Public Wise partnered with Change Research to poll BIPOC voters in Arizona, Georgia, North Carolina and Pennsylvania.
Change Research surveyed 4,814 likely voters from August 12 – 19, 2020 to yield a representative sample in Arizona, Georgia, North Carolina, and Pennsylvania. The margin of error is ± 1.3 for the full sample and larger among subgroups.
Key Findings
- There is a five-fold growth in voters’ plans to vote-by-mail or absentee ballot in the traditionally lower vote-by-mail states of Georgia, North Carolina, and Pennsylvania, from 4% historically to 20% who intend to vote-by-mail in 2020.
- However, 39% have not requested or received vote-by-mail ballots and 25% feel that they know nothing or little about their voting options for 2020.
- The data demonstrate that there are significant opportunities to create campaigns on the security and ease of both vote-by-mail and in-person early voting, especially among Democratic, Independent, and African American voters:
- Early in-person voting has a history in Georgia and North Carolina among Democratic and African American voters. The messaging on convenience addresses these voters’ concerns on mail-in voting and coronavirus concerns around in-person Election Day voting.
- Pennsylvania has no history of early in-person voting, but a sharp growth in mail-in voting interest among Democrats, from 6% historically to 48% in 2020. Mail-in voting is the best option to address voter concerns among Democrats.
- Voters in Arizona have a strong history of mail-in voting with only a 4% shift towards in-person voting. Promoting mail-in voting with messaging related to its convenience remains the best option here.
- Voters across demographics report being extraordinarily motivated to vote this year in comparison to previous years. 96% are an 8 out of 10 in terms of their motivation or higher. Voters are likely to be receptive to messaging that helps them engage that motivation and intention.
Voting Choices and Vote-by-Mail
42% of voters overall believe that contracting the coronavirus is a very serious or serious concern with in-person voting. The concern increases to 60% among Hispanic voters, 67% among African Americans, and 78% among Democrats. Long lines are also a concern among 50% to 60% of voters in these three groups. However, getting time off from work and transportation to the polling sites are a material concern to only 4% to 8% of voters.
There are concerns about mail-in-voting as well. While it is not a surprise that 70% of Republicans are concerned about the safety and security of vote-by-mail, 58% of Independents, 37% of African Americans, and 24% of Democrats also share this concern.
In the historically low mail-in or absentee voting states of Georgia, North Carolina, and Pennsylvania, there is a five-fold jump from 4% of voters usually voting by mail or absentee to 20% who intend to vote that way in 2020. This shift is from in-person on Election Day to vote-by-mail with in-person early voting at similar levels. Arizona has a strong history of mail-in-voting and there is no material change.
For voters who intend to vote-by-mail, 39% have not requested or received their mail-in ballot.

In Georgia and North Carolina, early in-person voting has been the usual method of voting for 26% to 33% of voters, rising to 32% to 43% among African Americans and 31% to 50% among Democrats.
- Arizona has always been a high mail-in-voting state and voters show only a 4% shift from mail-in voting to in-person Election Day or in-person early voting.
- Georgia and North Carolina have a strong history of early in-person voting, especially among African Americans and Democrats. Even though there is an increase in the desire to vote-by-mail, a more successful approach may be to continue to emphasize the early in-person voting that voters are comfortable with.
- Pennsylvania has little history of mail-in voting and none for early in-person voting. Mail-in voting has been made easier in 2020. Considering the increased interest in Pennsylvania in mail-in voting from 6% to 48% among Democrats, and from 1% to 24% among African Americans, awareness and promotion of mail-in voting is the best option.
About 25% of the voters overall, and 25% of persons of color, know little or nothing about the options that they have for voting this year and how they work.
State Specific Voting Plans
Arizona
- All voters: 49% by mail, 35% in person EDay, 11% early voting
- As is generally true elsewhere in 2020, Democrats lean 73% towards voting by mail vs. 9% in person EDay
- Republicans lean 57% in person EDay and 29% by mail
- Independents: 40% by mail and 45% in person EDay
- Voters 65 or older prefer mail in voting over in person Eday 60% to 26%.
Georgia
- All voters: 49% in person EDay, 29% in person early, and 17% by mail
- Democrats: 28% in person EDay, 31% in person early, and 33% by mail.
- Republicans intend to vote 67% in person, 27% in person early, and 4% by mail.
North Carolina
- All voters: 49% in person EDay, 33% in person early, 11% and by mail
- Democrats: 22% in person EDay, 48% in person early, and 22% by mail
- Republicans: 74% in person EDay, 22% in person early, and 3% by mail.
Pennsylvania
- All voters: 69% in person EDay, 25% by mail
- Democrats: 42% in person EDay, 48% by mail
- Republicans: 94% in person EDay, 5% by mail.
Motivation to Vote
Voters report being remarkably and extraordinarily engaged in voting this year in comparison to previous years. 92% are a 10 out of 10 in terms of their motivation and 96% are an 8 or higher. 91% of voters of color are a 10/10 with Democratic women the highest at 96% a 10/10. While voters are overly optimistic in predicting their own chances of voting, voters are likely to be extraordinarily receptive to messaging that helps them engage that motivation and intention.
84% have also thought a great deal about how they’re going to vote in the past month and those numbers span demographic groups. 87% believe that their vote definitely matters and when including those who say it probably matters that’s 96% of the electorate. More than a third would even be interested in serving as a poll worker this year if asked.

Over 95% of all voters across various demographics believe that their vote in November definitely or probably matters. Likewise, over 95% of voters responded with an 8 or higher motivation to vote in 2020 on a 1 to 10 scale.
Even though 44% of Democrats, and 34% of voters overall say that they have encountered false or misleading claims about voting this year, it has not affected their motivation to vote.
Coronavirus
A 59% majority of voters view the coronavirus as a very serious or somewhat serious concern, with 71% of Hispanics, 85% of African Americans, and 94% of Democratic voters in agreement, but only 28% among Republicans. Likewise, 79% of Democratic and only 23% of Republican believe that the virus will still be a major concern in November.
State-Specific Information
Arizona
Presentation | Crosstabs
Change Research surveyed 1,160 likely voters online across Arizona. Change Research uses Dynamic Online Sampling to recruit a representative sample. Weighting was done on gender, age, race, education, geographic region, and 2016 vote. The margin of error is ± 2.9% for the full sample and larger among subgroups.
Georgia
Presentation | Crosstabs
Change Research surveyed 1,393 likely voters online across Georgia. Change Research uses Dynamic Online Sampling to recruit a representative sample. Weighting was done on gender, age, race, education, geographic region, and 2016 vote. The margin of error is ± 2.6% for the full sample and larger among subgroups.
North Carolina
Presentation | Crosstabs
Change Research surveyed 1,077 likely voters online across North Carolina. Change Research uses Dynamic Online Sampling to recruit a representative sample. Weighting was done on gender, age, race, education, geographic region, and 2016 vote. The margin of error is ± 3.0% for the full sample and larger among subgroups.
Pennsylvania
Presentation | Crosstabs
Change Research surveyed 1,093 likely voters online across Pennsylvania. Change Research uses Dynamic Online Sampling to recruit a representative sample. Weighting was done on gender, age, race, education, geographic region, and 2016 vote. The margin of error is ± 3.0% for the full sample and larger among subgroups.
Share Findings
We are happy to share findings and data free of charge to any organizations.
Please feel free to contact info@publicwise.org.