Democrats are twice as likely as Republicans to worry about in-person voting — with nearly two in three seeing it as a large or moderate risk to their health — according to this week's installment of the Axios-Ipsos Coronavirus Index.
Why it matters: This could pose a significant disadvantage for Joe Biden and other Democratic candidates in November if the pattern holds — especially in states where high infection rates persist, or where there are significant hurdles to mail-in, absentee or early voting.
- It doesn't necessarily mean those Democrats won't vote in person. Many of them could decide it's worth the risk.
- But their fears almost certainly increase the odds that some won't vote in person — and if those voters don't have good alternatives, that could present Biden and other Democrats with a turnout challenge they don't need.
What they're saying: Cliff Young, president of Ipsos U.S. Public Affairs, says these risk perceptions around voting may have less to do with conditions inside particular polling locations than with the general gulf between President Trump's supporters and other Americans when it comes to most aspects of the pandemic.
- "I think what we're seeing is a more generalized sense of risk," Young says. "Democrats and independents tend to be closer together on the risk stuff, and Republicans are father away. Voting is just one of those domains."
- "This goes to Trump's narrative, which is, 'Why do we have to make any exception?' 'It's not that dangerous.' 'Oh, they just want to steal the election.'"
By the numbers: Overall, 52% of the respondents in Week 19 of our national survey ranked in-person voting as risky. Just 21% said it was "very risky" — a level of concern higher than shopping in grocery stores, but less intense than the idea of sending their children back to school, or going to see family or friends outside the home.
- But it's the big differences by party ID — more than age, gender, race or geography — that shows how unevenly this could impact election contests.
- 64% of Democrats, 59% of independents and just 29% of Republicans see in-person voting as very or somewhat risky.
- More women than men see it as risky, 57% to 47%.
- 65% of Hispanic, 63% of Black and 45% of white respondents see it as risky.
- Age matters in terms of risk perception, but less than you might think: 56% of seniors 65 and older saw it as risky, compared with 46% of respondents ages 18-29.
Between the lines: Perceiving that an activity is risky doesn't mean people won't do it.
- We saw this in earlier surveys during the height of the protests over the killing of George Floyd. Some respondents saw significant risk in demonstrating but decided the cause was important enough to do it anyway, and wore masks and gloves to protect themselves.
- One key question is whether voters who don't have good alternatives to in-person voting will decide the election is worth the risk they believe they'd be taking.
- Another is whether they feel they can sufficiently protect themselves with precautions such as a mask and gloves.
- "Just because it's risky doesn't mean they can't deal with it," Young says. But they may respond more if their concerns about how to "vote safely" are acknowledged. "If your message is that it's not that big a deal, you're falling on deaf ears except with Republicans."
The big picture: This week's survey captures new highs and lows as the U.S. approaches 4.7 million cases and surpasses 155,000 deaths.
- 89% of Americans now say they wear a mask all the time or sometimes outside the home, a new record in our poll.
- In another milestone, 19% now say they know someone who has died from the virus — and one-fourth of those were family members.
- At the same time, only 17% say they are self-quarantining, staying home and avoiding contact with others for 14 days, the lowest share since mid-March.
Methodology: This Axios/Ipsos Poll was conducted July 31-August 3 by Ipsos' KnowledgePanel®. This poll is based on a nationally representative probability sample of 1,129 general population adults age 18 or older.
- The margin of sampling error is ±3.0 percentage points at the 95% confidence level, for results based on the entire sample of adults.