On election night, when will you find out who won?
For the past twenty years, after the polls closed in Britain, at 10 p.m., there has been a surprisingly good answer.
That’s when the three major broadcasters release the results of national exit polls, the work of a group of statisticians and political scientists that begins hours before an election and that in recent years has produced increasingly accurate descriptions of the outcome before the votes are counted.
In the past five UK general elections, exit polls have predicted the winning party to win an average of four seats out of more than 650 in parliament. The last time, in 2019, exit polls predicted a total of just three seats.
Here is a guide on what to expect and how it works.
Let’s start from the beginning. What are exit polls?
It is a survey taken shortly after voters have cast their ballots. In the UK, the survey is conducted as voters leave the polling station: field workers ask more than 20,000 people at around 130 polling stations across the country to fill in copies of their ballot papers. An exit poll has been conducted at every general election in the UK since 2005, paid for by the three major broadcasters: the BBC, ITV and Sky.
Why are British measuring instruments so accurate?
But this was not always the case. In the 1992 general election, BBC exit polls predicted that no party would win an outright majority of seats in Parliament, but early results soon showed that the Conservatives were on track to retain control. Exit polls in some previous elections have been even more inaccurate.
Participants said the key change came in the 21st century when broadcasters focused their resources on adopting statistical methods pioneered by academics David Firth and John Curtice. The success of this approach helped Professor Curtice become Election Radio Star.
Previous exit polls have attempted to collect a representative sample of polling locations in each election and use the vote totals in the sample to predict the vote share of each party elsewhere.
The new polls still seek to have a representative sample, but they also return to the same polling stations as much as possible each time. Now, instead of focusing on totals, researchers can make direct comparisons and study how voting patterns change.
They then used statistical models to predict how the changes they found would play out across the country based on further analysis of each region’s demographics and previous election results.
Jouni Kuha, a professor of social statistics at the London School of Economics who has worked on the exit poll team since 2010, said focusing on the same location was key.
“When you look at these changes, there’s less noise in the data than when you’re trying to estimate the stocks themselves,” he said in a telephone interview.
Professor Feith said not much has changed since the rethink in the early 2000s. “Even the software I wrote in 2001-2005 is still in use today,” he said in an email.
So what could go wrong?
Like all statistical estimates, UK exit polls have a margin of error: around 20 seats.
In a tight race, 20 seats can be a lot. In 2015, the center-right Conservatives unexpectedly won a slim majority in parliament, five years after forming a coalition with the centrist Liberal Democrats. Polls that year underestimated their performance by 15 seats — within the expected margin of error, but enough to falsely suggest they might still need help from another party to govern.
Polls suggest this year’s election will not be close. However, there is still a chance of luck. The polling stations chosen may not be representative.
“People think there’s some kind of magic involved,” Professor Curtis said. told the New York Times “But our strength lies in the data.”
In 2019, before the exit polls were released, I wrote an article addressing the dramatic shift in the UK electoral map. Professor Firth pointed out: “There is nothing in the new method that can guarantee such amazing accuracy!”
The biggest challenge was time pressure. In the UK, most people vote in person on election day, and that day is a weekday, so the number of voters surges in the evening. This leaves a short window of time to collate and analyze data before 10 pm.
The redistricting exercise that has been underway in much of the country since 2019 could also prove difficult this time around.
Why don’t all places have exit polls like this?
American Experts Be cautious with poll results surveysand for good reason.
The main exit polls in the United States are conducted jointly by multiple news organizations (mainly broadcasters) and Edison Research, and are designed to achieve a wider range of goals under more stringent conditions.
Instead of a single question appearing on the mock ballot, voters surveyed typically receive 20 questions that collect demographic and issue data. The survey results are used not only to help project winners, but also to contribute to a broader analysis of why people vote on election night.
There is also a major obstacle to replicating the UK approach: absentee and early voting are much more common in the US. 41% The 2016 vote was as follows; 70% In 2020, by contrast twenty one% In the last election in the U.K., exit polls in the U.S. reflected this, using telephone, email and text message surveys as well as face-to-face conversations with voters.
“While our work in 2004, 2016, and 2020 was marred by specific errors in particular states and campaigns, the overall average error in the survey is smaller than it was in decades,” Joe Lenski, co-founder of Edison Research, noted in the report. Interviews in 2021 “The real issue is educating people about what level of precision we can and cannot demand from these data,” said the American Enterprise Institute researcher.
Complaints about the voting results have been even louder elsewhere. India’s election this year has seen huge stock market swings. Allegations of election fraud Exit polls wrongly predicted a strong majority for the incumbent Bharatiya Janata Party, but in fact the party Forced to form a coalition government.
Does having a really good exit poll result make any difference?
Election night is not usually that volatile, and some British viewers will Turn off the TV coverage The exit polls were taken immediately after they were concluded.
But it can still provide entertaining moments. In a 2015 BBC election broadcast, former Liberal Democrat leader Paddy Ashdown scoffed at predictions that his party would be left with just 10 seats, down from 57. “If this exit poll is correct,” he said, “I will publicly apologise for not appearing on your show.”
Ultimately, the Liberal Democrats won eight seats, and the BBC awarded Mr Ashdown Hat shaped cake.
For Professor Kuhar and his team, the crunch time came just minutes before the 10pm deadline. “It was a very strange experience for academics who are used to a different time scale,” he said. “So it was a bit stressful, but also exciting.”
