In the 2016 US presidential election, both national and state surveys left the impression that election night would bring clear Democratic gains, but this was not the case.
Pollsters indeed underestimated Donald Trump’s support in critical swing states that year. However, the national polling averages were fairly accurate compared to the popular vote, which Hillary Clinton won. However, the national polls in 2020 had an even larger margin of error.
The American Association for Public Opinion Research, a panel of the polling industry’s professional group, said the public opinion surveys ahead of the 2020 presidential election were the most inaccurate in 40 years.
“The problem is substantially a matter of reaching respondents who are reluctant to participate in polls,’’ said Charles Franklin, a professor of law and public policy at Marquette University who has run the Marquette Law School survey since 2012.
Pollsters believe they have mostly pinpointed the reasons behind the 2016 polling errors. According to a report from a group of experts published in Public Opinion Quarterly, a key factor was the oversight of voters’ education levels.
Charles Franklin, a polling expert, explained in an interview with Euronews that Trump’s supporters exhibit a unique distrust toward politics, the news media and polling organisations, a factor that may have contributed to inaccuracies in 2016 and 2020.
To further address the issue of underrepresented Trump supporters, some pollsters are now weighing their data based on the 2020 election results, aiming to better reflect the views of voters who may be less inclined to respond to surveys.
“Pollsters have adopted new methods of reaching samples using email and text in addition to traditional telephone calls,” Franklin explained.
These expanded contact methods are complemented by modified sampling procedures intended to reach respondents who have previously shown reluctance to participate, especially in areas with strong past support for Trump.
Many pollsters have adjusted their methodologies, largely due to the inaccuracies observed in the 2020 election.
“We will know after the election iff these efforts have worked,” Franklin said.