Skip to main content

View Diary: So, about that CNN Snap Poll (290 comments)

Comment Preferences

  •  Fair point . . . (3+ / 0-)
    Recommended by:
    wishingwell, MHopeful, Larsstephens

    if the composition of the audience skews Republican, then what the polls are really showing is that Romney bested Obama in front of a Republican leaning audience.

    Huge shocker?  Not really.

    Part of reporting facts, is to report them in context.  So, assuming that the interpretation is correct that there was a GOP skew -- and that I'm not simply misreading the data set -- the skew should be reported as such along with the top-line number.

    •  A 10 or even 15% skew (4+ / 0-)

      might reflect a self-selection bias--more Rs watched than Ds.  A 90% skew, all in R-favorable demographics, simply stinks.

      •  Yes, (2+ / 0-)
        Recommended by:
        wishingwell, Larsstephens

        It doth have a stench, methinks.

        "We have always known that heedless self-interest was bad morals; we know now that it is bad economics" - F.D.R.

        by biscobosco on Thu Oct 04, 2012 at 12:30:29 AM PDT

        [ Parent ]

      •  Not sure where you are getting a (0+ / 0-)

        90 percent skew.

        The composition of the viewing audience skewed GOP by about 15 percent relative to national polls.  Romney's favorables were about 15-20 percent higher.  

        All of that points to the idea that the audience was unrepresentative of the electorate as a whole.

        This is entirely possible.  Maybe people who are committed to Obama wanted to watch other programming (or not watch TV at all).  There was no new information that they needed to see in order to make a judgment about the race, and they aren't political junkies like us who watch simply to see how our guy performs.

    •  NO (2+ / 0-)
      Recommended by:
      wishingwell, wu ming

      The polls are showing that the demographic which favors romney far more than the nation does thought Romney did a better job.

      In my left leaning household of 15 who watched the debate we came to a completely different conclusion. Romney looked like a nervous attack weasel to us.

      "We have always known that heedless self-interest was bad morals; we know now that it is bad economics" - F.D.R.

      by biscobosco on Thu Oct 04, 2012 at 12:29:57 AM PDT

      [ Parent ]

      •  I came to a different conclusion too . . . (0+ / 0-)

        that was more in line with your experience as far as my personal reaction goes.

        However, I'm not going to say from my own perception of the debate is representative of the entire audience that watched the debate (or even a majority of viewers).

        Was the audience that watched the debate representative of the entire voting population?  No, it probably wasn't.  

        The CNN poll shows a GOP partisan advantage pre-debate, even though the national polls indicate a Dem advantage.  Romney's net favorables also skewed more positive than they do in national polls by a significant degree.  All of those factors point toward the idea that the viewing audience was more GOP leaning than the population as a whole.  This is almost certainly true.  If the viewing audience skews Republican, then perceptions of who the winner was are likely to skew in the same direction.

        •  if the sample is representative of the viewing (0+ / 0-)

          audience.

          It seems like if you are using 430 people to represent 56 million, you have a chance of being off a little.

          "We have always known that heedless self-interest was bad morals; we know now that it is bad economics" - F.D.R.

          by biscobosco on Thu Oct 04, 2012 at 01:58:55 AM PDT

          [ Parent ]

          •  oh and also (0+ / 0-)

            There is also the chance that the selection was biased. (by that I don't mean to imply intentionally biased, only that there was selection bias for some other reason.)

            "We have always known that heedless self-interest was bad morals; we know now that it is bad economics" - F.D.R.

            by biscobosco on Thu Oct 04, 2012 at 02:00:06 AM PDT

            [ Parent ]

        •  If you assume (0+ / 0-)

          That CNN's methods of selecting the sample truly produced a representative random sample.

          We don't know that this is true yet. It would be interesting to see the other polls which diverged considerably from this one, and also to see what the exact demographics of this poll were, and compare that against what Nielsen ratings say.

          "We have always known that heedless self-interest was bad morals; we know now that it is bad economics" - F.D.R.

          by biscobosco on Thu Oct 04, 2012 at 02:04:27 AM PDT

          [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site