New CBS polls show suspected Democratic presidential candidate Joe Biden is 6 percentage points ahead of President Donald Trump among likely voters in Wisconsin and Pennsylvania, key battlefield states Hillary Clinton lost by less than 1 point in the 2016 election.
There is still plenty of time to change things before the election and the battlefield state elections should be carried out with a grain of salt. However, the results suggest that Biden may have an advantage in the rust belt states that helped Trump win his 2016 victory – and that Trump’s botched response to the coronavirus crisis plays a prominent role in Biden̵
While the polls YouGov conducted on behalf of CBS between Aug. 4th and 7th show Biden at the top, it’s important to note that the former Vice President’s lead is within the margin of error on both polls, which means that Trump actually might do a poll a bit better than Biden. In Pennsylvania, polls found Biden ahead of Trump with 49 percent support for the 43 percent president. This poll has an error rate of 3.7 percentage points, which means Trump could have up to 46.7 percent support and Biden could only have 45.3 percent support.
The Wisconsin poll, which had an error rate of 3.8 percentage points, found that Biden leads Trump at 48 to 42 percent.
Both Wisconsin and Pennsylvania found that independent voters favor Biden. Clinton lost this group to Trump in both states; and Biden also surpasses Clinton’s share of the vote among white and college-educated voters. It should be noted, however, that the two data points are not directly comparable – while in this case the poll data is from likely voters who may or may not actually vote, the 2016 voting data is from voters who actually did .
Respondents found that views on the coronavirus pandemic are closely related to candidate preference. In fact, when CBS analyzed the survey data, it found that views about the pandemic are more closely related to voting than views about the economy.
“Those who say the Wisconsin’s outbreak is a crisis vote even more in favor of Biden than those who say the economy is very bad. The small group that doesn’t think the outbreak is a big problem has Mr Trump outnumbered voters who say the state’s economy is good, ”the analysis said.
The public’s perception of the president’s pandemic response is highly polarizing – and polls over the past few months have shown the public to see this as the nation’s top concern. CNN poll expert Harry Enten has argued that this is bad news for Trump, as historical poll data suggests that “whoever is most trusted on the non-economic issue is likely to win the election”.
Government polls should be conducted with a grain of salt
Polls in battlefield states are important, especially given that the US presidential election is determined by the electoral college, not a referendum. However, the state polls also have significant limitations, and Biden’s consistent lead in those states (including other states like Michigan, Florida, and North Carolina) should not be taken as a surefire sign of his victory in those states and the general election.
Note that in a poll by the Marquette Law School in Wisconsin in late October 2016, Clinton gained 6 percentage points – the same advantage Biden had in CBS polls in Pennsylvania and Wisconsin – but Trump ultimately won the state by 0.7 Points.
As Vox’s Li Zhou explained, there are many reasons why a number of polls in 2016 did not hit the mark when compared to the final election results. Some of these were corrected during this election cycle – for example, some polls leading up to 2016 overrepresented Clinton voters because they weren’t weighted for education, and that’s no longer the case. (The CBS survey is weighted by education.)
But there are still many obstacles. The poll is always a snapshot of a specific time and ultimately cannot provide any definitive insight into the likelihood that someone who shares their preference with a pollster will actually show up at the polling booth on election day, nor can it necessarily predict the patterns of recent times – breaking voters deciding on their candidate in the final days leading up to the election (which played a crucial role in Trump’s victory).
Adding to the uncertainty is that the pandemic makes predictions based on surveys particularly difficult, as Zhou explains:
In particular, the use of voting by email due to the coronavirus pandemic makes it much more difficult to predict the composition of the electorate. It is unclear how closely voter turnout will align with previous years as there are public health concerns about physical polling stations and questions about the number of people using postal ballot paper instead.
“It’s difficult to make a switch model because you’re not sure who will turn out to be. It’s even more difficult with an election that involves extensive email voting, ”said Lonna Atkeson, professor of political science at the University of New Mexico.
Bottom line: The survey looks promising for Biden, but surveys shouldn’t be confused with perfect predictions of the outcome.
Support Vox’s explanatory journalism
At Vox, we want to answer your most important questions every day and provide you and our audience around the world with information that can save lives. Our mission has never been as important as it is right now: to strengthen it through understanding. Vox’s work reaches more people than ever before, but our distinctive brand of explanatory journalism is draining resources – especially during a pandemic and economic downturn. Your financial contribution does not constitute a donation, but it enables our employees to continue to offer free articles, videos and podcasts in the quality and quantity required for this moment. Please consider contributing to Vox today.