Skip to main content

View Diary: Punish CNN for Bogus Poll: Remove Candy Crowley as a Moderator (281 comments)

Comment Preferences

  •  It's your choice (20+ / 0-)

    We can see from the exclusive and exhaustive income categories that SEs of 8.5% and 6.5% combine to an overall SE of 4.5%.  That suggests that the urban/suburban and moderate/conservative pairs pretty much exhaust the population of the sample.

    Attended college is 5.5% SE; overall is 4.5%.  The marginals are about the same.  There would be a smattering of non-college people there.  Ditto for the 50 & older.

    It strikes me that this might be a sample of CNN viewers alone, although that should be stated, if so, on page 1.

    It is completely irresponsible to take a sample that out of whack and publicize it as representative of the general public.  I look forward to your concession on that point.  This sample is so screwed up that I don't think that the decision to use it as a false portrayal of the general public could be an accident.

    Pro-Occupy Democratic Candidate for California State Senate, District 29 & Occupy OC Civic Liaison.

    "I love this goddamn country, and we're going to take it back." -- Saul Alinsky

    by Seneca Doane on Thu Oct 04, 2012 at 04:22:15 AM PDT

    [ Parent ]

    •  so you refuse to correct your diary? (5+ / 0-)
      Recommended by:
      soros, Bob Duck, Argyrios, Zornorph, distraught

      Seriously, I'm shocked. At best, you're moving the goalposts all over the place.

      it polled only white college-educated Southerners over age 49 who self-identified as either "moderate" or "conservative"
      Status: false.

      It is past time for you to reassess your evidence and revise your diary accordingly.

      It strikes me that this might be a sample of CNN viewers alone, although that should be stated, if so, on page 1.
      Page 1 is very clear. Clearly it isn't a sample of CNN viewers alone, unless you think that they are lying, badly.
      It is completely irresponsible to take a sample that out of whack and publicize it as representative of the general public.  
      Debate viewers aren't representative of the "general public" to begin with.

      We can talk about the propriety of reporting snap poll results, but we can't do it very effectively as long as you are engaging in, frankly, screwball speculations about the poll being exclusively of Southerners, or CNN viewers, or whatever.

      Election protection: there's an app for that!
      Better Know Your Voting System with the Verifier!

      by HudsonValleyMark on Thu Oct 04, 2012 at 05:01:09 AM PDT

      [ Parent ]

      •  You're right -- ENORMOUSLY DISPROPORTIONATELY (14+ / 0-)

        of white, college-educated, 50+, Southerners who consider themselves either liberal or moderate, and COMPLETELY UNREPRESENTATIVE of the U.S. electorate as a whole.

        I recognize that there was a little noise in categories such as age (a few at least in the under-50 set, though not enough to more than caress the marginals), but by and large this is not, one bit, at all, the sample one would get if one were honestly trying to come up with something representative even of those watching the debate.

        But it was not entirely composed of those people -- just enough so to get the job done.  That's the real scandal here and I hope that you will join me in recognizing it -- and in calling for them to divulge the numbers of respondents in each cell.

        Pro-Occupy Democratic Candidate for California State Senate, District 29 & Occupy OC Civic Liaison.

        "I love this goddamn country, and we're going to take it back." -- Saul Alinsky

        by Seneca Doane on Thu Oct 04, 2012 at 08:13:50 AM PDT

        [ Parent ]

        •  what a mess (1+ / 0-)
          Recommended by:
          Bob Duck

          If you can support your assertion that they somehow rigged their callbacks, please do. If you don't provide any supporting evidence, I damn well can't join you in recognizing it, can I?

          I think this is very plausibly the sample one might get if one were calling back respondents in a previous survey under time pressure. Whether and how they should have weighted it is a whole 'nother question.

          I don't think you've really thought through your demand for more information. It's already pretty obvious (pending Nielsen data) that their sample is older than the debate viewership at large; what exactly do you expect to learn from knowing what their tiny subsample of under-30 respondents said? What inference could you draw, beyond the sound and unsound inferences you have already drawn?

          I don't think there is a real scandal here. I don't think ordinary people actually care what percentage of respondents in some poll think Romney won last night's debate.

          Election protection: there's an app for that!
          Better Know Your Voting System with the Verifier!

          by HudsonValleyMark on Thu Oct 04, 2012 at 09:59:46 AM PDT

          [ Parent ]

      •  I agree with Mark (4+ / 0-)

        You are obscuring a good argument by making a false one.

        1) The simple, good argument is that the  poll oversampled Romney voters compared to the general population of registered voters by (um someone else should figure out the percentages here) alot anyway. We know this because of the pre-debate favorable/unfavorable ratings of 54/42 towards romney.  (when, for general population of voters those numbers are approximately reversed).

        2) This means either that the pollsters had really bad methodology which did not accurately get a sample of people who were demographically similar to the actual people who watched the debate. or

        3) Way  more Romney supporters watched the debate than Obama leaners.

        4)If you look at the pre and post debate favorable and unfavorables they really don't change. What that means is that this is not a poll of true undecideds, or even though people are polled as saying "more likely to vote for x after the debate" it really did not change their basic opinion.

        So basically if you are trying to get a message out, which is obscured it is that

        A group of people who favored Romney much more than the average US voter were more likely to think he won the debate.
        and
        In the end it didn't change anyone's mind much about whether they liked/disliked Romney or Obama,
        they both gained a tiny bit, (Romney 2% fav, Obama 1% less unfav) - But given that this was such a small sample, that is essentially meaningless.

         

        "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 09:14:02 AM PDT

        [ Parent ]

        •  That's a good critique, but a different one (5+ / 0-)

          I think that CNN knew that it had obtained -- whether intentionally or fortuitously -- a sample that was highly discrepant from that of the overall population.  I think that CNN deliberately obscured that fact, particularly in its headlines, to make it seem like it was a far more representative sample than it was.

          Pro-Occupy Democratic Candidate for California State Senate, District 29 & Occupy OC Civic Liaison.

          "I love this goddamn country, and we're going to take it back." -- Saul Alinsky

          by Seneca Doane on Thu Oct 04, 2012 at 10:44:35 AM PDT

          [ Parent ]

          •  They are not claiming that the sample represents (1+ / 0-)
            Recommended by:
            Seneca Doane

            The general population. However, many people just getting this in a headline probably miss that distinction.

            The sample should, if it was done right, represent the people watching the debate.

            However with the low attention span of the average person (IE two seconds for soundbite) The "nuance" of the distinction will be lost.

            "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 03:48:55 PM PDT

            [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site