Media induced voter suppression
- 1 Origins of the phenomena
- 2 Exit polling
- 3 Early declaration of victory & U.S. Time Zones
- 4 Reporting long lines
- 5 Notable incidents of media voter suppression in recent elections
- 6 Possible Solutions
- 7 References
- 8 See also
Media induced voter suppression refers to media coverage that may result in the lowering of voter turnout on election day.
Origins of the phenomena
Media induced suppression is a relatively new phenomenon resulting from the increased access to media reports, the increased speed at which media reports and polling data are compiled and released, the confusing nature of collecting polling data, and possibly of internalized bias within media organizations.
The main origin of the phenomena appears to be the existence of multiple time zones, both nationally and within single states as is the case in Florida. Often, election result and exit poll data is released from Eastern and Central time zone states while polls are still open in states located in the Mountain and Pacific time zones. Media outlets report this information to the public as soon as it is received, and in increasingly fast mediums such as real-time Internet coverage. Because of this, news reporting has played a key roll in determining voter turnout in states beyond the Eastern time zone.
Exit Polling is a poll of voters taken immediately after they have exited a polling location. Exit polls are used to compile early, immediate data on election results on Election Day by the National Election Pool and a consortium of media organizations. This data is then used to report on election day activities.
Unlike pre-election polls taken by professional polling groups, exit polls do not factor in statistical data about voter demographics, nor is extensive research done in compiling voter pools before the poll is conducted. The exit poll is drawn from a purely voluntary pool of voters who divulge information about their recent vote to pollsters.
The primary value of an exit poll is its usefulness in determining early results of an election. Under the model for an exit poll, voters exiting a polling place would report truthfully to pollsters about the vote they recently cast. Pollsters would then be able to compile accurate information about election statistics from a particular precinct. That data would then be used to make an informed prediction as to the winner of that precinct.
The "shy elephant" effect
Because of predispositions and the voluntary nature of exit polling,these polls tend to be skewed to the disadvantage of Republican voters. This has been colloquially termed the Shy Elephant Factor, a general theory that states that Republican voters are less likely to divulge information on their voting choices than Democratic voters. During the Presidential election of 2004 it is estimated that, because of the tendency for Democrats to report to pollsters, John Kerry was shown to be over-performing by 5.5% in early exit polls.
“Our investigation of the differences between the exit poll estimates and the actual vote count point to one primary reason: in a number of precincts a higher than average Within Precinct Error most likely due to Kerry voters participating in the exit polls at a higher rate than Bush voters. There has been partisan overstatements in previous elections, more often overstating the Democrat, but occasionally overstating the Republican.”
The phenomenon is not isolated to national exit polls. The 2004 Election System Report found that results for John Kerry were over-stated by at least one standard error in 26 states. Results for George W. Bush were over-stated by at least one standard error in four states.
The "Bradley effect"
The Bradley effect, less commonly called the Wilder effect, is a theory proposed to explain observed discrepancies between voter opinion polls and election outcomes in some elections where a white candidate and a minority candidate candidate run against each other. The theory proposes that some voters tend to tell pollsters that they are undecided or likely to vote for a black candidate, and yet, on election day, vote for his white opponent. It was named for Tom Bradley, an African-american who lost the 1982 California governors race despite being ahead in voter polls going into the elections.
The Bradley effect theorizes that the inaccurate polls were skewed by the phenomenon of social desirability bias. Specifically, some white voters give inaccurate polling responses for fear that, by stating their true preference, they will open themselves to criticism of racial motivation. The reluctance to give accurate polling answers has sometimes extended to post-election exit polls as well. The race of the pollster conducting the interview may factor in to voters' answers.
Some analysts have dismissed the theory of the Bradley effect, while others argue that it may have existed in past elections, but not in more recent ones. One analysis of 133 senate and gubernatorial elections between 1989 and 2006 suggests that "before 1996, the median gap for black candidates was 3.1 percentage points, while for subsequent years it was -0.3 percentage points."
Effect of exit polling on election results
A Cornell university study recently stated that such skewed exit polling results are known to influence voters at unclosed polls in western U.S. timezones. Because they perceived, based on early data, that the election would not result in a favorable outcome, or because they perceived their candidate had already won enough votes to solidify an electoral win, voters were discouraged from casting ballots in these western states.  Both Republican voters and Democratic voters were shown to be susceptible to this phenomenon.
2008 elections and exit polling
A poll conducted by Rasmussen Reports in key 2008 presidential battleground states of Colorado, Florida, Missouri, North Carolina, Ohio and Virginia found that Democrats are more eager to take exit polls than Republicans In five of the six states, a majority of Democrats say they would be "Very Likely" to participate in the exit polling process, while fewer than 40% of Republicans would be willing to do the same.
A poll conducted by Fox News shows that 46% of Presidential candidate Barack Obama's supporters were "very likely" to participate in exit polling, whereas 35% of Presidential candidate John McCain's supporters.
Both campaigns have warned about possible skewed results from exit polling. Senator John McCain's campaign send out an email alert on the eve of Election Day warning supporters of the Shy Elephant Factor and the tendency of exit polling to over-state favorable results for Democratic candidates. Senator Barack Obama's campaign sent a message to supporters through its local campaigns warning about the possible "Bradley Effect" and warning supporters not to be overconfident if early election results showed a significant lead.
Early declaration of victory & U.S. Time Zones
Television media often "projects" state winners on Election Night using exit poll data and extrapolation based on early precinct returns. Such "early vote projection" allows media and professional polling organizations to "call" certain states for candidates before all of the ballots cast are counted.
According to Rasmussen, this early vote projection model has an effect on states located in western time zones. Because early vote projection allows media to claim that large electoral states like Florida, which have a significant influence on the election, have already been allocated to one candidate, voters in western states may pass up the chance to cast their ballots. Although these uncast ballots are the result of, primarily, Presidential election projections, voters in these states also fail to cast ballots for local candidates and local referendum resulting in skewed results inside states.
The problems of early declaration and early voter projection were especially pertinent in the 1964 Presidential election and the 2000 Presidential elections. In 1964, CBS news projected Lyndon Johnson the winner of the Presidential election at 6:04 Pacific time, while polls were still open -- and would remain open -- in the central west and on the west coast. In 2000, NBC, CBS, CNN, Fox News, ABC and the Associated Press declared that Vice President Al Gore had won electoral votes from the state of Florida before 8p.m. Eastern time. This prediction, however, came while polls in Florida's panhandle remained open. The election was called, recalled, called again and, eventually recounted.
Studies have varied on whether the impact of projecting state-by-state results has an effect on voter turnout. Several studies, published in the Harvard and Univ. of Cincinnati law reviews, respectively, have found that the impact is "slight to negligible.". However, at least one report documenting a series of statistics from Presidential elections in the 1980s suggests that voter turnout in the west could have been depressed as much as 12% by early projections. .
Reporting long lines
Media reports of long lines or election delays may discourage voters and therefore suppress a number of voters who might otherwise vote in a particular election.
This phenomenon is particularly common in elections with high voter turnout. This has been noted in most Presidential Elections since 1960..
In these instances, a court's ruling may avert this suppression. Courts may rule that polls remain open past closing times indicated in state law. This is usually common in mid-sized to large metropolitan areas in populous states like Ohio, Illinois, and California.
Notable incidents of media voter suppression in recent elections
- It was early poll numbers leaked in 2004 that showed Kerry leading Bush, contrary to the final results. One of those in charge of the polling later concluded that the discrepancy was because “Kerry voters were more anxious to participate in our exit polls than the Bush voters.”.
- During the 2004 Presidential election, media reports from Ohio, reporting long lines, resulted in alleged media induced voter suppression .
- In the 2000 United States presidential election it was alleged that media organizations released exit poll results for Florida before the polls closed in the Florida panhandle closed. People believe that this caused many people in the more heavily Republican Florida panhandle to not cast their vote.This may have caused mass confusion during 2000 Election Night when the networks erroneously called first for Al Gore then George W. Bush, then put Florida back in the undecided column before Florida went through a month long recount and court battle which determined who won the 2000 Presidential election. The chaos of the 2000 FloridaPresidential Election results led to widespread criticism by Democratic Party supporters over voting rights issues and the legitimacy of the electoral college system.
- In the 1980 United States presidential election, NBC News predicted a victory for Ronald Reagan at 8:15 pm EST, based on exit polls of 20,000 voters. It was 5:15 pm on the West Coast and the polls were still open. There was speculation that voters stayed away after hearing the results. Thereafter, television networks voluntarily adopted a course of not using exit poll data after the polls close in certain states..
Political speech, which includes reporting on the results of elections, is protected by the First Amendment. Government cannot effectively impede the reporting of election results, nor the collection of election data. States attempts at discouraging or barring the practice of exit polling have been repeatedly struck down by courts. 
After the 1984 election, at the insistence of legislators in the Pacific and Mountain time zones, both the United States House of Representatives and the United States Senate passed a joint resolution urging broadcasters to "voluntarily refrain from characterizing or projecting results of an election" based on data from exit interviews "before all polls for the office have closed." Network news executives in the United States and observers in the political community decried the resolution claiming that it would force broadcasters to refrain from giving the public news about presidential and other election results until every poll closed in every state, including Alaska and Hawaii. Opposition to the resolution also feared that insiders in the press and political campaigns, who would have access to exit poll results, would know who won the elections in each state, while the public would remain uninformed..
As of today, it is standard practice of the major United States News networks to project the presidential victory after polls close in the Western U.S. States excluding Hawaii and Alaska.
The inaccuracies which caused the events of the 2000 Presidential election were largely blamed on the Voter News Service, a media collective established by most of the major media outlets. The Voter News Service collected survey responses,exit polls, and actual vote tallies as well as the vote count from precincts and counties. Based upon this information, VNS provided data and analysis to all of the members equally. It was the VNS that predicted Gore had won Florida at 7:52p.m.
Following the events of the 2000 election, the VNS was effectively disbanded, and no longer reported results from precincts directly. The major media outlets also agreed not to call a state for a particular candidate until that state's polls had closed. CBS news is the only media outlet that has stated that it will not abide by this mandate in the 2004 and 2008 Presidential elections: CBS will refrain from "calling" a state, however it will not refrain from predicting that a state will "lean" one way or another.
Single time zone
The US Time Zone map shows how early poll closings can affect states in the Mountain and Pacific Time Zones. The United States State Department proposed the creation of a single, nationwide Time Zone following the events of 2000. Adjusting the time zones is the prerogative of Congress and would be accomplished only by passage of legislation. The Department of Transportation can also approve a request to change a time zone, provided that the Secretary of Transportation signs off on the matter. The earliest a change can take place, however, is approximately two years after the initial request is made.
Exemption from Daylight Savings
In situations where a state lies in two time zones, such as with the state of Florida, it has been suggested that the part of the state that lies in the time zone further to the west would be exempted from Daylight Savings Time. This way, the entire state would be in a single time zone during the election.
The Uniform Time Act of 196613 ("Act") governs the ability of states to exempt themselves from adherence to daylight savings time ("DST"). The Act permits a state to exempt itself from observance of DST "by law," i.e., by passing legislation to the effect that either the entire state will no longer observe DST, or (if the state lies in two time zones) that the portion of the state which falls in one of the time zones will no longer observe DST as specified by the Act. It would be up to each state where this is a problem to decide whether to exempt part of the state from DST.
Uniform polling times
In 1985, the US House Subcommittee on Elections proposed a uniform poll closing time due to previous and apparent landslide victories for the three out of the last four presidential elections in 1972, 1980, and 1984.
The Uniform Poll Closing report, authored in response to this proposed bill, found that different people vote at different times of day based on job, demographic, culture and other factors. The National Commission on Federal Election Reform in its 2001 report on this subject, relied on statistical data from the 1980's which indicates that changing the hours of operation of polling places will affect these distinct demographic groups depending on whether polls are opened later and closed earlier, or opened earlier and closed later. As such, care would have to be taken in order to establish a uniform polling time that did not further disenfranchise voters.
- Evaluation of Edison/Mitofsky Election System
- Kevin Drum, "East Coast Bias Watch", washingtonmonthly.com, July 23, 2008, citing a Google search: "3,820 hits for Wilder Effect compared to 44,900 hits for Bradley Effect"
- Payne, Gregory(1986). Tom Bradley: The Impossible Dream : A Biography Roundtable Pub. The chapter about Bradley Effect (Chapter 16 / pp. 243 - 288) is available online at http://www.gregorypayne.net/bradleybio.pdf
- Langer, Gary. (1989, November 8). “Election Poll Problems: Did Some Voters Lie?”, Associated Press
- Reddy, Patrick. (2002, January 20). "Does McCall Have A Chance?", Buffalo News, p. H1
- Fighting the Last War - TIME
- Perez, Simon. (2008, October 9). "Could Bradley Effect Change November Election?" KPIX-TV, "Political Consultant Don Solem explains: 'It's not so much they're afraid to say it as they think it might be taken the wrong way.' Solem said the Bradley Effect is also known as social desirability bias."
- Rojas, Aurelio. (2008, October 9). "California poll on Prop. 8 could show 'Bradley effect'" Sacramento Bee, "'Anyone who studies survey research will tell you one of the biggest problems we encounter is this notion of social desirability bias,' [Patrick Egan, a professor of politics at New York University] said."
- Daniel J. Hopkins (2008-10-04). "No More Wilder Effect, Never a Whitman Effect: When and Why Polls Mislead about Black and Female Candidates". Department of Government, Harvard University. Retrieved on 2008-10-10.</cite>
- Michael Dorf Law, "Cornell University" 2008
- "Rasmussen Reports", Beware of Exit Polls, November 2nd, 2008
- Fox News 2008 Poll, PDF format
- "Columbia Journalism Review", Exit Polls, Academy Awards, and Presidential Elections, May 1, 2000
- Reports by the Division of Elections on the possibility of uniform voting times, U.S. Department of State
- 98 Harv. L. Rev. 1927,1945 fn3; 58 U.Cin. L. Rev. 1003, 1005
- Uniform Poll Closing Report, p.3
- MSNBC Turning Points", 1972 Presidential Election, October 31, 2008
- MSNBC Turning Points", 2000 and 2004 Presidential Elections, October 31, 2008
- Ballotpedia Exit Polls
- MSNBC Turning Points", 2000 and 2004 Presidential Election, October 31, 2008
- Daily Herald v. Munro, 838 F.2d 380 (9th Cir. 1988).
- U.S. House of Representatives, Energy and Commerce Committee Hearing, p. 93 (AP), p.97 (ABC), p.114 (CNN), p. 117 (Fox)
- 15 U.S.C. 261
- Uniform Poll Closing Report, 2001