Skip to main content

On August 4th, Nate Silver published a post at 538 using conventional protocols, i.e., he made a tiny url and tweeted his new post.
But for some reason, the post was almost instantly retracted, and replaced with a different one, not matching the old tiny url.
Nate said this:

An earlier post in this space about poll oversampling was published in error and will be updated and published later this week.
But that was the 4th, THREE weeks ago.
So where is the oversampling post?
Well I managed to find a copy, because the internet is forevah.
And you can read it below the fold.  But why would Nate pull such an innocuous, innocent post?  Because it utterly debunks the screams of "dem oversampling" coming from conservative pundits.

FiveThirtyEight: Why Charges of Poll ‘Oversampling’ Are Usually Misguided

08.05.12 | New York Times

Its a quadrennial tradition for partisans and poll-watchers to complain about the number of Democrats, Republicans and Independents that are included in each survey.

Recently perhaps because Mitt Romney still narrowly trails President Obama in most state and national surveys we have seen a bit more of this from conservatives. They will sometimes allege that these polls are oversampling Democrats, including too many of them in their surveys, and perhaps biasing their results toward Mr. Obama because of this.
....
There are, nevertheless, elements of truth in these critiques. It is certainly the case that some polling firms consistently show more favorable results for Democrats and Republicans than the consensus of polling firms. We call these house effects and our forecast model adjusts for them; if a polling firm is consistently 2 points more Democratic-leaning that the consensus, we strip most of that right back out.
...
It is easier, of course, to identify these cases after the fact. Beforehand, the best you can usually do is to acknowledge that there is some possibility of their occurrence. Even in the waning days of an election, when we have surveys from dozens of polling firms that collected tens of thousands of interviews between them, their biases will not necessarily cancel out, and the error in the surveys may considerably exceed that from sampling error alone.

Still, I think the charges of oversampling mostly miss the point. Let me make 13 relatively brief but interrelated points that explains my philosophy on this issue, and where I see the theoretical and empirical evidence as guiding the debate.

1. Be careful if you see the term oversampled. It is probably being used incorrectly. In blogs, the term oversampled has come to be a shorthand for a poll that includes too many Democrats or Republicans. But thats not quite the way that pollsters use the term.

Instead, an oversample is a deliberate effort to include more of a certain population in a survey to permit more robust analysis of a particular demographic subgroup.

For example, say that a polling firm wants to study the views of Latino voters in more detail at the same time that it is conducting a national survey. Its initial survey of 900 adults may include about 130 Hispanics about their share of the United States population which is not really enough to analyze with much accuracy because of the high margin of error associated with a 150-person subsample. So the polling firm would take an oversample  until it got a total of 450 Hispanics on the phone, creating a respectable sample size. Then it might be able to report, say, how Hispanic voters preferences would be affected by the presence of Senator Marco Rubio on the Republican ticket.

Knowing that it has interviewed too many Hispanics, the polling firm would then down-weight the Hispanic voters when it rolled them back into its national survey and reported the results from all United States adults. In this example, they would reduce the weight associated with each Hispanic voter by two-thirds, since they interviewed 450 when there should be 150 based on their share of the U.S. population. This technique permits the pollster something of the best of both worlds: it can have a more robust analysis of hard-to-poll demographic subgroups without skewing the overall sample.
....

2. Be even more careful when you see terms like skewed or biased.

3. Party identification is not a hard-and-fast characteristic, as other demographic characteristics are.

4. Partisan identification measures are affected by sampling error.

5. Partisan identification is not the same thing as partisan registration.

6. There are many different ways of measuring and asking about party identification.

7. Polls of registered voters, or all adults, typically show a more favorable party identification spread for Democrats.

8. If you are going to scrutinize polls based on their partisan identification, do so equally.

9. Weighting by party identification puts the cart before the horse.

10. There is no absolute standard to measure party identification only other polls.

11. Taking a poll average especially with adjustments for house effects is usually a more elegant solution to the problem.

12. Pay relatively more attention to party identification when you have fewer polls.

13. There has not been any long-term bias in the polling average toward Democratic or Republican candidates.

Why someone want to suppress this excellent and sensible post?
Because if there is no "dem oversampling" then Obama has even more of a lock on re-election than it appears.  And this also means there are just more democrats in all the samples, except of course, Rasmussen's.  And most important of all, it means the horserace is over before it begins.

you're welcome.

7:13 PM PT: It was NOT A DRAFT
http://sandalwood.tweetwood.com/...
and after we talked about it at PEC the mirror site got wiped.

EMAIL TO A FRIEND X
Your Email has been sent.
You must add at least one tag to this diary before publishing it.

Add keywords that describe this diary. Separate multiple keywords with commas.
Tagging tips - Search For Tags - Browse For Tags

?

More Tagging tips:

A tag is a way to search for this diary. If someone is searching for "Barack Obama," is this a diary they'd be trying to find?

Use a person's full name, without any title. Senator Obama may become President Obama, and Michelle Obama might run for office.

If your diary covers an election or elected official, use election tags, which are generally the state abbreviation followed by the office. CA-01 is the first district House seat. CA-Sen covers both senate races. NY-GOV covers the New York governor's race.

Tags do not compound: that is, "education reform" is a completely different tag from "education". A tag like "reform" alone is probably not meaningful.

Consider if one or more of these tags fits your diary: Civil Rights, Community, Congress, Culture, Economy, Education, Elections, Energy, Environment, Health Care, International, Labor, Law, Media, Meta, National Security, Science, Transportation, or White House. If your diary is specific to a state, consider adding the state (California, Texas, etc). Keep in mind, though, that there are many wonderful and important diaries that don't fit in any of these tags. Don't worry if yours doesn't.

You can add a private note to this diary when hotlisting it:
Are you sure you want to remove this diary from your hotlist?
Are you sure you want to remove your recommendation? You can only recommend a diary once, so you will not be able to re-recommend it afterwards.
Rescue this diary, and add a note:
Are you sure you want to remove this diary from Rescue?
Choose where to republish this diary. The diary will be added to the queue for that group. Publish it from the queue to make it appear.

You must be a member of a group to use this feature.

Add a quick update to your diary without changing the diary itself:
Are you sure you want to remove this diary?
(The diary will be removed from the site and returned to your drafts for further editing.)
(The diary will be removed.)
Are you sure you want to save these changes to the published diary?

Comment Preferences

  •  Tip Jar (11+ / 0-)

    "When they ask how I died, tell them....still angry."

    by quell on Mon Aug 27, 2012 at 04:11:29 PM PDT

  •  hope you're right... (2+ / 0-)
    Recommended by:
    ericlewis0, Lujane

    "It is horrifying that we have to fight our own government to save the environment." *Ansel Adams* ."Even if you are on the right track, you'll get run over if you just sit there."*Will Rogers*

    by Statusquomustgo on Mon Aug 27, 2012 at 04:31:17 PM PDT

  •  I'm guessing (9+ / 0-)

    That this was a draft.  He pulled it because he didn't think it was ready for prime-time.  He said he would get it out that week, but other priorities have intervened.

    If I'm right, you do him a dis-service by posting it here.

    Numbers are like people . . . Torture them enough and they'll tell you anything.

    by Actuary4Change on Mon Aug 27, 2012 at 04:40:27 PM PDT

  •  If he said it's not done, why post it? (4+ / 0-)

    Because it doesn't appear to be anywhere near ready... I don't think this was suppressed, I think it's incomplete.  And I think it's crappy of you to post his half-finished work.

    •  because that is a lie. (0+ / 0-)

      he posted it, he made the tiny url for twitter, and then he yanked it.
      we were talking about it at PEC.
      Olav found a cached view, and voila! the mirror site got wiped.
      its pretty strange.

      I made several comments asking for the rewrite like Nate PROMISED within the week...
      all dseleted.
      im within a hairsbreadth of getting banned at 538.
      just ax Micah.
      /wicked grin

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 06:55:16 PM PDT

      [ Parent ]

    •  heres the PEC link. (1+ / 0-)
      Recommended by:
      Sandino

      75 days

      sowwy, but I think Nate Silver has been captured by market forces.
      If the horserace ends too soon, he don't get paid.
      Who do you think made him yank the post?

      and, Olav found this link to a mirror site.
      http://finance.xyrm.com/...

      its purged now.
      but heres another.
      http://walmartcommunityvotes.com/...

      can't stop the signal.

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 07:00:15 PM PDT

      [ Parent ]

    •  every comment i made at 538 asking about it (0+ / 0-)

      got deleted.

      he is NEVER going to post it.
      unless someone calls him on his bs.

      Nate Silver is the new Jonah Lehrer.

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 07:02:18 PM PDT

      [ Parent ]

  •  I think it may still fall under (3+ / 0-)
    Recommended by:
    lineatus, Actuary4Change, pdx kirk

    copywrite rules though. Probably not a good idea to run the whole thing like that.

    Most of the people taking a hard line against us are firmly convinced that they are the last defenders of civilization... The last stronghold of mother, God, home and apple pie and they're full of shit! David Crosby, Journey Thru the Past.

    by Mike S on Mon Aug 27, 2012 at 05:23:08 PM PDT

    •  i didn't run the whole thing. (0+ / 0-)

      i left out several paras.

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 06:51:29 PM PDT

      [ Parent ]

      •  Fair use typically means 3 grafs. nt (0+ / 0-)

        Most of the people taking a hard line against us are firmly convinced that they are the last defenders of civilization... The last stronghold of mother, God, home and apple pie and they're full of shit! David Crosby, Journey Thru the Past.

        by Mike S on Tue Aug 28, 2012 at 07:56:38 AM PDT

        [ Parent ]

        •  is that a rule? (0+ / 0-)

          i didnt include the whole post.
          but you need to understand, i included the salient content because Im making a mirror of the original post, in case all sources mysteriously disappear like the last one, and indeed, like the original.

          "When they ask how I died, tell them....still angry."

          by quell on Tue Aug 28, 2012 at 08:34:25 AM PDT

          [ Parent ]

          •  Yes, it is a rule: (0+ / 0-)

            http://www.dailykos.com/...

            Copying and pasting complete copyrighted articles without permission from the copyright holder is absolutely prohibited by both this site's policies and copyright laws. Copyright infringement can expose both you and the site's owners to financial liability. Just don't do it. And if you see someone else doing it, please politely ask them to edit their diary accordingly. This is a bannable offense.
             6. Limited copying within the bounds of the doctrine of "fair use" is permitted. A reasonable rule-of-thumb is that copying three paragraphs from a normal-length news article or editorial is acceptable. (This, however, is not a safe-harbor. If even three paragraphs seems like "too much," then copy less or nothing at all.) For more on fair use, please visit this site.

            It doesn't matter now, I think. There were enough comments that a front pager must have read it and since they didn't tell you to edit  you seem safe.

            Most of the people taking a hard line against us are firmly convinced that they are the last defenders of civilization... The last stronghold of mother, God, home and apple pie and they're full of shit! David Crosby, Journey Thru the Past.

            by Mike S on Wed Aug 29, 2012 at 08:16:37 AM PDT

            [ Parent ]

  •  Yeah, like NYT of all papers would be rooting for (0+ / 0-)

    Republican pollsters.

    "Let's put the jam on the lower shelf so the little people can reach it." - Ralph Yarborough

    by Zutroy on Mon Aug 27, 2012 at 06:45:57 PM PDT

    •  not for repub pollsters (0+ / 0-)

      for the Holy Horserace.

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 07:03:30 PM PDT

      [ Parent ]

      •  i htink nate's pretty clear (1+ / 0-)
        Recommended by:
        lineatus

        that teh race is NOT horse-related in any sense...

        he constantly pushes the idea that it's obama's race quite decidedly at this point in the game.

        i think posting this here and thinking some grand scheme is afoot is a tad bit of a    s t r e t c h.

        Blessed are the cheese-makers?!

        by pholkhero on Mon Aug 27, 2012 at 07:14:06 PM PDT

        [ Parent ]

        •  Nate is paid (0+ / 0-)

          to put artificial uncertainty into the race.....to keep hope alive.
          my comments are deleted at 538.
          the 538 mirror site was wiped.
          http://finance.xyrm.com/...

          read the PEC thread.
          http://election.princeton.edu/...

          "When they ask how I died, tell them....still angry."

          by quell on Mon Aug 27, 2012 at 07:20:01 PM PDT

          [ Parent ]

          •  LOL. Take it easy there, big fella... (1+ / 0-)
            Recommended by:
            lineatus

            You're wading deep into CT land. None of those sites corroborate any notion of Nate being "paid to put artificial uncertainty into the race.....to keep hope alive." The mirror site was wiped for the same reason the main site was. The article was an unfinished draft.

            Hope alive for whom? Dems? The GOP? What specifically about his model (I'm talking specific statistical variables here, not some wild inferrence of yours) is skewing his numbers away from the truth?

            Regarding the last thread you linked, the only person suggesting what you're suggesting is a lone nutter who offers as little factual leads as you. Of course, he offers plenty of motive (the CT brain always goes in the wrong direction to form its conclusions), but everything else is BS.

            "Let's put the jam on the lower shelf so the little people can reach it." - Ralph Yarborough

            by Zutroy on Mon Aug 27, 2012 at 07:37:44 PM PDT

            [ Parent ]

            •  PEC (1+ / 0-)
              Recommended by:
              Sandino

              http://election.princeton.edu/...

              im not the only one smelling the skunk.

              and im a grrl.

              "When they ask how I died, tell them....still angry."

              by quell on Mon Aug 27, 2012 at 08:57:16 PM PDT

              [ Parent ]

              •  Oooooooooh. Two's a crowd, eh? (0+ / 0-)

                I'm sure I can find two people who think all public figures are shape-shifting reptilians too.

                How about documenting a direct link between Nate Silver and the Electionati or whoever, instead of hearsay and random anecdotes used to feed a confirmation bias?

                "Let's put the jam on the lower shelf so the little people can reach it." - Ralph Yarborough

                by Zutroy on Mon Aug 27, 2012 at 09:43:22 PM PDT

                [ Parent ]

                •  sure there is a direct link. (0+ / 0-)

                  Its called Nate Silver's paycheck from the NYT.
                  Do you honestly doubt the media isnt invested in the horse-race narrative?

                  Nate was awesome in 2008.
                  I think this year he's been captured by market forces.
                  Look, all he has to do to disprove my thesis is post it like he said he would.  A lot of the regulars were asking for a post on "oversampling" because everytime Sully or Politico links 538 we get swamped by low information conservative trolls.
                  "dem oversampling", "Bradley effect", that nutty CU poly-sci professors BS analysis and voter suppression laws are main reasons the trolls are supersure Romney is going to win in a landslide.
                  Nate has debunked all the above with posts, except for "dem oversampling".  And actually he debunked that too, but someone made him pull it.

                  Im taking a page from Julian Assange's data model.
                  Im using Kos as a mirror site to discuss oversampling mythology.  Because Nate wont allow it to be discussed at 538.

                  "When they ask how I died, tell them....still angry."

                  by quell on Tue Aug 28, 2012 at 06:09:57 AM PDT

                  [ Parent ]

  •  where's the uncertainty (0+ / 0-)

    when for MONTHS now, it's showed obama increases his win percentage over time, increasing his EC count since june.  again, SHOW where the uncertainty he's supposed to be injecting is at in his posts or his numbers??

    your post would have more credence if Nate actually was showing any sort of uncertainty, but he doesn't.

    my guess is ur comments get deleted b/c you keep posting the same question that's barking at shadows...i believe that's called "trolling."

    Blessed are the cheese-makers?!

    by pholkhero on Mon Aug 27, 2012 at 07:29:19 PM PDT

    •  glad to. (1+ / 0-)
      Recommended by:
      Sandino

      One way Nate increases uncertainty is by using Rasmussen in his polling averages.  Another is reliance on a-priori economic data.
      There is no mapping of economic indicators from the past onto this race, because no matter how bad the economy gets, black people arent voting for Romney.
      Ditto browns, youth, college-educated and women.

      Do you know why Ras skews the polling?
      He robopolls.
      31.6 % of american households are cell-only, and they go Obama by 20 points.  But they dont get polled.
      If the race is over, Nate is irrelevant.
      That is why hes adding artificial uncertainty.

      The vanished post debunked conservative claims of dem oversampling in polls where Obama is ahead.
      If that is true, like Nate said in his post, then Obama is actually 6 or 7 points ahead-- just like 2008.

      What if....Obamas win probability is really +80% and that gets out?  Does Romney tank on Intrade?  Does Adelson shift his money into downticket races?  Do the super pacs quit pouring money into Romney if they think he cant win?
      Dr. Wangs (PEC) prob of win is 87%.  Why is there a difference if PEC and 538 are using the same poll data?
      Artificial uncertainty.

      "When they ask how I died, tell them....still angry."

      by quell on Mon Aug 27, 2012 at 09:12:42 PM PDT

      [ Parent ]

  •  AND (1+ / 0-)
    Recommended by:
    Sandino

    I will gladly stfu if Nate puts up the oversampling post.....like he promised.

    An earlier post in this space about poll oversampling was published in error and will be updated and published later this week.
    In the meantime I don't think he can wipe a Kos diary.

    "When they ask how I died, tell them....still angry."

    by quell on Mon Aug 27, 2012 at 09:17:40 PM PDT

  •  If the "horserace" meme is sustained and boosted (1+ / 0-)
    Recommended by:
    quell

    by those with the power to do so... it begins to look like their fallback strategy is just to keep the base turnout strong so they don't lose ground in the Senate if not kissing their slim chance of taking it goodbye and head off  losing the house too...

    If a Romney win is looking really unlikely... even with all the electoral flim flamming and piles of money they still want to keep their ALEC agenda alive... state and national and that means enough tools in place to stymie any return to or even just inching back towards moderate if not mildly progressive policy and legislation....

    But at some point the horse race meme will just backfire if pushed beyond the credibility line, Outside the die-hard right wing echo chambers people who can see what side their bread is buttered on, Republican moderates, independents and normally stay at home Democrats will be panicked about a Romney win and come out and vote... and turn the "horse race" into a Republican rout...

    Pogo & Murphy's Law, every time. Also "Trust but verify" - St. Ronnie (hah...)

    by IreGyre on Tue Aug 28, 2012 at 04:45:48 AM PDT

    •  I believe this already happening. (1+ / 0-)
      Recommended by:
      IreGyre

      And that is why someone told Nate to yank that post and why he hasnt replaced it.
      All conservatives have is enthusiasm to beat demographic evolution.
      The election is already over over, it just depends on turnout.
      And not on republican turnout, because that is already maxxed.
      This election depends on turning out democrats.
      If all the polling (except Rasmussen) shows Romney losing by a widening margin the conservative base will lose enthusiasm, and Romney's funding will move down ticket.
      Even Sheldon Adelson isnt going to keep throwing good money after bad if Romney begins to look like a sure loser.
      PEC has Obama's probability of a win at 87%, and Dr. Wang's model uses the exact same poll data as Nate's model does.

      "When they ask how I died, tell them....still angry."

      by quell on Tue Aug 28, 2012 at 05:57:43 AM PDT

      [ Parent ]

      •  correction (0+ / 0-)

        PEC's model uses the exact polling data that 538's Nowcast does.
        So why is there such a gap?
        PEC:  Obama's win prob 87%
        538:  Obama's win prob 71.3%

        My hypoth is that Nate adds artificial uncertainty with Rasmussen and economic a-prioris to promote the horse-race narrative.
        Because Romney having a 1 in 3 chance looks a lot more possible than a 1 in 5 chance.

        "When they ask how I died, tell them....still angry."

        by quell on Tue Aug 28, 2012 at 06:26:33 AM PDT

        [ Parent ]

    •  and im not the only one that (0+ / 0-)

      thinks the race is already over.
      Michael Tomansky does too.

      So does PEC.

      All the statnerds know the race is over for Mitt, barring a black swan event.

      "When they ask how I died, tell them....still angry."

      by quell on Tue Aug 28, 2012 at 06:42:57 AM PDT

      [ Parent ]

Subscribe or Donate to support Daily Kos.

Click here for the mobile view of the site