2014 Exit Polls and Election Maps
Highlights from the Facebook page: Polit Soc.
(If some of them don't open from these links,
try the links on the Facebook page.)
Problems & Debates
about Polling Accuracy
- FiveThirtyEight, August 25, 2014, Is The Polling Industry In Stasis Or In Crisis? Essay on polling accuracy by one of the top analysts, Nate Silver of FiveThirtyEight.
- Pew, July 28, 2014, Q/A: What the New York Times' polling decision means. The New York Times and CBS News made big news in the polling world this weekend when they announced that they will begin using online survey panels from YouGov as part of their election coverage. YouGov, a U.K.-based research firm founded in 2000, uses such panels rather than traditional telephone surveys; the panel the Times and CBS are using has more than 100,000 members. The Times, citing concerns about the dearth of high-quality, non-partisan survey data, particularly at the state level, says it plans to include YouGov results as part of "a diverse suite of surveys employing diverse methodologies." While panels have long been used by market researchers, they're relatively new in the opinion-research field, and views on them are sharply divided. We asked Scott Keeter, the Pew Research Center's director of survey research, to explain the issues at stake and give us his preliminary thoughts.
- NY Times, August 5, 2012, Political Pollsters Struggle to Get the Right Cell Number. Analysts are tweaking methods to better include the views of voters who only have cellphones: some hang up on those who also have landlines, others turn to the Internet. For all you polling geeks (like me) ...
October 4, 2010, Understanding
Gallup's Likely Voter Models. Since 1950, Gallup has used likely voter
models to identify Americans who are most likely
to vote in a coming election. These models involve
asking poll respondents a series of questions about
their interest in the coming election, their past
voting behavior, and their current intention to vote
in the election.
Blumenthal, 10-5-10, 'Likely'
Voters: How Pollsters Define And Choose Them.
We have seen the "likely voter" polling
problem rear its head several times in recent weeks,
but few examples have been as vivid as three national
surveys released in the last 24 hours.
- FiveThirtyEight -
Nate Silver started his meta-analysis site in 2008.
The New York Times now presents it.
14, 2010, Bypassed
Cellphones: Biased Polls? On Wednesday,
Pew Research issued a study suggesting that
the failure to include cellphones in a survey
sample — and most pollsters don’t
include them — may bias the results against
Democrats. Pew has addressed this subject a
number of times before, and in their view,
the problem seems to be worsening. Indeed,
this is about what you might expect, since
the fraction of voters who rely on cellphones
is steadily increasing: about 25 percent of
the adult population now has no landline phone
installed at all. Clearly, this is a major
problem in survey research — and one
that, sooner or later, every polling firm is
going to have to wrestle with. What isn’t
as clear is how much of a problem it is right
now. I have written about this in the past,
and I encourage you to review those articles.
But let me try and come at it from a couple
of fresh directions.
4, 2010, The
Uncanny Accuracy of Polling Averages*, Part
IV: Are the Polls Getting Worse? There
is another type of argument, however, that
is potentially more troubling. It could be
that, irrespective of the character of this
political cycle, polling itself is in decline.
This is a widely held view among political
elites and many polling professionals — and
quite a few of the readers of this blog, I
3, 2010, The
Uncanny Accuracy of Polling Averages*, Part
III: This Time, It’s Different? In
Part III, we take up one type of critique that
I encounter frequently — that 2010 is
an unusual political cycle, and that its idiosyncrasies
may render the polling less accurate. While
this is not an unreasonable hypothesis, we
found it does not have any grounding in the
evidence: the polls have done no worse in “unusual” political
cycles like 1992, nor in “wave” years
like 1994 and 2006, than in routine-seeming
ones like 1996 and 1998.
30, 2010, The
Uncanny Accuracy of Polling Averages*, Part
II: What the Numbers Say. In Part II, I
demonstrated, by contrast, that a simple average
of polls has performed very well over the past
six election cycles in determining the winner
of the contest. For example, Senate and gubernatorial
candidates who have trailed by 6 to 9 points
in the polling average with a month to go until
the election have won their races only about
10 percent of the time in recent years.
29, 2010, The
Uncanny Accuracy of Polling Averages*, Part
I: Why You Can’t Trust Your Gut.
In Part I, I explored why our intuition may
mislead us when it comes to forecasting the
outcomes of elections — for a variety
of reasons, we may tend to assume that there
is more uncertainty in the forecast than there
Langer, The Numbers,
A Run at the Latest Data from ABC's Poobah of Polling,
Langer, August 30, 2010, This
I Believe. It’s
quickly mushroomed into the
summer’s hottest data
point: A boatload of Americans
believe Barack Obama’s
a Muslim. Except that, maybe,
they don’t. Consider
this instead: They’re
just willing to say it. This
is useful in understanding
public opinion and its measurement.
Yet the punditry and pronouncements
that have followed the Obama/Muslim
numbers mainly have missed
the point, falling instead
into the trap of literalism.
They say, so they believe.
Not necessarily so. People
in fact may voice an attitude
not as an affirmed belief – a
statement of perceived factual
reality – but rather
as what my colleagues and
I have taken to calling “expressed
belief” – a statement
intended to send a message,
not claim a known fact.
the big section in our 2008 election page, here,
methodological statements from Gallup (their
methods are typical of industry standards)
on factors that may distort polling's accuracy
- The "Bradley
Effect" - do survey respondents lie about
Game:" Turning out the Vote
Plus: Defense against Voter Fraud & Vote Suppression
- NYT, August
25, 2010, Shaping
Tea Party Passion Into Campaign Force, By KATE
ZERNIKE. On a Saturday in August when most of the
political class has escaped this city’s swelter,
50 Tea Party leaders have flown in from across the
country to jam into a conference room in an office
building on Pennsylvania Avenue, apparently unconcerned
that the fancy address does not guarantee air-conditioning
on weekends. They have come to learn how to take
over the country, voter by voter. ...This is a three-day “boot
camp” at FreedomWorks, the Washington advocacy
group that has done more than any other organization
to build the Tea Party movement. ...The goal is to
turn local Tea Party groups into a standing get-out-the-vote
operation in Congressional districts across the country.
- WP 8/23/10. Primary
turnout shows big GOP enthusiasm edge. By Aaron
Blake.Three-quarters of the way through the 2010
primary season, the so-called "enthusiasm
gap" appears to be playing out across the
country with turnout in GOP contests exceeding
previous highs and beating Democratic turnout by
unprecedented margins in many targeted states.
on Karl Rove's successful "Ground Game"
that worked for George W. Bush ... until it didn't
of the Rove/Bush Strategy
Party Country: The Republican
Plan For Dominance in the 21st
Century" at Amazon
September 28, 2006 article in
tries to answer these questions:
G.O.P. still raises more money
than the Democrats, but the
Democrats are hardly short
of cash. How significant is
the G.O.P. advantage in terms
of sheer dollars? Are they
simply raising more money,
or are they also doing a better
job of spending it?
successful has the G.O.P. been
in eating away at Democratic
support among core constituencies
like African Americans and
say that Republicans have surpassed
the Democrats in mobilizing
their voters on election day,
in part by using databases
such as Voter Vault, which
allows party activists to track
voters by personal hobbies,
professional interests, and
even by their favorite brand
of soda. How does that bank
of personal data translate
into an advantage on election
day? Are Democrats responding
with similar programs of their
structural advantages the Republicans
have, hasn't the G.O.P. also
sought to gain an electoral
advantage by suppressing Democratic
turnout? How significant are
those efforts on the part of
the G.O.P., and are we likely
to see new and improved methods
down the road?
would no doubt argue that their
policies and ideology are simply
more popular with the public
than Democratic policies. Do
ideas still play a role in
electoral success or is it
all about money and organization?
Rehm show, 27 July 2006, here
Air with Terry Gross, July 24,
November 15, 2005. By Jim Rutenberg, Voter
Profiles for Bloomberg Went Beyond Ethnic
Labels. Throughout this year's mayoral
campaign, Mayor Michael R. Bloomberg's spending
records included something called "voter
list development." It looked ominous
to Democrats - especially as Mr. Bloomberg
poured millions into it. Lists like this
usually include voters' personal data - the
magazines they buy, the cars they drive,
their political affiliations. But as the
cost of compiling Mr. Bloomberg's list inched
up toward $10 million, not even aides to
President Bush, who perfected this sort of
voter identification last year, could figure
out where the money was going.
Angeles Times July 24, 2005. By Tom
Hamburger and Peter Wallsten, Parties
Are Tracking Your Habits. Though both
Democrats and Republicans collect personal
information, the GOP's mastery of data
is changing the very nature of campaigning.
COLUMBUS, Ohio — At first glance,
Felicia Hill seems to fit the profile of
a loyal Democrat: She is African American,
married to a General Motors union worker
and voted for Dukakis, Clinton and Gore
in past presidential elections. But in
the weeks before election day 2004, the
suburban mother of two was deluged with
telephone calls, invitations and specially
targeted mailings urging her to support
President Bush. The intense Republican
courtship of Hill, 39, was no coincidence.
A deeper look at her lifestyle and politics
reveals a voter who might be persuaded
to switch sides. Among the clues: she is
a church member uneasy about abortion;
she lives in a growing suburb and she sent
her children to a private school. ...For
the first time, she sees the GOP as a place
where black women can be comfortable. "I
saw people I could relate to," she
said, describing conversations she had
with Republican professional women during
telephone outreach calls and at party events.
...Hill and millions of other would-be
Bush backers in closely contested states
were identified by a GOP database that
culled information ranging from the political
basics, like party registration, to the
personal, such as the cars they drive,
the drinks they buy, even the features
they order on their phone lines. The "micro-targeting" effort
was so effective that the party credited
it with helping to secure Bush's reelection.
December 6, 2004. By Katharine Q. Seelye, How
to Sell a Candidate to a Porsche-Driving,
Leno-Loving Nascar Fan. After the 2000
presidential campaign, strategists for President
Bush came to a startling realization: Democrats
watch more television than Republicans. So
by buying millions of dollars' worth of television
advertising time, Republicans were spending
their money on audiences that tended to vote
Democratic. What to do? With the luxury of
four years until the next election, the Bush
team examined voters' television-viewing
habits and cross-referenced them with surveys
of voters' political and lifestyle preferences.
This led to an unusual step for a presidential
campaign: it cut the proportion of money
that it put into broadcast television and
diverted more to niche cable channels and
radio, where it could more precisely reach
its target audience.
November 19, 2004. By Adam Nagourney, Bush
Campaign Manager Views the Electoral Divide.
After two years of polling, market testing
and up-close demographic scrutiny of American
voters, the manager of President Bush's re-election
campaign, Ken Mehlman, offered another way
Thursday to view the divide between the American
electorate. "If you drive a Volvo and
you do yoga, you are pretty much a Democrat," Mr.
Mehlman told an assembly of the nation's
Republican governors here. "If you drive
a Lincoln or a BMW and you own a gun, you're
voting for George Bush." ...Rather than
dispatching troops to knock on doors in neighborhoods
known to be heavily Republican, Mr. Mehlman
said, the Bush campaign studied consumer
habits in trying to predict whom people would
vote for in a presidential election. "We
did what Visa did," Mr. Mehlman said. "We
acquired a lot of consumer data. What magazine
do you subscribe to? Do you own a gun? How
often do the folks go to church? Where do
you send your kids to school? Are you married? "Based
on that, we were able to develop an exact
kind of consumer model that corporate America
does every day to predict how people vote
- not based on where they live but how they
live," he said. "That was critically
important to our success."
Post, November 7, 2004. By Dan Balz
and Mike Allen, Four
More Years Attributed to Rove's Strategy.
Despite Moments of Doubt, Adviser's Planning
Paid Off. Admired, disparaged, respected
and feared, [Karl] Rove joins an elite
cadre of political strategists who can
claim two presidential victories. Bush's
adviser can now look toward the goal he
has pursued since he was an obscure direct-mail
specialist in Texas: the creation of a
durable Republican majority in Washington
and across the country.
Post, November 4, 2004. By John F.
Bears Out Emphasis on Values. GOP Tactics
Aimed At Cultural Divide. ...The results
appeared to validate several of the pet
theories of [GOP campaign director Karl]
Rove, including his belief that politics
is as much science as art. Presidential
stops in swing states, and the route of
campaign bus trips, rarely included the
largest cities. That was because Rove
chose them scientifically, using three
criteria that he explained to reporters
in the waning days of the campaign.
Rove said his targets were areas where
Bush had underperformed in 2000, whether
Republican or Democratic, and where the
campaign's target for votes was higher
than the number that showed up. Second
were fast-growing exurban areas or Republican
places where there were a large number
of people who ought to register to vote
and do not -- what Rove calls "a large
gap between participation and potential." Third,
he said, he paid attention to areas "that
have a significant number of swing voters,
and swing wildly from election to election."
November 4, 2004. By Elisabeth Bumiller, Turnout
Effort and Kerry, Too, Were G.O.P.'s Keys
to Victory. In the closing hours of President
Bush's campaign for re-election, Karl Rove,
his chief political adviser, was obsessed
with turning out Republican votes. Late on
Monday night, Mr. Rove stood in the cold
at a rally in Albuquerque and pulled scraps
of paper from his pocket covered with numbers
that reassured him that his ground army was
in full assault.
July 18, 2004. By Jim Rutenberg, Campaigns
Use TV Preferences to Find Voters. When
deciding where to run his television advertisements,
President Bush is much more partial than
Senator John Kerry to crime shows like "Cops," "Law & Order" and "JAG." Mr.
Kerry leans more to lighter fare, like "Judge
Judy," "The Ellen DeGeneres Show" and "Late
Show with David Letterman." Those choices
do not reflect either man's taste in television,
but critical differences in the advertising
strategies of their campaigns, which are
spending more money for commercials than
any other campaigns in presidential history.
Crime shows appeal to the Bush campaign because
of its interest in reaching out to Republican
men who are attracted to such programming.
By contrast, the Kerry campaign is more interested
in concentrating on single women, who tend
to be drawn to shows with softer themes.
April 7, 2004. By Joyce Purnick, Data
Churners Try to Pinpoint Voters' Politics.
There's this great story making the Washington
political rounds about the Conservative Party
in Britain. It is that fund-raisers in London
found a strong correlation between Conservative
Party donors and people who buy garden bulbs
by mail. Far-fetched? Maybe not, because
people who plant spring bulbs tend to be
more suburban and rural than urban, more
wealthy than poor and, with time to garden,
older. Hence, a likely Conservative, right?
April 6, 2004. By Joyce Purnick, Foraging
For Votes: One-Doorbell-One-Vote Tactic Re-emerges
in Bush-Kerry Race. They call it the
ground war. And as anticipated, it is back
after a long hiatus, subtly changing politics
as we know it. Or trying to. After decades
of playing poor relation to television advertising,
grass-roots politics has become a campaign
star this year, as many political pros predicted
it would be in the aftermath of the Bush-Gore
face-off of 2000. And today it ranges from
old-fashioned shoe leather to Web technology
that can make a precinct captain of anyone
with a computer.
Post, November 10, 2002. In
GOP Win, a Lesson in Money, Muscle, Planning.
[Karl] Rove, [Rep. Tom] DeLay and others
concluded that Republicans had lost the
turnout battle in recent elections by focusing
too much on paid advertising and too little
on the ground war that Democratic allies
such as the AFL-CIO do so well: getting
potential voters to the polls. Beginning
in early 2001, the party registered thousands
of new Republican voters, particularly
in fast-growing states. It invested heavily
in a program, dubbed the "72-hour project," that
would later help spur record turnout in
key regions. The Republican National Committee
spent millions of dollars honing a system
to identify voters, down to specific households,
and contact them repeatedly with phone
calls, mail and visits from party activists.