So unbelievable.

VPR rolled out its latest poll today, conducted by the Castleton Polling Institute. I dutifully pored over the results, about which more later. But for now, one simple stupid thing.

There was an impressive array of questions about issues of concern, Vermonters’ impressions of candidates, how they feel about the presidential election…

… but nothing about voter preferences on the August primary races. No head-to-head numbers. No question asking “If the primary were today, who would you vote for?”

(Or, “for whom would you vote” if you insist.)

So I Tweeted an inquiry and got the following, stupefying response.

“We didn’t do a head-to-head.”

As in, “We didn’t ask the question that would be foremost on the minds of those who care about the polls.”

Double-u Tee Eff.

The question was further aired during “Vermont Edition.” Pollster Rich Clark said that the primary electorate is too small and unpredictable. With no real idea how many people will vote and who’s more likely to show up, “We’d need a much larger sample size” to get accurate results, he explained.

Well, shit, folks. VPR and Clark did a poll back in February and they asked for voters’ preferences for governor and Lite-Guv. Why was it okay months ago, but not okay immediately before the primary? Wouldn’t this poll be at least as accurate as its predecessor?

I accept the idea that the results would be particularly iffy, but reading poll results is always an exercise in camel-swallowing. Now you’re straining out a gnat?

And even if the results were only rough indications of voter sentiment, to be taken with a truckload of salt, isn’t it better than nothing?

VPR is the only media outlet that’s paid for an opinion poll so far this year. This is the only available window onto the mood of the electorate.

And they didn’t ask the respondents how they plan to vote for governor.

This may go down in the headlines as “the poll that didn’t ask the biggest question.”

The Things. You Say.

Advertisements

8 thoughts on “So unbelievable.

  1. John Dillon

    please post this

    Hi John,

    Unfortunately the one question that everyone wants to ask can’t be answered with reliable data, and we’d rather not distribute unreliable data to satisfy the curiosity. As you note, pollster Rich Clark explained on Vermont Edition that incredibly low voter turnout in our state primaries, coupled with a lack of data and exit polls from past primaries, means there is no reliable way to ask “who are you voting for?” without using a prohibitively enormous sample size.

    To your point about our previous poll, the head-to-head matchup question can be asked in the presidential primary in early March because turnout for that election is reliably high. (Among the reasons: holding the presidential primary on Town Meeting Day boosts turnout, whereas the August primary stands alone amidst summer vacations.)

    VPR will be conducting another poll in the fall that will ask, “Who are you voting for?” and it will yield head-to-head polling numbers. Again, fall election turnout is high enough and there is enough past voter data to make these numbers possible.

    Thanks for paying attention to our work.

    Reply
  2. Jonathan Butler

    Thanks for the questions about the latest VPR Poll. I know we discussed this on Twitter, but I wanted to capture some of the points here.

    It’s true. The one question that everyone wants the answer to – who is leading in the VT primary races? – is one that we can’t answer with a reliable data. We discussed this while planning the Poll because we had the same question, but we were advised against it by our partner, Rich Castle from the Castleton Polling Institute, because it wouldn’t be statistically accurate or reliable. According to Rich, the issue isn’t the race or the question, it’s the turnout in the vote we’re trying to predict.

    In the VPR Poll in February 2016 we were able to ask about the races in the March primaries because the turnout in March is over 50%. Rich speculates that Town Meeting and related activities drive up turnout; if we just had primary voting that day with no Town Meeting, turnout might be much lower, compromising polling before the vote.

    In the Fall VPR will conduct another Poll before the general election, and we’ll ask those “horse race” questions – for President, Governor, Senate, Lt. Gov, etc. – because turnout in the general election after the poll is usually greater than 50%.

    The Poll just released today precedes the Vermont Primary on 8/9. Traditionally, turnout for these primaries is very low (if I’ve got this right it was under 10% of registered voters in 2014, according to VT Sec State http://bit.ly/2auY5Rk). If we poll c. 700 people, and less than 10% are likely to actually vote in the primary, it makes the poll results unreliable for predictive purposes. We would have had to poll a prohibitively large sample size to correct for this, so ultimately Rich advised that the best practice at this stage of the election cycle is to poll for awareness and favorability.

    I hope this explains the thinking and decisions that went into crafting this particular poll. There’s more on this issue here: http://blogs.castleton.edu/polling/the-perils-of-polling-in-low-turnout-primaries/ Some organizations ignore this advice, we chose to follow it.

    Thanks for the questions and for following out work. We sincerely appreciate all feedback.

    Jonathan Butler
    VPR

    Reply
    1. John S. Walters Post author

      I guess you’re missing my point about the February poll, which not only asked about the March primary race — it also asked directly about preferences in the gubernatorial race. So there’s one unanswered question: Why could you ask that question in February but not in July?

      Reply
  3. Brooke Paige

    The Most Accurate Poll – Election Day!

    John,

    The finance reports show that several of the candidates, most probably, paid for polling data; I know that Donna and I have answered two telephone poles within the past month or so, not sure if our still having a landline increased our chances of being called?

    In any case, my suspicions are that those who paid for the numbers were displeased with the outcomes, especially Mr. Lisman whose shills have been circulating a 28% figure gleaned from a Castleton poll last fall that only asked about name recognition and not preference in a head to head question. I suspect that the results of the poll, paid for by Bruce, probably revealed that he has captured an amazing 8% – 10% among likely Republican primary voters (there is a rumor circulating to that effect) with his “dark side” campaign – which will mean that he has spent around $100 – $200 PER VOTE based on 2014 Republican turnout figures (15,000 participated) and that figure would only hold if he got ALL of the votes, not 10% of them. If the 10% figure holds he will have spent an astronomical $1,000.00 (or more) per vote, which would qualify his “layout” as the worst political investment in campaign history – not just Vermont history, but the history of the entire civilized world !

    In any case, the only poll that really counts will be in 13 days and that one won’t cost Bruce, Sue, Peter “G”, Phil, Cris, Matt or even little old me a single penny and the results will have a margin of near +/- 0% – oh wait, Jimmy Condos and the boys are running the election, make that +/- 5%.!

    All the Best,

    Brooke

    Reply
  4. Steve Beck

    You know, who gives a rat’s you know what about VPR. I read somewhere, 7Days I think, in the Best of VT edition, that the host of Vermont Edition was VT’s own Terry Gross and that pushed me over the edge. I had been thinking of going cold-turkey and stop listening to VPR/NPR and when I read that I said enough. Cold Turkey. I have listened to some great Shostakovitch and some wonderful audio books. Now if I could just stop reading the digital edition of the New York Times, I might start to think/feel that the state and the country is not having a collective nervous breakdown.

    Reply
  5. Robert Haskins

    Lisman, who has enough Wall Street cash to buy VPR, certainly has head-to-head polling his campaign has conducted. He knows what we know, he’s still drowning in low single digits against a very popular incumbent. If Lisman had remarkable gains after stuffing everyone’s mailbox with toxic sludge these last three months, he’d be the 1st person to crow about it.

    Reply
  6. Hayden Dublois

    I have no idea what VPR was thinking. They thought it was smart to commission a head to head poll on the VT Senate race which is months away, but not on the statewide primaries which are two weeks away? At least they gave us favorability numbers for the gubernatorial candidates….but still, come on. They could’ve easily done head to head match-ups for the Gov and LG races. This leaves political junkies like me unsatisfied.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s