[Nettime-bold] [ot] [!nt] \n2+0\ nietzsche !zt dead

[integer@www.god-emil.dk]

this travelled across another list.



>From: "David Funder"
>Subject: Statistics and the Florida fiasco
>Date: Thu, 9 Nov 2000 09:48:40 -0600
>
>Dear list members,
>I thought you might be interested in the following analysis, done by
Dr.
>Leslie Eaton, a former student of mine.  She noticed that in Florida
the
>more votes Bush got in a county, the more votes for Buchanan, whereas
the
>more votes Gore got, the less for Buchanan.  (This is probably because
Bush
>counties tend conservative and Gore counties tend liberal).
>
>But if you include Palm Beach, suddenly the correlation comes out
showing
>pro-Gore counties tending to be pro-Buchanan!  This not only makes no
sense
>politically, it is inconsistent with the pattern if you remove this one

>county from the analysis.  That is, remove Palm Beach and again you see

>Gore
>totals correlated with lower votes for Buchanan.
>
>This might be a great teaching example for a stat class on regression;
also
>of course it seems rather important for the future of the country at
the
>moment.
>
>Leslie's analysis follows:
>.  If you
>haven't been following the story, Palm Beach did a horrible job
>constructing
>their ballots.  Several residents of the county have complained that
the
>ballot was confusing and they accidentally voted for Buchanan when they

>meant to vote for Gore.
>
>So, I put that premise to the statistical test.  In all likelihood they
are
>correct.  Here's some evidence.  Using the percentage of votes, per
county,
>for Bush, Buchanan, Gore, & Nader:
>
>The correlation between Bush/Buchanan votes, R(b,b) should be positive.

>The correlation between Gore/Buchanan votes, R(g,b) should be negative.

>
>For the great State of Florida: R(b,b) = .27 R(g,b) = -.29
>
>Now, that pattern should hold for both the counties that predominantly
>voted
>in favor of Gore AND those that predominately voted in favor of Bush.
>
>Counties in favor of Bush: R(b,b) = .11 R(g,b) = -.13
>
>Counties in favor of Gore: R(b,b) = -.21 R(g,b) = .18
>
>Why the reversal in the signs of the correlations?  Because Palm Beach
was
>a
>county in favor of Gore.  If you re-calculate the correlations for
Gore,
>removing Palm Beach, you get:
>Counties in favor of Gore: R(b,b) = .10 R(g,b) = -.14
>
>So, without Palm Beach, all the correlational patterns hold, the
pattern
>for
>the State, the pattern for the counties in favor of Bush, and the
counties
>in favor of Gore.  Palm Beach is a statistical outlier and a BIG one.
>
>Based on the voting patterns for the great State of Florida (via
>regression), Gore was likely robbed of approximately 2,287 votes (or
that
>many votes were aberrant for Palm Beach County).  Gore may have been
robbed
>of the approximately 1700 votes he needed from Florida just because of
a
>poorly designed ballot.
>

--
William MacAllister

"Human beings, who are almost unique in having the ability to learn from
the experience of others, are also remarkable for their apparent
disinclination to do so." -douglas adams


"Battle not with monsters lest you become one." - nietzsche






nn lookz at nietzsche - battle not with life lest you dead become.


nietzsche at lookz nn - [nn gave h!m l!f] flap flap


nn lookz at nietzsche - spontaneous infringement of my symmetry


nietzsche at lookz nn - battle not with life lest you dead become


nn lookz at nietzsche - i want 2 live with you


nietzsche at lookz nn - m9ndfukc.macht.ganz.gluck!ch+fre!


nn lookz at nietzsche - nietzsche !zt dead  [run!ng aua! u!th mad abandon]



242.zkueek++







kuot!ng 01 !mage

dreamz = 2 dze m9nd
uat 01 karess = 2 01 m9ndz 
bod!

kl!k


nn






                                            pre.konssept!*n
                                                  meeTz ver!f1kat!*n.
 
 
 
  -
 
  Netochka Nezvanova    -  eksplod!ng galax!ez + !nflaz!onar! un!versez
  f3.MASCHIN3NKUNST
  @www.eusocial.org
  17.hzV.tRL.478
                                                      e
                                                      |
                                                       |  +----------
                                                      |  |     <
                                     \\----------------+  |  n2t^P
                                                         |       
                                                         e








_______________________________________________
Nettime-bold mailing list
Nettime-bold@nettime.org
http://www.nettime.org/cgi-bin/mailman/listinfo/nettime-bold