--NETWORK NEWS HEADS in the DOCK--
TAUZIN
HEARINGS CONFUSED BY RED HERRINGS & ELECTION WOES
by Michael
Hammerschlag
TABLES + TIMELINES COMMERCE
COMMITTEE TESTIMONY CBS REPORT
HAMMERHOME CNN
REPORT
COMMERCE
COMMITTEE LINKS WHOOPS- HOW THE
NETWORKS GOT IT WRONG ABC REPORT
CAST
CNN: CNN Report TAUZ: Billy Tauzin- R-La. JON: Tom Johnson- CNN 
CBS: CBS Report WAT: Ben Wattenberg- AEI- LAK: Andrew Lack- NBC
ABC: ABC Report RIS: James Risser- Stanford U AIL: Roger Ailes- Fox
BIEM: Paul Biemer-RTI KON: Joan Konner- Columbia U HEY: Andrew Heyward- CBS
SAV: Ted Savaglio- VNS WES: David Weston- ABC BOC- Lou Boccardi- AP
CNN REPORT by WAT, RIS, + KON = my comments
For the first time since May 1985, network news presidents
were grilled by members of Congress, rather than the reverse, over the
appalling mistakes and miscalls on Election night. (TV liability hearings were
held after previous early calls in 1985*, 1981, and 1965). The independent CNN
report on the fiasco starts vividly: "On Election Day 2000,
television news organizations staged a collective drag race on the crowded
highway of democracy, recklessly endangering the electoral process, the
political life of the country, and their own credibility, all for reasons that
may be conceptually flawed and commercially questionable.” [Testimony is not
repeated in order- this is an attempt to provide a jigsaw narrative from a
massive body of information. Quotes are implied on all individual testimony.]
WAT: By calling
Florida wrong twice -- first for Al Gore, then for George W. Bush -- television
networks hit an avoidable iceberg that had been on their radar screen for many
years.
RIS: All of
this for no good reason. Americans are much more interested in having the
election returns reported accurately than they are in whether one network comes
in a few minutes ahead of the other in reporting the results. Very few people
know at the time which network is coming in first, and even fewer of them would
remember today… We question what purpose this hyper-competition serves, either
journalistic or commercial.
AIL: In the closest race in history, the wheels came
off a rattletrap computer system we relied on.
KON: CNN executives, correspondents, and producers themselves described election night coverage as "a debacle," a "disaster" a "fiasco."
FIRST AMENDMENT
The first question was whether the Committee had the
right to demand their presence, especially when colored by the dubious
Republican proposition that the networks had deliberately delayed reporting
Bush states. In capsule conduct: Ailes was pugnacious + funny, Boccardi
resentful, Weston uncomfortable, Heyward somewhat arrogant, Konner curt, Lack
ingratiating, Biemer overly academic, Wattenberg entertaining, and Johnson
chagrinned and most responsive.
BOC: AP has serious doubts that the Committee and its
staff, no matter how sensitive they may be, can avoid crossing the line between
appropriate government concern with the electoral process itself and, on the
other hand, inappropriate government involvement with the reporting on that
process by a free press. … To put it more plainly, we believe that such an
official government inquiry into essentially editorial matters is inconsistent
with the First Amendment values that are fundamental to our society. We agree
that there were serious shortcomings -- call them terrible mistakes -- in the
election reporting … and that these mistakes cannot be allowed to happen again.
But fixing them is a job for the nation's editors and news directors, not its
legislators. What we report and when we report it are matters between us and
the audience we try to serve, not matters between us and our Congressman.
GREENWOOD (R-PA): “This
member is uncomfortable grilling you.”
AIL:
I am deeply disappointed that this is being handled as an investigative
and not as a legislative fact-finding matter …(and) that this committee regards
its role as adversarial, requiring us to take an oath as if we have something
to hide. We do not… Everything our organizations did on election night was done
under the protection of the first amendment, which may become more relevant as
these discussions continue.”
TAUZ: “The
Commerce Committee has heard 178 witnesses, all under oath- it has no meaning
other than investigative meetings are taken under oath. VNS modeling is
seriously flawed: it underestimated poll numbers in 32 states for Bush and
..only 15 states for Gore. They
overestimated 15 states for Bush and 34 states for Gore.
DINGELL: “It
appeared by many that this inquiry was an attempt by the Republicans to shift
attention from election problems in Florida that cost Gore the Presidency… For
good motives and bad, elections have been called wrong by the media for a long
time.” =The 74 year old Dingell
reflected on the Alf Landon/Roosevelt and Dewey/Truman miscalls that he’d lived
through.
DELIBERATE BIAS for GORE
CNN: Tauzin said the subcommittee's "analysis of the networks' election night ‘victory calls' indicates an incontrovertible bias in the results which were reported." The networks "consistently reported Vice President Gore's victories earlier than Governor Bush's victories, portraying a skewed electoral picture and disenfranchising many American voters," he said.
STUPAK (D-Oh): You (Tauzin) seem to make a premature call
(of bias) just like the networks did on election night.
Impeachment Manager (prosecutor) Steve Buyer (R-In)
was the most aggressive and relentless in pushing the Repub hypothesis that a
cabal of liberal outlets had corruptly thrown Fl to Gore early and delayed
reporting Bush wins- incredible, since the greatest television mistake in
history had just happened because they reported results too fast; here
Repubs were attacking media for reporting results too slowly.
BUYER: We live in a republic, not a democracy (as
evidenced by Fl). In Indiana, it’s not even debated whether there’s a
bias in some of the news… so when someone acts irresponsibly, it’s an
invitation to government intrusion. You have an editorial problem!
SAV: The notion that some kind of political bias
enters our work is simply without foundation, and … that appears to be common
ground among those of us here today.
WAT: Many
people would say the networks are too liberal.
Mass. Rep. Edward MARKEY devastated that
theory: (The question is) whether early calls were a vast left-wing conspiracy.
Were dozens of network journalists, the staff of VNS, all network news
directors, a score of network anchors, and the President’s first cousin all
co-conspirators in an intricate plot to call key states early for Gore? It is,
of course, preposterous. Their only bias was their desire to be first – this
competitive urge which blurred their own judgment.
=Republican concerns were based on a few Bush
states: Ga, Al., NC, Co., Va that ended with sizable Bush leads, but were
called ½ hour to an hour after poll closing by the networks. http://com-notes.house.gov/bush_nine.pdf (only lists the later VNS call times). The
striking thing was that VNS didn’t lead the networks - the networks called
races before VNS (up to several
hours) or tied them in almost every instance, inc. Fl for Gore, except Wash.
and Wisconsin, so any bias was questionable. CBS/CNN explained that the exit
polls were terrible in those states.
CNN: Exit Poll Final Result
Alabama Gore by 1.2 Bush by 14.9
Arizona Gore by 3.6 Bush by 6.3
Colorado Gore by 3.1 Bush by 8.4
Georgia Bush by 4.7 Bush by 11.7
North Carolina Gore by 3.0 Bush by 12.8
As you can see the exit
polls in these five states were off by between 7 and 16(!!!) points. We waited
for actual results in each of these five states and called them as soon as we
were sure that Bush would win. In North Carolina our call was 28 minutes after
poll closing; in Georgia it was 32 minutes after poll closing. In Alabama VNS
made in our view a very reckless poll closing call for Bush while Gore was
actually leading in the best survey estimate at that time. We believe that this
is the first time that a projection was made using an exit poll that actually
showed the other candidate ahead. We could not allow this call to go on
the air until we had seen some actual returns. At 8:25 – 25 minutes after poll
closing – Bush had achieved a lead and we called the race for CBS & CNN.
=CNN seemed to reap the most
enmity of the Republicans for it’s alleged liberalness- interesting, because it
was considered an inexperienced corporate mouthpiece when it started 2 decades
ago.
CBS:
On
some 20 occasions, based on the data before it, CBS News actually preceded VNS
in its calls for one or the other of the candidates. This year, VNS called 28
races at poll closing. CBS News differed from VNS at poll closing in two
states. On the other hand, CBS News called Louisiana for Bush at poll closing,
while VNS waited 21 minutes. The networks and VNS were far from being in
constant lockstep. In fact, throughout the night, the networks and VNS made 26
calls at different times. Ironically, VNS never called Florida for Bush, even
though all the networks did so, based largely on data from VNS. [see chart of
relative times of state calls by different networks in tables
and timelines]
‧‧‧‧
CBS News was the first to make
15 calls--including the first to withdraw the call for Bush in Florida. Three
of the first calls were made at the same time as other networks, but CBS News
was first, alone, in 11 calls. CBS was the second network to make 7
calls--including the second to retract the Gore call in Florida. (Because CBS
News shared its Decision Team with CNN, the timing of CNN’s calls was
essentially the same as CBS News’.)
‧‧‧‧
Fox had 8 first calls--9
when we add the Election Night call for Bush in Florida-- and 4 of the 9 were
made at the same time as another network. Fox was the second to call the race
in three states.
‧‧‧‧
NBC had 7 first calls--8
if we count the Gore call in Florida. Five of the first calls were made at the
same time as another network. There were 5 second calls--6 with the call for
Bush in Florida. One of those calls was made at the same time as another
network.
‧ ABC
had 2 first calls and 11 second calls--12 with the pullback
from Bush in Florida.
=CBS is quite proud that
they had the most first calls- in this case, though, the winners were probably
the losers.
ABC: ABC News never projected a winner in the presidential race in Oregon, which was resolved several days later, after the last of the absentee ballots was counted. In addition to the projections in the race for President, ABC News made 45 projections in races for the Senate and for Governor, all of which were correct. ABC News never projected a winner in the key Senate race in Washington. ABC News made 49 other projections of state races for President on November 7-8, and each one was correct.
EARLY
GORE FL CALL
Tauzin then played a 10 min. Repub prepared
videotape of election highlights (or lowlights), including Dan Rather’s
resplendent array of metaphors: “Bush wins, that’s it.. sip it, cup it, photostat
it, underline it in red, press it in a book, put it in an album, hang it on the
wall.” Nothing equivocal about Dan. The tape seemed to show an unstoppable
momentum for Gore, with talk of the paramount importance of Pa, Mi, Fl, and
Gore’s wins rolling in succession, the point being to show how the slightly
early call for Gore (7:48pm- 8-12 min. before polls closed in 10 W Fl counties)
caused Repubs to not vote across the country- a highly dubious proposition,
since Bush had been claimed to be ahead 3-6 points in polls- Gore wins might
pull Bush voters into the polls or make Gore people vote for Nader. 10
minutes before poll close, any voting voters are in a car, not watching TV.
LAK: Let me
respond by stating, unequivocally, that I am absolutely certain that political
bias played no role in NBC's election night reporting.
HEY: The
video that said Gore had it in the bag was misleading: 15 times between 7 and
11pm we displayed the popular vote graph- President Bush was ahead every
single time and on the electoral count- 75 out of 100 times.
WES: That was
a good editing job. (chuckles)
KON: It’s up to the voter to vote, period… that is their responsibility.
WITTS (R-Pa):
In my state people were still voting as late as 9 or 9:30.. thousands of Pennsylvanians
voted after the media called it. (Polls closed at 8 but people in line could
still vote)
Jane HARMAN (D-Ca): (Rep.)
Jim Corman lost because Carter conceded early in 1980 and no Democrats showed
up after. Since 1992, I’ve never been elected by more than 5%, so am extremely
sensitive to the impact of early election calls.
=Great debate exists over the
effect of early calls on remaining voters. Several studies were cited- this one
from 8-10pm PST in Oregon in the ’84 Reagan/Mondale race seemed to show no
effect:
CNN: Early TV verdicts of
Reagan’s victory did not depress turnout anywhere near the extent critics had
predicted: only 2.6 percent of the non-voters blamed TV for their failure to
vote – roughly less than one-quarter of one percent of the entire electorate.
Most non-voters had not heard the TV projections.
The
same conclusion was reached in eastern Oregon where a natural experiment
allowed contrasting turnout and time of voting in Malheur County (MST) with
that in Grant County (PST). The counties were demographically similar, but
Grant received an extra hour of projection news before the polls closed at 8:00
p.m. However, turnout after 5:00 PST was actually higher in Pacific-Time Grant
County than it was in Mountain-Time Malheur County.
KON: We found no convincing evidence that calls made
before polls are closed within a state or in another state have an impact on
voter turnout.
=Every news head agreed that there should be a
uniform poll closing and that they wouldn’t call a state before its poll
closed, but all intended to continue exit polling.
WAT: There are indications that that balance has
shifted and that commercial and business values are holding much more sway over
the behavior of journalists. On one hand they want to be accurate, on the other
they want to be fast. They are in a mode where they say ‘Stop me before I kill
again’. They can’t stop because of competition, so if there’s an outside force
that says, ‘You’re not going to get those ballots till later, they can go about
doing their job.’
BOC: You start down a (dangerous) road: What about
pre-election polling- if there’s a poll that says that this candidate is 20
points ahead, is that keeping people from the polls? Well, maybe we
shouldn’t do those.
CNN: Dec 15, 2000, Rep. Ed Markey (Mass.) introduced a bipartisan Uniform Poll Closing Act (H.R. 5678). This bill requires polls in all 50 states to close at 9 p.m. eastern standard time (8 p.m. central standard time, 7 p.m. mountain time) during Presidential elections. In order to set a 7 p.m. Pacific Time zone closing, daylight savings time would be extended for two weeks during Presidential election years, allowing for a West Coast poll closing of 7 p.m. Pacific daylight time.
VOTING FOUL-UPS
Denied any forum for investigating the multiple
egregious voting problems in Florida, Dems on the Committee used the network
hearings to vent their outrage at them, the decisive bad call for Bush, and
networks poor coverage, especially their reluctance to question outlandish
Republican assertions.
DEUTSCHE (D-Fl): If this was a foreign country, no one would
consider this a legitimate process. In 19 of 20 precincts in black
neighborhoods in Duval Co. 20% of the votes were thrown out. Research
showed you needed a 4th grade education to get through these
ballots.
SHERROD: Media has the responsibility to tell the people
more of these voter suppression stories.. like Gov Pataki as he repeatedly said
‘4 recounts had been conducted.’ Lugar said his state doesn’t do hand recounts
when a simple call to Indiana election officials proves they do. Media allowed
Lugar + Pataki and countless other Republicans to repeat this mantra
unchallenged. These statements resulted in a series of distortions, backed up
by a conservative corporate-owned media too lazy to scrutinize such allegations
and too eager to manufacture drama. The media have a responsibility to check
the facts. Refrain from the ‘he said, she said’ approach to news, which causes
politicians to exaggerate, to distort, and even to lie. Don’t accept what we
say- make us tell the truth.
DEUTSHE: The exit polling was probably more accurate than
the actual count. It was 50 to 75 thousand people net who thought they were
voting for Al Gore whose vote didn’t count. It is no longer debatable that if
the votes were counted, Al Gore would be President.
WAXMAN (D-Ca): The later rescinded late-night calls of Fl
for Bush were the basis of perceptions that George Bush was the winner and Al
Gore was the spoiler. It immeasurably helped Bush maintain the idea that he was
the man who won the election.. (and) set in motion a chain of events that was
devastating to Gore’s chances.
=NBC’s Andrew Lack hit every hot button issue in a
seemingly earnest mea culpa designed to spread blame.
LACK: Where was our reporting before Nov 7 about the impact of ineffective voting machines or confusing ballots? We have a system that might in fact be protective of felons that vote. I don’t know why we didn’t know much about automatic recount standards. We know if you’re registered to vote, it doesn’t mean you’ll be permitted to vote. If you’re in the military and mail in an absentee ballot, it doesn’t necessarily mean it’s been counted. We know if you’re poor, it likely means you will have a little more difficulty voting than if you are rich. Millions of votes are thrown out in election after election. Now that’s a story and there’s a screw-up… We booted it in more ways than one. I ask that you extend your focus beyond what the networks experienced, but look at the problems that the voters experienced. As a journalist… I wish I had.
VNS
BLUNDERS
SAV: Our (VNS) work involved
more than 40,000 people who staffed nearly 28,000 individual precincts and went
to some 4,600 counties to obtain the information that we needed. The plain fact
is that, despite our best efforts, the Voter News Service let down its members,
subscribers, and ultimately the American people, on Election Night 2000. = VNS has a $35 million operating budget.
MARKEY: I think a terrible set of mistakes were made at
VNS… We pay for pollsters as politicians, and if a pollster makes a mistake,
I’m going to make all these expenditures based on that erroneous polling. I
want to place the Queen of Spades squarely in front of VNS. (in Hearts
the Queen of Spades is fatal).
WAT: VNS, is, am, are the networks- they own it lock
stock and barrel.
CNN: Regarding the mistaken
call for Gore in Florida, VNS editorial director Murray Edelman reported to the
VNS owners that its analysis at 7:50 p.m. indicated a Gore victory in Florida
of 7.3 per cent. An accurate analysis would have shown a dead-even race.
Edelman determined that the 7.3 per cent error resulted from:
(1) Sampling of voters in
exit polls, 2.6 per cent.
(2) Use of the 1998 Florida gubernatorial
election to construct computer models for making vote estimates on Election
Night, 2.6 per cent.
(3) Estimation of the
absentee vote, 1.7 per cent. (VNS had estimated that absentee votes in
Florida would be 7.2 per cent of the total, whereas there actually was a 12 per
cent absentee vote, which was 23% more Republican)
(4) Quality control procedures working improperly, 0.4 per cent. This illustrates, Edelman reported, “that the bad estimate was the product of many factors all of which overstated the vote for Gore.” The “most unexpected effect” came from the choice of the 1998 race as the one for comparison. Later study showed that had the 1996 presidential race or the 1998 Senate race been used, “we would not have reached a call status in Florida” for Gore, he said. He said, “The odds are more like 1 in a million for a miss of 7.2 percentage points.”
CNN: Fluctuations, variations and contradictions had begun to appear in the basic system, including statistical models. From that point on, extreme caution should have been exercised in making any calls. Prudence demanded waiting for much more complete hard information.
=It is notable that
secretive VNS didn’t report the swapped exit poll numbers (between Gore and
Bush) in a Repub district we heard from a highly placed source that may have
led them to call the election for Gore. They never reported an alleged similar
mistake in the ’96 NH Senate race either, and with only 6 mistakes in 2000
elections, had a virtually perfect record to protect. Then again, it was the
Repubs on the warpath about the Gore call (with Cong hearings): arcane
statistical polling mistakes are far less actionable than accidentally
swapping numbers in a column. Or my source (or his source) may have confused
that mistake with one of the 4 or 5 other numerical counting mistakes in
Florida, but he reconfirmed the story in response to pointed inquiries.
That call remains a stunningly arrogant mistake on VNS
and the networks behalf: it was made with only 4% of the actual vote counted
and Bush, not Gore, was then ahead by 6%; and from only 45 Florida precinct
exit polls out of almost 6000 total.
CNN: The Decision Team determines
that the probability of a Gore win in Florida is within its guidelines for a
call, which means there is a 99.5 per cent probability of a Gore win (or)… only
a 1 in 200 chance that such a call could be wrong.
WES: We would be happy in a
world in which we don’t make any projections- in which there’s a uniform
poll closing time and a system of voting that is so reliable and instantaneous
that we can say, “here are the results”
CNN: (from RTI report) RTI noted that VNS had explained to the owners “that inadequacies in their budget have prevented them from making many needed improvements,” in such things as methodology, equipment and even adequate pay to hire qualified field interviewers. RTI warned that the standard errors reported on the VNS decision screens “tend to importantly underestimate the true total error in the estimates.” The information provided on the screens is so voluminous and complex as to be “prone to misinterpretation. Another area of concern is the absence of any direct quality control check on the interviewers' data collection activities (which was displayed in not catching bad mistakes). Exit poll non-response has become a “serious problem.” (those who refuse to answer). The average response rate in 2000 was 51 per cent, a drop from 55 per cent in 1996 and 60 per cent in 1992. The exit polls have consistently over-represented Democrats, which could reflect a bias from non-response.
AIL: When Republicans come out
of polls, if you ask them a question, they tend to think it’s none of your
business and Democrats want to ‘share their feelings’. (laugh)
CNN: “The exit poll is a blunt instrument,” said Mitofsky. Lenski (Decision Desk partners) said he believes the polls are getting less accurate. They noted that in eight states the final result did not turn out the way exit polls had indicated. They believe that they would have been derelict in their duties had they not made the two calls on Florida, based on the data supplied to them. The (CBS/CNN) Decision Team first alerted VNS to the Duval error, and the team complains that VNS did not promptly alert it to the Volusia error once it was found and processed into the VNS system. They believe that current VNS exit poll models are inaccurate in states where the absentee vote is more than 10 per cent.
WAT: Exit polls, whether accurate or not, are self-generated news. Their use by television networks to project election results is an attempt to forecast what is not yet known -- the actual vote count.
OTHER
RETRACTIONS: NEW MEXICO AND WASHINGTON STATE
CBS:
CBS News called New Mexico for Gore at 10:21 PM on Election Night, nearly an
hour and a half after the polls had closed there and with about half the
precincts included in the count. The call was withdrawn on Friday, November 10,
at around 3:00 PM. CBS News also called Maria Cantwell the winner in the
Washington Senate race at 12:52 AM, nearly two hours after the polls had closed
there. That call was retracted four hours later. After several weeks of
counting, the calls were confirmed. But, given the closeness of both races,
neither call should have been made.
In both
cases, the problem can be traced to bad information about which votes were
counted. In the case of New Mexico, there was a faulty report from Bernalillo
County, the state’s largest. The county reported to VNS on Tuesday night that all
but 2,000 absentee votes had been counted. Later, the county found some
software problems in the vote-counting program, so on Wednesday officials
removed 67,000 of the absentee and early votes for another count.
The
slow recounting in the next few days, accompanied by misplaced ballots and
accusations of partisanship, eventually dropped Gore’s lead to less than 200
with 1,800 votes yet to be counted, and the call was withdrawn. Gore was finally
certified on December 5 as the winner of New Mexico’s five electoral votes.
In
Washington State, the call was based on a combination of information and
assumptions: an exit poll of those who voted at their polling places on
Election Day, a telephone poll of those who voted by absentee ballot, and tabulated
votes from approximately 26 percent of the precincts and a large share of the
already-counted absentee vote. The CBS News Decision Team assumed that about
half of all votes (polling-place and absentee) were counted when CBS News
called the race and that Cantwell could safely be declared the winner. But, in
fact, only about 40 percent of the eventual total had been counted. That fact,
combined with an exit poll and a pre-election absentee (estimate) that were at
the outer limit of sampling error, made the election closer than it (appeared).
Cantwell was declared the winner in mid-Dec.
BUSH MISCALL
=VNS’s true crime may have
been holding the rotten numbers from the Volusia machine mistake (16,022 vote
drop for Gore, 2813 gain for Bush) for 2 to 4 hours (reported on County website
at 10pm and removed at 12, but copied onto Daytona Beach News-Journal till the next afternoon), then
inexplicably releasing them into the VNS system at 2:08. The crucial question-
why?- was never asked at the hearings.
A high source claimed that VNS phoned a Volusia official after 2am, but Volusia
spokesman Dave Byron is adamant that that didn’t occur. “No one from VNS or any
network contacted Volusia on election night to find out anything about this
number. .. All it would have taken is 5 minutes to get this genie back in the
bottle.” “We think..
VNS was picking it up from the Daytona News/Journal site that had taken it off
of ours," Volusia Supervisor Deanie Lowe explained. “The County stopped
updating their web site (removed Presidential race at midnight) and we were
left with the incorrect numbers till the next morning,” said the News/Journal’s
Matt Grimison, who “left (the Deland elections bureau) around 1:40am”. After sitting on these
numbers for hours, VNS had to know they were wrong- how could a 16,000 vote drop
not trigger human or computer alarms at VNS or the networks who all received
that same info? The corrupted Deland precinct was the last one reporting. The
19,006 vote erroneous net gain for Bush provoked the fateful Bush victory call only
8 minutes later, it was the difference between an indication and a
call. VNS presumes all precincts to be the same average size, which caused them
to massively underestimate the remaining vote in giant S. Florida counties.
The News-Journal also had the wrong
numbers listed through Jan for those errors (and the original 19,000 vote
mistake on their web site through Nov 8th or 9th), which
led me and others to initially report it as a 25,000 vote error, when the
County supervisor was unreachable. It was presumable that any error in a local
paper would be quickly corrected. Great confusion existed about vote changes,
because it depended on the time interval- a 19,000 mistake doesn’t cause
a 10,000 or 25,000 vote change, but in the reporting interval, other votes may
also be added. The CBS Report doesn’t specify the Volusia mistake; even the Feb
CNN Report wrongly lists it as 20,000 when it was almost exactly 19,000 (with
the actual votes from that precinct added), confirmed twice by Volusia election
officials.
CNN: In the Bush call, there was ample reason not to trust the VNS figures, which they now admit. Nevertheless, for whatever reasons, they overrode their best judgment.
ABC: From 10:10 p.m.
until 1:40 a.m. EST, the data and statistical models continued to be mixed,
with all but one model (the one that included actual vote count from the
counties) indicating that Mr. Gore would prevail. By 2:10 a.m., however,
the data provided by VNS indicated that a full 96% of all precincts had
reported and that, given the projections of the remaining outstanding vote, Mr.
Gore would have had to win over 63% of the remaining vote to prevail.
MARKEY:
It’s clear that models used by VNS were deeply flawed. This flawed methodology and
shoddy VNS data misled the networks news divisions.
CNN:
· There
was almost no communication between VNS and its membership in the critical
hours leading up to the Bush call.
· Inadequate quality control
at VNS (later admitted in a self-study) failed to catch the computer error in
Volusia County. Consequently, the error entered the system. The error was not
communicated to the members, even after it is was detected.
· Other errors entered the VNS
system for a variety of reasons. VNS became increasingly aware of this, but it
did not provide warnings to its members.
· One report points out that
some VNS personnel were released early that evening. We believe that all hands
should have been on deck when faced
with the uncertain situation
in a key state like Florida, particularly following the faulty Gore call and
the mistake in Duval County.
· It wasn’t until after 3 a.m.
that John King made the first reference of the evening to an automatic
mandatory recount in Florida. That was critical information that should have
been reported much earlier.
CNN (Decision desk failures): In the Bush call, there was ample reason not to trust the VNS figures, which they now admit. Nevertheless, for whatever reasons, they overrode their best judgment.
· It failed to take into
account the warning flags raised by previous errors in the VNS system that
evening.
· It failed to bring the
proper journalistic skepticism to the VNS numbers. Despite the previous
mistakes of the evening and despite the lack of communication, the decision
desk accepted VNS numbers as "convincing." At least one other news
organization was not convinced about making a Bush call; that is the AP, which
was already seeing discrepancies in the numbers.
· It failed to check with VNS
before it made its call.
· It failed to check other
available sources of information, namely the AP and the official state tallies
on the Web.
· It failed to be in touch
with Edelman at VNS until one full hour after the call was made.
CBS:
Had CBS News been checking two sources of the vote count--the Associated Press
and VNS totals--the Bush call at 2:17 AM might not have been made. Instead of a
nearly 50,000-vote margin statewide, the margin was just 30,000 votes. The AP
entered a correction in the Volusia County totals one minute (2:16am) before
CBS News made the call (simultaneously to the Fox call). VNS would
not make its correction until 2:51 AM.
The two sources might have corrected each other. However, if we had believed the AP data, we might have made the Bush call in Florida even earlier. It appears that bad Volusia County vote data had entered the AP vote total even earlier than it entered the VNS total. Beginning at 12:27 AM, Bush held a narrow lead in historically Democratic Volusia County in the AP count. Gore reasserted his lead with the AP’s 2:16 AM correction. Bush led in Volusia in the VNS tabulated vote beginning at 2:09 AM. That error was not corrected until 2:51 AM.
=They
called Florida for Bush after only 36 minutes of him leading on AP- before that
Gore was supposedly projected to win and then was ahead by 220,000 in the
national vote! No one watched the single most respected and reliable news
source on the planet- AP, the content of which fills 70% of most papers, because
they were so wedded to and confident of their autonomous VNS, though CBS may
have wanted to:
BOC: The margin separating the two candidates had descended in the AP tabulation from over 100,000 at 1 a.m. on November 8 to about 45,000 shortly after 2 a.m. In the next 15 minutes, it plummeted to less than 16,000. AP continued to report that it was too close to call.
CBS:
As it turned out, the VNS Florida state manager did not
use the AP as a resource either on Election Night or on the
day following the election. On Election
Night, the AP feed was unavailable to the vote-collection operation, and the
planned comparison checks (?) between it and the VNS count were never made. Had
the feed been working, the disparity between the AP and VNS Volusia County
results would most certainly have been caught, though perhaps not in time to
avoid the 2:17 AM call. However, there would certainly have been other county
vote differences that would have been noticed, and if members had been made
aware of those errors, they might have been more cautious in looking at the
Florida vote count.
There is yet a third source for vote-counting
information--secretaries of state and county Web sites. These were monitored at
VNS on Election Night, although apparently with varying skill and mixed
success. Despite the increased number of state and county Web sites, most ran
behind VNS in their vote totals. The most frequent VNS use of these sources was
for verification.
The Florida Secretary of State’s site ranked
somewhere in between, and it was used mainly for verification and comparison.
However, apparently at the critical period in the early morning hours of
election night, the site was overloaded and could not provide results.
WAT: Ironically, an election that is too close to call is itself a unique news story that should have kept viewers rapt for as long as it lasted. CNN, along with others, was driven by what appears to be a compulsion.. to provide an election result — even if there was not one.
POLLS + METHODS
Apart from the hideously wrong exit polls, only one
mention was made of the wildly gyrating national polls that led into election
night with Bush the preordained winner by 3-6% points.
STUPAK: I don’t know how you can have such wild
swings (in pre-election polls).
KON: We note that polls, in general, are statistical
calculations, not factual realities; and as such, that they are an imperfect
measure of voter intent and actual voting, especially in close elections.
CNN: The supposedly sophisticated system of polling is not nearly sophisticated enough. It is a flawed system that fails to take into full account many dynamic factors—absentee balloting, early voting, demographic change in key precincts, a declining response rate to polling generally, the quality of questionnaires, vote undercount, mistaken balloting, computer error, human error and more.
ELLIS’s ISLAND
AIL: All four members of the Decision Desk had
to agree on a call before it was recommended.
John Ellis has almost 23 years of experience
in calling elections. I am aware that Mr.
Ellis was speaking to then Governor George
W Bush and Jeb Bush on election night. I
do not see this as a fault or shortcoming
of Mr. Ellis. Quite the contrary, I see this
as a good journalist talking to his very
high level sources on election night. By
the way, Mr. Gorman and Mr. Mishkin (Decision
Desk) were speaking to high level Democratic
sources throughout the evening. To be clear,
there was no information which John Ellis
could have given to anyone nor was there
any unilateral decision which Mr. Ellis could
have made which would have affected the outcome
of the election.
=The spectacular impropriety of Ellis’s position
amazed some. “(in wonder) I can’t believe they let the cousin of
the candidate call the election,” said a mostly Republican editor of a
newsweekly. Only Rep. Sherrod plumbed the bizarre depths of Fox’s news
practices.
SHERROD: Anybody else think… that there’s a policy that you
don’t hire in a decision making capacity on election night a close relative of
the Presidential candidate? =nobody
answered, finally Konner did
KON: If
there’s a perceived conflict of interest, I would think there would be some
layers of insulation.
BIEM: When the
Bush call was made, there was only a .6% difference; looking at the data on the
screen- I wouldn’t have made that call- The data didn’t support the call.
I don’t understand why the call was made.
AIL: (Totally missing the point) We at Fox News do not discriminate against
people because of their family connections. I am more than happy to give you
examples of offspring of famous politicians who are employed at Fox News.
=In fact, because of the misestimations of the
remaining vote and the 23,000 vote errors for Bush, Ellis may not have been
derelict at that point in calling the election, by their weak standards. But
given the previous blown Gore call, given the multiple mistakes they knew
about, given the fact Gore was then ahead by almost ¼ million, given the
Democratic vote coming in from the Gold Coast- it was a catastrophic and
outrageous mistake, and no doubt colored by Ellis’s passionate desire to see
his cousin win. After being burned once, the other nets may have held off-
every second of delay would have given more chance for the tumbling AP or Fl
Sec State Bush lead to come to their attention: 47,000 lead at 2:12am, 30,000
at 2:16 (meaning automatic recount), 19,000 at 2:30, 11,000 at 2:52. AP
allegedly sent out an urgent bulletin at 2:37 warning there was “the narrowest
of margins” between candidates, which may have penetrated the Chinese Wall of
network infallibility. That deathly final call halted all analysis or urgency
of the incoming vote till they realized their exposure an hour later. Longtime
Republican political operative Roger Ailes and Fox “News” had no second
thoughts about Ellis- after all, they had wanted to make an impact in
their first election outing.
They
did- they (and all networks) convinced Al Gore and most of the country that
Bush won, and that lie ultimately became a self-fulfilling prophecy.
CNN: It also threw one state and the nation… into confusion and helped lead one candidate into prematurely conceding to the other.
WES: To some extent we were a victim of our own
success. The long record of accuracy led to hubris.
KON: We reported an impulse to speed over accuracy, and we attributed that impulse to the business imperatives of television news--to win the highest ratings, which is not a journalistic standard but a commercial standard. Ratings, that is the size of the audience, drive the price of the commercials, and commercials determine the bottom line profits of the corporations that own the news networks.
WELCHING IN ON THE NEWS
One of the most intriguing queries came from Waxman
(D-Ca), who asked Lack about whether GE Chairman Jack Welch had pressured the
news division in Bush’s favor.
WAXMAN: There’s an allegation that Jack Welch actually
intervened in NBC to call the election for George Bush… from several sources.
We’ve heard that it was captured on tape- filmed by NBC’s advertising +
promotion dept.
LACK: (chuckling) He was there, but that’s just a dopey
rumor.
WAXMAN: We request that .. you produce the tape, if such a
tape exists.
=According to a Waxman aide, NBC missed the deadline
for complying with the demand from the committee, but were sending a letter of
explanation.
CONCLUSIONS + PROMISES
SAV: VNS is actively pursuing numerous improvements
· using larger sample sizes for exit polls;
· developing new procedures to account for the
effects of a growing absentee and early vote, including more extensive
telephone polling of absentee voters;
· rewriting of the VNS projection and
statistical models;
· working to improve the exit poll accuracy and
response rate;
· completing work on the integration of The
Associated Press's tabulated vote as a second source of information;
· developing more sophisticated quality control
in the tabulated vote system and the rest of the VNS systems; and
· upgrading and modernizing the VNS technical
capabilities and infrastructure.
We are determined to do everything humanly possible
to make sure that these mistakes will never be made again.
=All participants agreed that it was better to fix
VNS than try to create something else.
WAT: It’s a problem to have a single source, but if you
have 5 separate polling operations, it might not be an improvement.
CNN (Appendix 5): The standard for an estimated vote margin
will be increased from 2.6 standard errors to four standard errors – i.e. a
Critical Value of 4.0 instead of 2.6 will be necessary for a “Call Status”
based entirely on exit poll data. (sounds impressive).
Mitofsky favors having VNS
abandon the release of “waves” of exit poll results to the networks early in
the day.
PROMISES
of CNN: 1. CNN's Continued Involvement with VNS is Conditional
CNN will remain within VNS if -and only if -
significant changes are made to assure that the errors that plagued election
night coverage in 2000 do not recur. To accomplish these goals and to implement
the RTI Report, additional financial contributions will be required of its
members.
2. A Second Source
for Data to Make Projections
CNN will fund a sample key precinct vote reporting
system in the states expected to have the closest races. This will ensure that
the network has a second source of data to use as a crosscheck against the VNS
data. In addition, CNN will insist that the Associated Press vote tabulation
system be better integrated into the election night data collection system.
3. CNN Will Not
Use Exit Polls for Projections in Close Races
Exit polls will not be used for projections in close
races.
CNN will only use exit poll survey data to project a
winner when the data indicates one of the candidates has a "large
margin" at the time that the polls close in that state. If CNN can't make the call in a state at
poll closing, it will then only project a winner in that state using actual
vote data from the statewide vote tabulations and key precincts.
If these standards had been
in place on election night 2000, CNN's projection of a winner would have been
delayed by at least 30 minutes in 10 states and Florida would never have been
called early in the evening for Vice President Gore. As for calls made at poll
closing in the 2000 election, 26 states were called by CNN, using VNS exit poll
data right when the polls closed in those states. All were correct calls.
Despite that, CNN will raise significantly the criteria for these exit poll
projections above what was used on election night.
4. Withholding a
Call When the Vote Margin is Extremely Close
CNN will not project a winner in a state, even if it
is reported that all the outstanding ballots have been accounted for, if the
balloting shows that there is less than a 1% margin between the candidates.
5. Multiple Poll
Closing States
CNN will no longer project the winner in a state
until all the polls are closed within that state.
6. Demystifying
Exit Polling and Projections
CNN will peel back the curtain for our viewers on
the exit poll and projection process, assigning correspondents to report how
the key precinct and exit poll workers do their jobs and providing other
"behind the scenes" reporting to show viewers how the projection
process works.
7. Election Night
Language and Graphics
CNN will change its language and graphics on
election night regarding projections. Until all the votes have been counted in
each state, CNN will no longer call anyone the winner of a state. For example,
anchors will say "CNN projects, that based primarily on exit poll
estimates, that Al Gore will win Michigan," On election night, CNN's full
screen graphics included the words "CNN Estimate" but the words were
not as prominent as they should have been.
8. Other Changes
Regarding VNS
CNN strongly urges that there will be one or more
outside respected academics, journalists or research professionals added to the
board
9. Uniform Poll
Closing
CNN enthusiastically supports the adoption of a
nationwide Uniform Poll Closing Act by Congress. If so adopted, CNN pledges not
to make any projections until all the polls are closed nationwide.
= The CNN Report went even farther in recommending
restricting dangerous practices.
CNN: Exit polling for
projection purposes: Cease the use of exit polling to project or call
winners of states. The 2000 election demonstrates the faults and dangers in exit
polling. Even if exit polling is made more accurate, it will never be as
accurate as a properly conducted actual vote count, and the current network
practice of sacrificing accuracy to speed should be reversed.
Sample precincts for
projection purposes: The use of returns from sample, or key, precincts for projecting or
calling winners also should be halted. These precincts, thought to provide a
representative sample from which larger results can be extrapolated, are
subject to numerous errors.
Accuracy over speed: Whether networks are trying
to come in first, or simply trying not to come in last, it is very clear that
the competitive zeal to report election returns quickly led the networks badly.
CNN should clearly emphasize that it prefers accuracy over speed in reporting
election returns, and its performance should make it clear that this emphasis
is genuine. We hope that all networks adopt such a policy. All would no doubt
say that they follow such a policy now, but their actual performance on Election
Night 2000 makes such an assurance hollow.
Woodruff said: “I’m one of the Neanderthals who think we shouldn’t call any states until votes are counted.”
Early calls serve no particular public or journalistic purpose. Taking a long view, it appears they do not even serve a commercial purpose.
CBS: A Cautionary Note on Uniform Poll Closing
While
uniform poll closing (and the consequent delay in calling elections) sounds
like a perfect solution, there might be unintended consequences. It opens up a much
larger window for misinformation and even disinformation to appear in
uncontrolled outlets. In 2000, various Internet sites published information
leaked from VNS exit polls. This practice began in February, when slate.com
published exit-poll results about the New Hampshire primary.
Slate did so again in South
Carolina, and was followed by nationalreview.com
and the Drudge Report on other primary days.
On
Election Day, exit polls were cited in midafternoon on-line (the Drudge Report
and inside.com) and on talk radio (Rush
Limbaugh), with results attributed to “campaign, sources” or simply “sources.”
Some of the information was correct (“Hillary Clinton is ahead in New York”),
while some was either wrong or very premature (“Bush holds an edge in...New Mexico,
Wisconsin and Iowa”).
=No one else agreed to the 1% rule or to not
use exit polls unless there was a large margin. All pledged to not call races
until polls closed, all supported uniform poll closing times, promised to
integrate AP and possible other sources, all resolved to fix VNS, and pledged
to elucidate the difference between projections and actual vote totals.
ABC had the novel and
seemingly risky idea to further isolate the decision desk, though it was too
little info and too few sources that caused the blunders, but CBS recommended moving in the
opposite direction:
ABC: Until now, however, ABC News has not
sought to restrict access of members of its Decision Desk to the reporting of
other news organizations, including competing television news organizations. It
was thought that the knowledge of what other credible news organizations were
and were not projecting could be helpful to the Desk in determining when it was
appropriate to make a projection. Competition, in news reporting as in other enterprises,
can be a good thing. But competition that encourages a journalist to report a
story prematurely is bad. In the particular instance of the Decision Desk, it
is most important that the individuals making projections do so based on two
things: the data provided to them (from exit polls, from actual vote tallies,
and from statistical models), and their own experience and judgment. They
should not be distracted or influenced by the decisions of other news
organizations.
CBS:
Move the Decision Desk into the Election Night studio. This
will promote constant contact between the newsgatherers and the analysts. If a
story is breaking, as it was in Florida this year, there will be constant
interaction, instead of the Decision Desk functioning in a vacuum, as it did
this time in an office three floors from the studio.
CNN: Television interfered with
the electoral process and the election result. In our opinion, that constitutes
an abuse of power, if unintentionally so, by CNN and by all the mainstream
television news operations. Television news did not adequately carry out the
role of free journalism in a free society in the 2000 election.
AIL: In my heart I do believe that democracy was harmed by my network and others on November 7. I do believe that the great profession of journalism took many steps backward.
*In ’85, controversy about the early call for Reagan
in his blowout against Mondale lead to hearings in the Election Subcommittee,
where NBC, CBS, + TBS News Presidents testified, according to Arturu Silva of
the Commerce Com.. Press Dept.. Other election hearings were also held in Feb
’84 and July ’83.
Michael Hammerschlag has written commentaries and articles for Seattle Times, Providence Journal, Honolulu Advertiser, Columbia Journalism Review; Moscow News, Tribune, and + Guardian; was a TV reporter and produced a documentary series on the Presidential primaries. He broke stories about the election night TV errors and Duval Co. vote irregularities. His website is http://mikehammer.tripod.com e-mail hammerschlag@bigfoot.com