of “opinions about the shape of the earth differ,” in which a reporter (or more likely his or her editor) thinks that accurately quoting both Sen. Smith saying that the earth is round and Sen. Jones insisting that it is flat is doing the job.

It isn’t. That’s not journalism; it’s stenography. Reporters are allowed to know empirically verifiable fact: The earth is round (or ovoid, but not flat); the Donovan campaign did not take a push poll; transparency is inconsistent with being gagged.

Comments

  1. Amelia Silver :

    This story, and the other stories on this subject are all Much Ado About Nothing. There was no gagging of the press. There was no push poll. The self-absorption of political campaigns is de rigeur, and in this case, insignificant. Journalists should move on to real stories. Campaigns should continue to do the good work of articulating the particular stances and goals of their candidates. The people should listen closely to the substance of what the candidates say, and VOTE.

  2. Jeffrey Marshall :

    The real issue is that voters feel manipulated by these polls, whatever you want to call them, and a candidate who uses this tactic without identifying him or herself is behaving in a way that most people would consider very annoying if not unethical.

  3. Marshall Weiss :

    Probably only political junkies really care about this but…alas, I am one of them.

    Margolis is right that campaigns tend to leak polling information all the time, as long as “it is in their interest to do so.” That’s not transparency. Transparency isn’t picking and choosing only the positive elements of a candidate/campaign to send out to the public. Transparency is showing your cards even if you don’t know the outcome (i.e. how they will be reported).

    I think Margolis is wrong here. This was a transparent action simply because Donovan had no idea how reporters would react to reading through a long list of poll questions that were specifically chosen to understand what Vermont voters are thinking about this AG race. He didn’t have to release the questions and could have buried the story. But he didn’t. He met it head on and opened his hand to the table.

    Also, oddly enough what’s not being talked about here is the fact that polls provide a competitive advantage (if done correctly) to the campaign that issues them. The information contained within polling results gives any campaign a leg up, which is why polling ain’t cheap — $10,000 for Donovan. Not only is that information understandably valuable, it’s proprietary. Releasing it is obviously a bit more tenuous, even if transparency is your top issue.

    I think Donovan deciding to release the questions despite not knowing how the media would cover it, shows a degree of transparency I’ve not seen before. If he just released the positive aspects of the poll, then it would just be business as usual within a typical campaign spinning chosen elements of an internal poll.

  4. Marshall Weiss :

    P.S. I find it hilarious that the only poll Sorrell is willing to talk about is one that is not his own. Has he issued a poll in Vermont? I mean, for the sake of transparency, why not share that information Mr. Attorney General?

  5. Jane Stein :

    Margolis, “Though they are both center-left Democrats who agree on most public policy questions, these two candidates display very dissimilar approaches and attitudes toward the office they seek, and some differences on specific issues. But none of this has attracted as much ink or air time, or as many pixels, as the non-scandal over the poll.”

    Right-o, including this post complaining about all the attention being paid to this story.

    How about one of those clear, concise pieces you’re so good at, Jon, about those “very dissimilar approaches and attitudes”?

  6. Donald Kreis :

    Thank you, Jon Margolis, for identifying the real story here, which has to do with the possibility that journalists were regrettably complicit in agreeing to the Donovan campaign’s demand that they not report information they were supposedly given in the name of transparency. Though some may find my opinion to be too unsympathetic, as a former reporter myself I’d like to think that any decent journalist would have the gumption, in real time, to respond to such a demand with nothing more agreeable than a familiar two-word scatalogical phrase. If reporters can’t think on their feet then the case for their necessity in the internet age is undermined.

    Earth to T.J. Donovan: You are a good guy, who would almost surely be a great AG, and you were put in a bad spot because of this non-story. But telling journalists they can see but not report something because it’s “proprietary” was a big error. You should say you made a mistake.

    • Christian Noll :

      Mr Kreis,

      IF it is true that Mr Donovan paid “$10,000″ for it would it not be “proprietary?”

      Where’s the “mistake” in that?

  7. Brian Harwood :

    I was polled. I found the final positional questions confusing and tough to answer. Not sure what TJ got out of it,but if could unwind the answers to those questions he deserves to be the new AG.

  8. Robin Windemere :

    I guess I don’t understand the world of politics. In the world I live in, if I call up 400 strangers and tell / ask them something for 15 minutes, I can’t claim with a straight face that what I told them was “confidential and proprietary.”

    I assume any of the persons Donovan’s pollsters had called could have recorded the call or written down some of the poll questions verbatim and posted either on the web. If that’s true, then how do you claim the questions are confidential?

    How does this sort of thinking play into what Donovan would consider confidential under the public records laws? Could the state still claim a government record is confidential if one of its employees had read it over the phone to 400 strangers?

    Politics is its own world, i guess.

  9. William Boardman :

    Very long story about a very short subject: reporters going for the push poll bait got suckered again.

    The primary complainer cited has a son who works for Sorrell.

    If you haven’t seen the questions, you don’t know what kind of poll it is.

    Basic reporting isn’t nearly as hard as it is uncommon.

Comments

*

Comment policy Privacy policy
Thanks for reporting an error with the story, "Margolis: When push comes to shove"