What is an exit poll?

Exit polls are public opinion surveys conducted with voters immediately after they cast their ballots.

How accurate are exit polls?

Exit polls face challenges with accuracy and reliability. They don't accurately predict the winners of elections, the data set changes throughout the election, and participant bias may occur.

Are exit polls reliable?

Exit polls are more reliable in providing information about the success of an elected official's campaign, shedding light on who voted for the winner, and providing insight into their support base than they are in determining election results.

Do exit polls include early voting?

Exit polls often do not include mail-in voting or early in-person voting.

Where are exit polls conducted?

Exit polls are conducted outside voting locations.

Exit Polls: Definition & History

Exit Polls

If you've ever followed a close election on a television network, you've probably seen them announce the projected winner. This information likely came, in part, from an exit poll. While we may view the data exit polls provide as factual, exit poll data is preliminary information based on surveys of voters as they leave the polls.

Get started

The Definition of Exit Polls

Exit polls provide a "snapshot of the electorate" and measure public opinion by asking people how they voted immediately after casting their ballots. Exit polls are distinct from opinion polls in that they measure a voter's response in real-time after the fact rather than predicting votes or opinions. Exit polls are useful because they offer the public an early idea of which candidate is winning and how specific demographics voted. Like other public opinion metrics, exit polls can shape future political campaigns, policies, and laws.

How Exit Polls are Conducted

Trained canvassers conduct exit polls and surveys on Election Day after voters have cast their ballots. These surveys provide valuable information to political analysts and media networks who use exit poll data to project election winners. Each survey records which candidates voters cast their votes for along with important demographic information such as gender, age, level of education, and political affiliation. The canvassers survey approximately 85,000 voters during each exit poll.

In recent years, exit poll workers have also contacted voters by phone. About 16,000 exit polls are conducted in this way to account for early voting, mail-in, and absentee ballots.

The media organizations (e.g., CNN, MSNBC, Fox News) working in partnership with Edison Research regulate the exit polls and determine the questions voters will be asked. Edison Research also decides which polling locations to conduct surveys and hires canvassers to conduct the exit polling. Throughout Election Day, canvassers report their responses to Edison, where the information is analyzed.

However, because exit poll data changes as the day goes, the earliest poll numbers, usually reported around 5:00 pm, are generally unreliable and do not take into account the complete demographic picture. For example, the first wave of exit polls often reflects older voters who tend to vote earlier in the day and do not consider younger, working-age voters who arrive at the precinct later. For this reason, Edison Research cannot glean a clearer picture of which candidates might win until the polls are nearer to closing.

Nevertheless, National Election Pool employees examine the information collected from exit polls in secret. No cell phone or internet access is allowed. After the analysis, employees report to their respective media outlets and share this information with the press.

When the polling has concluded for the day, Edison obtains voting records from a sample of polling locations to examine them side by side with exit poll data. The research company updates the results and disseminates the data to media outlets.

Finally, media outlet "decision desks," comprised of political experts and professional journalists, determine the election results. They work together to project winners using the information from exit polls alongside actual data from the exit polls.

Exit Poll Data 80s StudySmarter

Exit poll data for Blue Collar Voters, 1980 Presidential Election, Wikimedia Commons. Photo by NBC News. Public Domain

Exit Polls: Challenges

Exit polling presents many challenges. Thus, it is important to emphasize that exit polls aren't necessarily a reliable indicator of the winner of an election. Since data changes throughout Election Day, early predictions are often incorrect. As election day progresses and more data is gathered, exit poll data accuracy also increases. Only after the election can it be determined whether or not an exit poll accurately predicted the winners. Mail-in ballots and other factors further compromise the usefulness of exit polls as a predictive tool.

This section will highlight some of the main challenges with exit polling.

Exit Polls: Accuracy

Bias

The main intention of exit polls is to provide information about the success of an elected official's campaign, shed light on who voted for the winner, and provide insight into their support base, not determine election results. Furthermore, like most surveys, exit polls can result in participant bias — when survey data becomes skewed as it relies too heavily on information collected from a similar subset of voters sharing similar demographics.

Participant bias can occur when a polling or research company selects at random a polling precinct that is not as representative of the electorate as was expected, which can lead to polling error.

COVID-19

The COVID-19 pandemic also has complicated exit polling. In 2020, fewer people voted in person, as more voted remotely by mail. As a result, there were fewer voters to conduct exit polls with. Additionally, the 2020 election witnessed a record number of mail-in votes being cast because of the pandemic. In many states, these votes were not counted until days later, making it difficult to make early predictions of election winners.

Methodology

There are doubts about the quality of data obtained in exit polls. Five-Thirty-Eight statistician Nate Silver criticized exit polls as being less accurate than other opinion polls. He also pointed out that while exit polls are supposed to represent voters equally, Democrats more commonly participate in exit polls leading to Democratic bias, further eroding the usefulness of exit polling. It is also important to remember that surveys have inherent flaws and don't 100% accurately represent the entire body of voters.

Democrat Bias in Exit Polling

According to Five-Thirty-Eight, exit polls have routinely overstated the Democrats' vote share. In the 2004 Presidential Election, exit poll results prompted several political pundits to believe John Kerry would be the victor. The exit polls were inaccurate, as George W. Bush eventually came away the winner.

In the 2000 Presidential Election, Democrat Al Gore appeared to be leading in heavily Republican states like Alabama and Georgia. In the end, he lost them both.

Finally, during the 1992 Presidential Election, polling data suggested that Bill Clinton would win Indiana and Texas. Ultimately, Clinton would go on to win the election but lost in those two states.

Polling Place Entrance StudySmarter

A polling location. Wikimedia commons. Photo by Mason Votes. CC-BY-2.0

History of Exit Polling

The history of exit polling spans several decades. In this section we will highlight the evolution of exit polling and retail how the procedure has grown increasingly sophisticated over the years.

1960s and 1970s

The United States first used exit polling in the 1960s. Political and media groups wanted to better understand voter demographics and uncover any variables that might connected to why voters chose certain candidates. The use of exit polls increased in the 1970s and have been employed regularly during elections ever since to help gain insight into voters' decision-making processes.

The 1980s

In the 1980 Presidential Election, NBC used exit poll data to declare Ronald Reagan the winner over incumbent Jimmy Carter. This sparked great controversy because the polls hadn't yet closed when the winner was announced. In the aftermath of this incident, a congressional hearing was held. Media outlets then agreed to forego announcing the election winners until all the polls closed.

The 1990s - Present

During the 1990s, media outlets and the Associated Press created the Voter News Service. This organization enabled the media to access more accurate exit poll information without receiving duplicate reports.

Controversy struck again during the infamous 2000 Presidential election, during which Al Gore's loss was misconstrued by the Voter News Service. They mistakenly announced Gore as the winner over George H. W. Bush. That same evening, the announcement was made that Bush had won. Later, the Voter News Service teeter-tottered again saying the presidential winner was undetermined.

The Voter News Service disbanded in 2002. The National Election Pool, a new polling consortium, was created in 2003, in partnership with mass media outlets. Some mass media networks have left the group since that time. The National Election Pool employs Edison Research to carry out exit polls.

Exit Polls - Key takeaways

Exit polls are public opinion surveys conducted with voters immediately after they cast their ballots.
Originally used in the 1960s, exit polls were designed to provide demographic information about voters.
Today, they are used along with other data to predict election outcomes.
Exit polls differ from opinion polls because they collect data from voters after they vote instead of trying to predict who voters will support in advance of the election.
Exit polls face challenges with accuracy and reliability. They don't accurately predict the winners of elections, the data set changes throughout the election, and participant bias may occur. There may be a bias that favors Democratic voters inherent in exit polling. Further, the impact of the COVID-19 pandemic on top of the margin of error that comes along with any survey impacts their usefulness as a tool in understanding voter behaviors.
Exit polls have incorrectly announced presidential winners on two occasions.

Exit Polls

StudySmarter Editorial Team

The Definition of Exit Polls

How Exit Polls are Conducted

Exit Polls: Challenges

Exit Polls: Accuracy