Really, likely voters?

With just 26 days remaining before Election Day, another poll shows Democrat Terry McAuliffe leading Republican Ken Cuccinelli. But that’s not what caught my eye.

QMcDReally? 21% of likely voters haven’t heard enough about Governor Bob McDonnell to form an opinion of him?  They are planning to vote?

But it gets worse.



The majority of those surveyed  – and in most cases, the large majority – haven’t heard enough about the down-ticket candidates to form an opinion. Adding insult to injury is that self-identified Democrats and Republicans don’t know enough about their own party’s candidate to form an opinion!

I can only hope that the Quinnipiac likely voter screen is way off. Because if this is true, things are worse than I thought: voters will choose on E-Day just based on party ID.


8 thoughts on “Really, likely voters?

  1. Vivian, I’ve had the same argument. Simply asking someone if they plan to vote is NOT a likely voter screen. You model the likely voter screen, or you eliminate unlikely voters from the phone pool. If all you do is ask someone who only votes in Presidential elections if they’re voting in November, they’ll say yes. They aren’t showing up.

  2. How about this?

    Q2 On November 5 of this year, there will be a general election for Governor, Lieutenant Governor, Attorney General and other offices. What are the chances of your voting in the November 5th General Election? Are you almost certain to vote or will you probably vote or [sic] in the November 5th general election?

    Actual question from actual poll released a few days ago. Poll shows 100% said yes. (Of course, I know they only used the Yes answers, but still… )

    1. It really is a worthless question–it reads more like a generic GOTV script than a research question.

      If, on the other hand, the poll asked

      Q1 Can you recall off the top of your head when the next general election in Virginia will be held?

      Q2 Can you name at least one office that’s on the ballot in that election?

      I’d feel a lot more confident.

      I sent at least one angry letter to Quinnipiac last year to point out a gigantic flaw in their research methodology. I seem to recall that their Virginia surveys were generally okay, but they somehow kept ending up with samples in…Colorado, I think? where their likely voter screen kept predicting an electorate that skewed older and whiter than even the 2010 midterm.

  3. “[If] this is true, things are worse than I thought: voters will choose on E-Day just based on party ID.”

    Pardon my confusion, but didn’t you ADVOCATE for putting party affiliation on the ballot, so that uninformed voters could do EXACTLY THAT?

    Am I incorrect in thinking you had advocated for party labels on the ballot?

    1. Either all candidates chosen by a party process should have the labels on the ballot or none should. It’s just that simple. I will always advocate for that.

      Which is not the same as advocating for uninformed voters. I’ve been pretty clear that voters should be well-informed: look at the TJ quote which graces my blog.

      But, of course, you know that.

Comments are closed.