Forty-nine percent of likely Minnesota voters support former Vice President Joe Biden, while 44 percent support President Donald Trump, a poll by Change Research commissioned by MinnPost finds.
The online poll was conducted among 1,021 likely voters in Minnesota last week, from Oct. 12 to Oct. 15, and the results’ margin of sampling error is 3.1 percentage points.
Separated by just five points and within the margin of error, this represents the closest the two presidential candidates have been in a recent public poll.
Fewer pollsters have surveyed Minnesota lately compared to in August and early September, when the state had a brief flirtation with national media outlets as a more than remote possibility for flipping from blue to red. On Aug. 31, FiveThirtyEight gave Biden a 69 percent chance of winning the state, giving Trump a little less than a third, the closest the site has had the race before or since.
More recently, pundits have placed Minnesota in the much more likely blue column. As of Tuesday, FiveThirtyEight gave Biden a 92 percent chance of winning the state, his highest odds since the race began. The site’s polling average put Biden at 50.7 percent, compared to Trump’s 41.6 percent — a spread of 9.1 points. Polls done in September and October have suggested Biden ahead by as little as 4 points and as many as 17.
Some recent polls suggest the race may be tightening as Election Day nears, though. A SurveyUSA poll released in early October found Biden ahead of Trump by 7 points, a slightly larger margin than the one in the MinnPost/Change Research poll, compared to 9 points in early September.
Change Research’s methodology involves targeting survey takers based on geography, age, gender, race and partisan identification via advertisements on websites and on social media. The company argues this allows it to “cast a net that is wider” than pollsters who rely on phone-based surveys. This poll was weighted — meaning the results were corrected to better represent the electorate — by gender, race/ethnicity, age, education and 2016 presidential vote.
Online polls have gained popularity as it’s gotten more difficult, and more expensive, to get poll respondents to pick up the phone. But critics of this method say that probability-based random sampling achieved by random-digit phone dialing is still the gold standard for building a sample that accurately reflects the electorate, and that online polls that use non-probability based samples don’t achieve results with the same level of fidelity.
Smith versus Lewis
The poll also found that 48 percent of likely voters favor incumbent DFL Sen. Tina Smith, compared to 44 percent for her challenger, Republican former congressman and former conservative talk show host Jason Lewis, and 4 percent in favor of other candidates and five percent not sure. This also represents the closest these two candidates have been in a recent public poll.
Sixteen percent of likely voters had never heard of Lewis, while 9 percent had never heard of Smith.
Little national attention is being paid to Minnesota’s U.S. Senate race. The nonpartisan Cook Political Report rates the seat solidly Democratic, and the race between Smith and Lewis is seeing scant outside spending.
Early vs. Election Day votes
This election year, a record number of Minnesotans are expected to vote early or by mail, a side effect of the COVID-19 pandemic and encouragement by election officials and Democrats to keep traffic at polling places down.
This poll finds a stark divide in partisan preferences between people who have already voted and those who have not.
Of those who already voted, 79 percent said they had voted for Biden, while 17 percent said they had voted for Trump. In the Senate race, 76 percent said they had voted for Smith, compared to 17 percent for Lewis.
Those who had not yet voted heavily favored Republicans: 61 percent said they would vote for Trump if the election were held today, and 60 percent said they would vote for Lewis.
This poll also found Minnesotans split on how best to deal with the pandemic.
Nearly two-thirds of respondents (61 percent) said they had very or somewhat serious concerns about the coronavirus outbreak, while 39 percent said they had minor concerns or no concerns at all.
Forty-one percent favored taking all possible precautions to limit the spread of COVID-19 and waiting as long as it takes before safely resuming normal life, compared to 31 percent who said everyone is making too big a deal of COVID-19 and normal life should be resumed even if some get sick. Twenty-eight percent favored keeping some precautions in place while returning to work and public places to keep the economy afloat.
Sixty percent of respondents strongly or somewhat supported mandating masks in indoor public places, while 39 percent strongly or somewhat opposed such a mandate. Fifty-seven percent strongly or somewhat supported in-person school resuming in their county, compared to 31 percent who strongly or somewhat oppose it.
Sixty percent of respondents were very or somewhat confident that a vaccine that may be approved by the U.S. Food and Drug Administration would be safe and effective, compared to 41 percent who were not very or not at all confident.