-Caveat Lector-
Errors Plagued Election Night Polling Service
By Howard Kurtz
Washington Post Staff Writer
Friday, December 22, 2000 ; Page A01
The polling organization responsible for the biggest blunder in
television history on election night was plagued by a series of
errors that distorted the Florida vote all night long, a
confidential report concludes.
The internal investigation by Voter News Service also makes clear
that its techniques were inherently risky for the networks that
rely on its data.
The group had no reliable way of estimating the number of
Florida's absentee ballots in the presidential race, which were
almost double what it had expected. What's more, the news
service dramatically underestimated the number of Florida votes
still uncounted at 2 a.m.
"Budget limitations" have "placed heavy burdens on all VNS staff
and [have] made the task of covering elections far more difficult
than necessary," says the report, a copy of which was obtained by
The Washington Post. VNS was created in 1990 as a cost-cutting
measure by the major television networks and the Associated
Press.
While CBS, NBC, ABC, CNN and Fox decided to project Vice
President Gore and, six hours later, George W. Bush the Florida
winner -- and had to retract both calls in humiliating fashion --
those decisions were based heavily on bad VNS data.
This was more than just a media embarrassment. The calling of
Florida for Gore gave many viewers the impression, especially
after the vice president won Michigan and Pennsylvania, that he
was on his way to the White House, a situation Republicans say
may have discouraged some Bush voters from turning out. The
later projection that Bush had won Florida fostered a national
mind-set that he had been elected president, which Gore
supporters say made their recount battle that much harder.
VNS spokeswoman Lee C. Shapiro said she could not comment on the
inquiry.
At 7:50 p.m. on Nov. 7, when the network calls for Gore began,
VNS was wildly underestimating the size of Florida's absentee
vote. The group thought absentee ballots would make up 7.2
percent of the overall vote, instead of the actual figure of 12
percent.
VNS also projected that absentees would vote 22.4 percent more
for Bush than Election Day voters, when the actual figure was
23.7 percent. That mistake alone accounted for 1.3 percentage
points of the 7.3 percent lead that Gore was projected to hold at
that moment.
VNS was operating in the dark because the organization did no
telephone polling in Florida to try to estimate the size and
shape of the absentee vote, largely because of "the very
considerable costs" involved. The group did such surveys in
California, Oregon and Washington because of the traditionally
heavy absentee balloting in those states.
"The absentee vote has been growing over the years, and we have
had to deal with it in a patchwork method," the report says.
Another 2.8 percentage points of Gore's projected lead was
inflated by problems with the exit polls, specifically the
sampling of voters in the group's 45 selected precincts. The
report says this degree of error is "within the normal range" for
exit polls.
The remaining 3.2 percentage points of the Gore lead were due to
flaws in the exit poll "model" itself. One of VNS's key
techniques is to compare its exit-poll findings to the results of
past elections.
VNS says it used Florida Gov. Jeb Bush's 1998 victory as the
best predictor of how his brother would fare this year, but that
Robert J. Dole's 1996 bid -- more voters turn out in a
presidential year -- would have produced a better estimate.
There also "may be errors in the past vote file for the 1998
gubernatorial race," the report says.
Finally, VNS uses raw vote totals to help correct any exit-poll
errors. At 7:50, the exit poll in Tampa was off by 16 percentage
points, inflating Gore's estimated lead. But Tampa and Miami,
which had the biggest overstatement of Gore's lead in the exit
polls, had not reported any votes at 7:50, leaving VNS unable to
modify its errors.
If any one of these four mishaps had not occurred, the report
says, VNS might not have called Florida for Gore. While some
"bad luck" was involved, says the report by editorial director
Murray Edelman, the networks also bear responsibility for making
projections without consulting VNS.
"It would appear that calls are being made at the minimum
acceptable tolerances for risk, with very little allowance for
error," he writes. "If we are to continue in this manner, our
decision procedures must be redesigned."
The network projections that Bush had won Florida, and with it
the presidency, began at 2:16 a.m. They also were based on bad
VNS numbers (although neither the news service nor the AP
declared Bush the winner).
At 2:10, with 97 percent of the state's precincts reporting, VNS
estimated that there were 179,713 votes outstanding. In fact,
more than 359,000 votes came in after 2:10. In Palm Beach County
alone, VNS projected there were 41,000 votes outstanding, but
129,000 votes came in.
This was compounded by local problems with reporting the vote.
At 2:08, Gore's total in Volusia County mysteriously dropped by
more than 10,000 votes, while nearly 10,000 votes were added to
Bush's total. This mistake boosted Bush's lead by 20,348 votes,
giving him a 51,433-vote lead over Gore -- or so VNS believed.
Brevard County later increased Gore's total by 4,000, with none
for Bush, in what appeared to be a correction of an earlier
mistake. Given the tightness of the contest, Edelman writes, "I
was very concerned to see the race called by the networks."
Among other problems, VNS's quality-control system was so
inadequate that it failed to reject an early report that 95
percent of Duval County had voted for Gore.
And thanks to exit-poll samples that are smaller than the
networks used before VNS was created, "there is some evidence
that we overstated the size of the black vote and underestimated
the size of the Cuban vote in Florida, and both of these errors
could have contributed to the overstatement of the vote for
Gore."
Network executives and anchors have repeatedly apologized for
their election night mistakes and launched internal inquiries.
ABC says it will insulate its decision desk from competitive
pressures and describe all future projections as estimates. Fox
says it will probably drop VNS and start a new polling
consortium.
=================================================================
Kadosh, Kadosh, Kadosh, YHVH, TZEVAOT
FROM THE DESK OF:
*Michael Spitzer* <[EMAIL PROTECTED]>
~~~~~~~~~~~~~~~
The Best Way To Destroy Enemies Is To Change Them To Friends
=================================================================
<A HREF="http://www.ctrl.org/">www.ctrl.org</A>
DECLARATION & DISCLAIMER
==========
CTRL is a discussion & informational exchange list. Proselytizing propagandic
screeds are unwelcomed. Substance�not soap-boxing�please! These are
sordid matters and 'conspiracy theory'�with its many half-truths, mis-
directions and outright frauds�is used politically by different groups with
major and minor effects spread throughout the spectrum of time and thought.
That being said, CTRLgives no endorsement to the validity of posts, and
always suggests to readers; be wary of what you read. CTRL gives no
credence to Holocaust denial and nazi's need not apply.
Let us please be civil and as always, Caveat Lector.
========================================================================
Archives Available at:
http://peach.ease.lsoft.com/archives/ctrl.html
<A HREF="http://peach.ease.lsoft.com/archives/ctrl.html">Archives of
[EMAIL PROTECTED]</A>
http:[EMAIL PROTECTED]/
<A HREF="http:[EMAIL PROTECTED]/">ctrl</A>
========================================================================
To subscribe to Conspiracy Theory Research List[CTRL] send email:
SUBSCRIBE CTRL [to:] [EMAIL PROTECTED]
To UNsubscribe to Conspiracy Theory Research List[CTRL] send email:
SIGNOFF CTRL [to:] [EMAIL PROTECTED]
Om
