Bush 45%
Kerry 44%
Nader 3%
Unsure 6%
Note: Colorado will vote on a ballot initiative that would split its
nine electoral votes proportionately. It is written to apply to this
election, but if passed, there may be a legal battle over that provision.
*Return to Top*
<http://www.latimes.com/news/politics/la-polldatapage,1,1906511.htmlstory#Top>
/American Research Group. Sept. 17-20, 2004. N=600 likely voters. (MoE ±
4):/
Bush 45%
Kerry 46%
Nader 2%
Other 1%
Unsure 6%
Note: Two earlier September polls taken during hurricanes are not shown.
------------------------------------------------------------------------
/Research 2000 for The South Florida Sun-Sentinel and The Florida
Times-Union. Aug. 23-25, 2004. N=600 likely voters statewide. MoE ± 4
(for all likely voters):/
Bush 46%
Kerry 46%
Nader 2%
Unsure 6%
Note: Voters were asked who would they vote for if Nader were not on
ballot. Net results were Bush 46%, Kerry 47%, Unsure 7%.
*Return to Top*
<http://www.latimes.com/news/politics/la-polldatapage,1,1906511.htmlstory#Top>
*Indiana*
/Research 2000 for The South Bend Tribune and WSBT-TV. Sept. 7-9, 2004.
N=600 likely voters statewide. MoE ± 4:/
Bush 54%
Kerry 38%
Unsure 8%
Note: The Nader campaign has abandoned a bid to get on the ballot here.
*Return to Top*
<http://www.latimes.com/news/politics/la-polldatapage,1,1906511.htmlstory#Top>
Bush 52%
Kerry 43%
Nader 1%
Unsure 4%
Note: Among all registered voters, poll has Bush at 48% and Kerry at 46%.
------------------------------------------------------------------------
/Mason-Dixon Polling & Research for Knight Ridder, MSNBC and The Las
Vegas Review-Journal. Sept. 13-14, 2004. N=625 likely voters statewide.
MoE ± 4:/
/Mason-Dixon Polling & Research for Knight Ridder, MSNBC, The Santa Fe
New Mexican, and KOB-TV. Sept. 15-16, 2004. N=625 likely voters
statewide. MoE ± 4:/
*Wisconsin*
/University of Wisconsin/Capital Times/Milwaukee Journal Sentinel Badger
Poll. Sept. 15-21, 2004. N=485 adults statewide who are registered to
vote or have definite plans to register so they can vote in the election
on Nov. 2 (MoE ± 4), including 468 who "said they would either certainly
or probably actually cast ballots" (MoE ± 4):/
Polls, Methodology and the Interactive Map
The U.S. presidential election is decided by the results of 51 electoral
vote contests. In all states except Maine and Nebraska the electoral
votes go to the winner of the popular vote in a "winner-take-all"
system. (Colorado is considering an initiative to create proportional
distribution of its electoral vote in November.) Thus, trends in the
campaign must be considered on a state-by-state basis.
The initial setting of the pop-up flash interactive map is colored red
or blue if either candidate in the latest major poll has a lead outside
the margin of error. Poll data are provided by PollingReport.com; only
polls with methodologies that meet its standards are used. The initial
setting also shows a recent national poll for reference, but note that
the national vote does not determine the winner. PollingReport.com lists
national polls on the free section of its site; see link at the top
right of this page.
The Times' map does not attempt to indicate states that are "leaning"
one way or another or make conclusions about trends from a series of
polls. The data used is from the most recent major poll only. In a few
cases, we've included other recent polls above.
When no recent poll is available but voting trends and patterns suggest
one candidate is likely to win a state, this page and the map may show a
state as "Bush" or "Kerry." Otherwise the state is white, indicated by
"swing" above or "Up for Grabs" on the map.
Polls are a snapshot in time; changes are expected. Notes below each
state may indicate a second poll taken within a few weeks or another
relevant factor. The campaigns, The Times and other observers have
identified 17 key battleground states that are indicated with an asterik
in the first list of states at the top of this page.
The poll data presented here are from probability-sample surveys,
conducted by telephone using live interviewers. Where available, surveys
of likely presidential election voters are used. However, users should
be aware that polling organizations screen for "likely" voters in
different ways. All generally use various combinations of three
criteria: Voting history, interest in the election and stated intention
to vote. But variations in the results of different surveys are to be
expected. For example, some organizations will never use a first-time
voter in a poll because they have no voting history while others may
include them if their answers to certain screening questions indicate
they are interested in the election.
In addition, some organizations weight their results to conform to
specific proportions of political party identification (sometimes based
on past turnout or other polls that ask about party identification),
while others consider that variable to be too changeable to use as a
weight.
In some cases, polls that have been reported in the news media are not
included here because details of the survey methodology are unavailable.
When a poll has asked presidential preferences both with and without
Ralph Nader, the survey that includes Ralph Nader is used. If Nader
ultimately does not qualify for a state's ballot, however, the survey
without him will be used. Other third-party candidates will be on this
page if they had at least one percent in the poll but added to
"undecided" on the map, due to space constraints.
Some polls used name the vice presidential candidates and some do not.
Full details are available from PollingReport.com.
The map, its colors and all notes are by latimes.com. Only the reported
polling data comes from PollingReport.com.
On the initial setting of the map, white states are within the latest
polls' stated margin of error (MoE ±). ( Definitions of margin of error.
<http://www.google.com/search?q=define%3A&q=margin+of+error&Overview=1>)
"Bush" states are states with red and "Kerry" states are blue.
Comments may appear as you roll over states. "Safe" is used when there
are large leads in polls and trends from previous elections for one
party or another.
(R) or (D) sometimes appears to indicate that the polling organization
is often associated with one party or the other.
Electoral Vote Changes for 2004
After the 2000 census, seats in the House of Representatives were
reallocated based on new population trends. Electoral votes are based on
those numbers. Here are the changes for 2004 from 2000:
*Gained Electoral Votes:* Arizona (+2); Florida (+2); Georgia (+2);
Texas (+2); California (+1); Colorado (+1); Nevada (+1); North Carolina
(+1).
Libertarian Mark Badnarik appears to be safely on the ballot in all
states except Oklahoma. Green Party candidate David Cobb is likely to be
on 30 state ballots and Constitution Party candidate Michael Anthonty
Peroutka is likely to be on about 40 state ballots. Of these, only
Badnarik has had as much as one percent in state polls covered by this
page since June. Those were in Minnesota and Iowa. There are other
third-party candidates as well, but they are on fewer state ballots.
Ralph Nader, however, is often included in these surveys and polls above
one percent routinely. However, Nader is unlikely to be on as many state
ballots as he was on in 2000, when he was on 43 ballots.
Although not all authorites agree, as of Sept. 22, 2004:
. Nader appears to be permanently blocked from the ballot in Arizona,
California, Georgia, Indiana, Massachusetts, Missouri, North Carolina,
Oklahoma and Virginia.
. Legal challenges continue in Arkansas, Hawaii, Idaho, Illinois, Maine,
New Hampshire, New Mexico, Ohio, Oregon, Pennsylvania and Texas.
. Nader appears to be qualified for the ballot in Alabama, Alaska,
Colorado, Connecticut, Delaware, D.C., Florida, Iowa, Kansas, Kentucky,
Louisiana, Maryland, Michigan, Minnesota, Mississippi, Montana,
Nebraska, Nevada, New Jersey, New York, North Dakota, Rhode Island,
South Carolina, South Dakoka, Tennessee, Utah, Vermont, Washington, West
Virginia, Wisconsin and Wyoming.
In states where Ralph Nader is not on a party ticket, his campaign had
to file independent petitions to appear on the ballot.
The Reform Party endorsed Nader, which enabled him to skip petition
drives. However, litigation and battles within the Reform Party in some
states made his status unclear. Other appeals and litigation continues
over several issues, including the issue of Republicans gathering
petitions for him.
Because Nader's former group, the Green Party, elected not to endorse
him this year, he will not appear automatically on the Green ticket in
the 22 states where the party has gained ballot access.
About "PollingReport.com"
PollingReport.com publishes data from opinion polls of the American
public. Many surveys are available at no cost at its website,
pollingreport.com <http://www.pollingreport.com>.
Additional data (including state-by-state presidential, congressional,
and gubernatorial polls), plus analyses by leading pollsters, are
available to subscribers. An introductory one-year subscription costs
$95 and includes access to its subscriber web site, plus its
twice-a-month newsletter. For details, see:
www.pollingreport.com/nletter.htm
<http://www.pollingreport.com/nletter.htm>.