ESOMAR’s “The Pollsters’ Survival Kit for Election Night… And the Days After” webinar Monday, October 12 illustrates how divided America is weeks before the 2020 Presidential election.
Based on an Economist/YouGov poll of 1,363 likely voters conducted October 4-6, 59% of Joe Biden voters will vote by mail, while 59% of Donald Trump voters will vote in-person on Election Day. The respondents also said they anticipated potential long delays on the overall final results.
Moderator Kathy Frankovic, a member of ESOMAR’s Professional Standards Committee and former Director of Surveys at CBS News, suggested the “toss-up” States are Arizona, Ohio, Georgia, Florida, and North Carolina, and that the weight of the massive population in California on past Presidential polls in relation to the actual Electoral College vote has been addressed by pollsters. She cited improved respondent profile weighting to properly balance the profile of respondents who are “likely voters”. Those that will show up on voting day or vote by mail — the population base for most election polls.
Clifford Young, President-Public Affairs at Ipsos USA, emphasized that the U.S. election is reflective of 50 state-by-state elections and that “the constellation of indicators has never been more complex.” However, he did posit that polls become more predictive the closer we get to Election Day. All the pollsters noted the stability of the polls over the past several months and that about 9 million Americans have already voted by mail to date. So, are the current results likely to hold?
Ipsos’ Young underscored that COVID-19, and its handling is unequivocally the No. 1 issue, restoring trust in government is second, and that Biden currently leads by 11 points on both issues. A strong economy and job creation are the No. 3 voter concern, for which Trump currently has a three-point lead. The relative importance of these issues is quite different for Trumpers versus democrats naturally.
Ipsos’ Young said the states to watch – Arizona, Florida, Michigan, North Carolina, Pennsylvania, and Wisconsin – are all currently leaning toward Biden.
Jean-Marc Leger, President of Leger, Market Research and Analytics, underlined his earlier position that turnout will be the cornerstone noting that it is closely related to “enthusiasm” levels that boost it. Likely voters of either candidate are showing remarkably high interest in the election: close to 90% for each group. Levels of the various measures of enthusiasm have all declined for Trump but have increased for Biden.
Leger echoed that Trump supporters “are far less afraid of the virus at the ballot box” as well as the huge mail-in versus in-person disparity between the parties in voting method. He also revealed that for 94% of likely voters, “my choice is final.” So, with three weeks to go, only 6% are likely to change!
He agreed with Ipsos’ Young that the Bill Clinton-era campaign mantra – “It’s the economy, stupid,” has been eclipsed by a more contemporary, “It’s Covid, stupid.” Jean-Marc also opined, “Usually an election starts with reason and ends in emotion. This time it’s the opposite.” So, will marginal extraordinary voters that did not vote in 2016 (or in 2018 for the Mid Terms) turn out in 2020?
Joe Lenski, co-founder and EVP, Edison Research, revealed a potential significant difference in final voting methods in 2020 between the two camps and the possible profile based on extensive experience from exit polling. Edison completed exit poll interviews of 37K democratic voters in 23 States during the Primaries earlier in the year. He said Edison estimates that in-person voting on Election Day will account for 40-50%, that in-person early voting will represent 15-20%, and absentee/vote-by-mail voters will be 40-45% of the final vote.
Due to COVID, Edison has established very very safe ways for the surveyors and the voters to complete in-person exit interviews and the overcome differential non-response. They will also complete telephone interviews with absentee/by mail voters. Non-respondents are profiled.
Lenski made it clear that with States having different procedures on how they process mail-in ballots. The final results in some States could take days, if not weeks. Some States have pre-processing ‘ready to count’ – Arizona, Florida, Ohio, North Carolina, and Virginia; some not. Apparently, some States have already been overwhelmed and remember that the deadline for acceptance can be ‘received by Election Day’ or ‘postmarked by Election Day’. It differs by State. He identified New York, Michigan, Wisconsin, and Pennsylvania as potentially the slowest to have final results of the largest States.
Based on these anomalies and the projected high mail-in proportion of votes cast, he noted that “some of the broadcast networks would be reporting results based on the “expected” vote”. Media predicted results may be made well off the ultimate 100% of votes cast in the initial days for some States. “Expect some result changes either way” as the mail-in votes are completed and reported!
Deborah Mattinson, founding partner of BritainThinks, offered potential parallels between the recent U.K. election (December 2019), in which the Labour party were trounced notably in their “Red Wall” stronghold areas, and the current U.S. situation, notably in the Rust Belt.
Do they hold insights for the Democrats? Labour lost in the Red Wall areas due to being taken for granted and not being wooed, failing to show empathy and respect, and lack of expected resources and investment in the local areas. I suggest her Red Wall parallel may well be more reflective of the 2016 Presidential Election although Labour’s omissions were fundamental strategic mistakes.
Mattinson’s most impactful U.K. research result was that 84% of those who wanted to remain in the European Union (remember that debacle) want Biden as the next U.S. President. Interestingly in the Red Wall strongholds two thirds wanted to leave the EU. Her analysis clearly does reflect that “both countries are divided. And that trend in both countries is increasing. The demographic and psychographic profile comparisons between the UK “remain” and the “progressive Democrats” would be fascinating. Maybe Biden could engineer the U.K.’s return to the EU?
Lenski indicated that Florida, North Carolina, and Texas will have close to 100% of the results by around midnight on November 3 with the mail-ins being reported early evening first and then the addition of the in-person. These States could make the ultimate result clear. He theorized that, “If Biden takes Florida and North Carolina he will win the Presidency!” If it is muddy or too close to call, Michigan, Wisconsin, and Pennsylvania become critical, and results from these states “may take a while.” In these States, the mail-ins would be revealed after the in-person so initial results could flip completely.
Already the Florida Democrats have requested ~50% more mail-in ballots, 2.5 million, than the Republicans, 1.7 million, have return rate of 28.1% to 22.8% as of October 9th. No party affiliation mail ballot requests were 1.2 million with 20.7 returned. Will some holding mail-in ballots actually vote early in-person or even in-person on Election Day?
Florida, despite its disparate collection of ethnicities and cultures, remains a bellwether State on Election Day. How the various Hispanic groups (Cubans & Venezuelans – lean GOP versus Puerto Ricans and Lin-Manuel Miranda – heavily Dems, Mexicans, etc.) vote, the turnout levels in the suburbs versus the rural areas, the simple age/sex and education breaks, etc. will be revealed in the Edison exit polls. They will provide invaluable clues to how the night and the next many days will ultimately turn out not only for the Presidency but for the all-important races in the Senate and the House.
The inevitable confirmation of Amy Coney Barrett to the Supreme Court by the Republican controlled Senate during an Election period against all precedents is likely before Election Day, November 3rd. Consequently, the control of the Senate from January 2021 becomes crucial to the Democrats in countering a US political shift that has moved even further to the right. It must be appreciated that based on a sophisticated report in New York Times based on an analysis of political party manifestos, the centre of the US political spectrum is already well right of the centre of those in most European countries and Canada.
Sadly, the whole US Election process and post-election shenanigans promised by the occupant of the White House may evolve into a brutal legal travesty for a democracy already under siege. However, as the Lincoln Project (Republicans for Biden) has championed, the Election is about saving America and its Democracy.
The post analysis of the US polls will provide tremendous insights for their continuing refinement and how to overcome extremely tough conditions and on-the-ground circumstances.
This is an enhanced version of an Op Ed in Media Post, October 13, 2020 https://www.mediapost.com/publications/article/356773/us-election-pollsters-keep-an-eye-on-florida.html
‘The Pollsters’ Survival Kit for Election Night…and the Days After‘ webinar is now available on demand and can be accessed it via this link (ed.)
