Monday, May 2, 2016

Nate Silver's 2008 debunking of exit polling which supposedly "overstated Obama's performance in the primaries [against Hillary] by an average of 7 points"

   
   In 2008, Nate Silver did a reasonable job predicting the actual results in different states, combining and assessing the work of other pollsters (he doesn't do polling, himself).  He was then lionized in the commercial press, and published a column in the New York Times.  On November 4, 2008, on 538 blog, he wrote a flip, self-serving “Ten Reasons why you should ignore exit polls” here.  A few exaggerate actual difficulties with exit polling.

***

        But his point 2 states, as if it were obvious why this should be the case, that "Exit polls have consistently overstated the Democratic share of the vote.” For the obvious alternate explanation, however, which Silver does not consider, is that exit polls are comparatively accurate;  in contrast, the vote recorded on electronic voting machines which leave no paper trail and have a secret, proprietary code that not even a state or county government, let alone  a committee of independent experts can examine, are not.   Silver then offers, against without so much as a comment, the Ohio, Pennsylvania and Florida “results” in 2004  – all 6-8 points off toward Bush (far beyond the margin of error, a vast statistical unlikelihood). to challenge exit polling. 

***

      His point 3 bald-facedly states:  Exit polls were particularly bad in this year’s primaries.(2008)  They overstated Barack Obama’s performance by an average of about 7 points. 

        But were they?

***

     And are the exit polls in 2016 which show that Sanders won Massachusetts, Illinois and came close in New York and Ohio also off?  See here.


Charnin Chart

Chart courtesy of Richard Charnin.

***

      Silver assumes that there is and can be no electoral fraud in the United States. But that is a curious and unargued - anywhere in the corporate media - assumption... 

***

       If right, Silver should campaign for the State Department to correct its use of exit polls to test the fairness of elections in a minimum of 14 cases abroad…See the State Department/AID report "Vote Count Verification: a User's Guide for Funders, Implementers and Stakeholders," pp. 52-54 here and my analysis of it here.  For instance,  Bjornland and Cowan, the authors, state:

       "In recent years, domestic and international organizations have increasingly turned to exit polls to verify the officially reported results in the transitional elections of emerging democracies. Outside observers have credited exit polls with playing a key role, for example, in exposing fraud in Serbia and Mexico in 2000, Georgia in 2003, and the Dominican Republic and Ukraine in 2004.  U.S.- funded organizations have sponsored exit polls as part of democracy assistance programs in Macedonia (2002), Afghanistan (2004), Ukraine (2004), Azerbaijan (2005), the West Bank and Gaza Strip (2005), Lebanon (2005), Kazakhstan (2005), Kenya (2005, 2007), and Bangladesh (2009), among other places."
  
***

       Note: if Silver were right, there would be no test of fairness of elections even in cases where theft of elections is obvious…

***

       A baseball statistician, it is quite possible that Silver does not know about State Department policy and of course, reading the Times would be no help…

***

     In 2012, Richard Charnin offered 25 rejoinders. See here and here.  He provides reason to suspect that many Republican "victories" have come not mainly because of special funding by the .0001%, but because of wide exit poll/machine deviations.  These are particularly stunning in the actual Obama elections, for instance, exit polling in November, 2008, recorded Obama with a 60% to 40% victory over McCain.  If right (and I should add, if right in one-tenth of the cases Charnin discusses), the bipartisan - and corporate media - betrayal of American "democracy" is shocking. See, for instance, here and here.  

      Silver has not deigned to reply…

***


      Nate also does baseball statistics.  It is as if many series were the Chicago Black Sox of 1919 and he reports often crooked results with a touching faith...

1 comment:

Sharukh Khan said...




usa foreign education consultancies in hyderabad do you want to study in abroad today or in the next intake. we are the best and top rated study abroad consultancies in india with good visa assurance.we help you in filing the study abroad consultants in hyderabad f1 visa for you in very less time. we are also help you with information needed to apply for the college university.

Post a Comment