MILWAUKEE – A new Marquette Law School Poll in Wisconsin finds slight change in voting preferences or attitudes in the wake of shootings and protests in Kenosha, Wisconsin, in late August.
In early September, Democratic former Vice President Joe Biden is the choice of 47 percent of likely voters and Republican President Donald Trump is supported by 43 percent. Libertarian candidate Jo Jorgensen is the choice of 4 percent, while 7 percent say they would vote for none of these candidates, didn’t know how they would vote or declined to say.
In August, before the events in Kenosha, Biden was supported by 49 percent and Trump by 44 percent, with 6 percent not choosing either. Jorgensen was not included in the August poll. In June among likely voters, Biden had 50 percent and Trump 44 percent, with 6 percent not choosing either.
Other findings from the new poll include:
- Approval of Black Lives Matter protests barely changed following the Kenosha events.
- Approval of President Trump’s response to protests rose following his visit to Kenosha among Republicans but shifted only slightly among other voters.
- Gov. Tony Evers’ job approval has fallen to pre-coronavirus levels after a period of increased approval.
- Parents of school-age children have grown more uncomfortable with reopening schools.
- While two-thirds of respondents say they would definitely or probably get a vaccine for COVID-19, a third say they would probably or definitely not get vaccinated.
The poll was conducted from Aug. 30 to Sept. 3, 2020. The sample included 802 registered voters in Wisconsin interviewed by cell phone or landline, with a margin of error of +/- 4 percentage points. There are 688 likely voters, with a margin of error of +/- 4.3 percentage points.
Table 1 shows the trend in the presidential vote among likely voters from May through September. Table 2 shows the trend in preferences among all registered voters over the same period. Jorgensen was not included before this September poll. (Note: Likely voters are those who say they are certain to vote in November. Registered voters refer to all respondents who are registered to vote or who say they will register by the election. Tables below are based on registered voters unless “likely voters” is indicated in the heading.)
Table 1: Vote among likely voters, May–September 2020
| Poll Dates | Joe Biden | Donald Trump | Jo Jorgensen | None/other (VOL) | Don’t know | Refused | n |
| 5/3-7/20 | 49 | 45 | NA | 3 | 2 | 1 | 650 |
| 6/14-18/20 | 50 | 44 | NA | 3 | 2 | 1 | 686 |
| 8/4-9/20 | 49 | 44 | NA | 3 | 2 | 1 | 675 |
| 8/30-9/3/20 | 47 | 43 | 4 | 2 | 3 | 2 | 688 |
Table 2: Vote among registered voters, May–September 2020
| Poll Dates | Joe Biden | Donald Trump | Jo Jorgensen | None/other (VOL) | Don’t know | Refused | n |
| 5/3-7/20 | 46 | 43 | NA | 4 | 4 | 2 | 811 |
| 6/14-18/20 | 49 | 41 | NA | 5 | 3 | 2 | 805 |
| 8/4-9/20 | 48 | 42 | NA | 6 | 3 | 1 | 801 |
| 8/30-9/3/20 | 46 | 40 | 5 | 3 | 5 | 2 | 802 |
Views of protests, BLM, and police after Kenosha
Approval of protests over police shootings of Black Americans declined among registered voters from June to early August, prior to events in Kenosha, but barely moved following the Kenosha shootings and protests, as shown in Table 3.
Table 3: Approval of protests against police shootings, June–September 2020
| Poll Dates | Approve | Disapprove | Don’t know | n |
| 6/14-18/20 | 61 | 36 | 2 | 805 |
| 8/4-9/20 | 48 | 48 | 3 | 801 |
| 8/30-9/3/20 | 47 | 48 | 4 | 802 |
Favorable views of the Black Lives Matter movement also declined from June to August, but did not change further in September, as shown in Table 4.
Table 4: Favorable or unfavorable view of Black Lives Matter movement, June–September 2020
| Poll Dates | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 6/14-18/20 | 59 | 27 | 6 | 5 | 805 |
| 8/4-9/20 | 49 | 37 | 10 | 5 | 801 |
| 8/30-9/3/20 | 49 | 37 | 8 | 5 | 802 |
There were small changes in favorable and unfavorable views of the police from June to September, shown in Table 5. Favorable views rose slightly, from 72 percent in June to 76 percent in August, falling to 73 percent in September. Unfavorable views of the police were 18 percent in June, 13 percent in August and 18 percent in September.
Table 5: Favorable or unfavorable view of the police, June–September 2020
| Poll Dates | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 6/14-18/20 | 72 | 18 | 3 | 5 | 805 |
| 8/4-9/20 | 76 | 13 | 5 | 6 | 801 |
| 8/30-9/3/20 | 73 | 18 | 5 | 4 | 802 |
Approval of Trump’s handling of protests
Approval of Trump’s handling of protests rose in September to 36 percent from 32 percent in August, while disapproval declined from 58 percent in August to 54 percent in September. Table 6 shows the trend in approval since June.
Table 6: Approve or disapprove of Trump’s handling of protests, June–September 2020
| Poll Dates | Approve | Disapprove | Don’t know | n |
| 6/14-18/20 | 30 | 58 | 11 | 805 |
| 8/4-9/20 | 32 | 58 | 9 | 801 |
| 8/30-9/3/20 | 36 | 54 | 8 | 802 |
Trump’s visit to Kenosha occurred during the field period of the poll, with 441 respondents interviewed before his visit, on Sunday and Monday, Aug. 30-31, and 361 interviewed on Tuesday-Thursday, Sept. 1-3, following the visit. Tuesday interviews were conducted in the evening after the conclusion of the president’s visit.
Approval of Trump’s handling of protests rose slightly following his visit to Kenosha, although the difference is not statistically significant, as shown in Table 7.
Table 7: Approve or disapprove of Trump’s handling of protests, pre- and post-Trump Kenosha visit, September 2020
| Pre-Post Trump visit | Approve | Disapprove | Don’t know | n |
| Pre Trump visit | 34 | 54 | 9 | 441 |
| Post Trump visit | 38 | 53 | 7 | 361 |
Reactions to Trump’s visit varied by party. Republicans were more approving of Trump’s response to protests after his visit, and to a statistically significant degree. Independents and Democrats, by contrast, barely budged following his visit. These results are shown in Table 8.
Table 8: Approve or disapprove of Trump’s handling of protests, pre- and post-Trump Kenosha visit by party identification, September 2020
| Pre-Post Trump visit | Party ID | Approve | Disapprove | Don’t know | n |
| Pre Trump visit | Republican | 65 | 20 | 13 | 134 |
| Post Trump visit | Republican | 87 | 6 | 6 | 104 |
| Pre Trump visit | Independent | 28 | 59 | 9 | 186 |
| Post Trump visit | Independent | 31 | 62 | 6 | 141 |
| Pre Trump visit | Democrat | 5 | 90 | 5 | 111 |
| Post Trump visit | Democrat | 4 | 88 | 8 | 113 |
Views of Trump, Biden, Pence and Harris
Favorable and unfavorable views of the presidential candidates have been stable in recent months. Trump has held a 42 percent favorable rating since June, with 54 or 55 percent unfavorable, while Biden’s favorable rating has varied between 42 and 45 percent since May, with 46-48 percent giving him an unfavorable rating. Trump’s net favorable rating has been lower in the last three months than in the winter and spring, while Biden’s net favorability rating was lower in the winter than it has been since March. The full trends are shown in Tables 9 and 10.
Table 9: Favorable or unfavorable view of Trump, January–September 2020
| Poll Dates | Net | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 1/8-12/20 | -5 | 46 | 51 | 0 | 1 | 800 |
| 2/19-23/20 | -5 | 45 | 50 | 3 | 2 | 1000 |
| 3/24-29/20 | -5 | 45 | 50 | 2 | 3 | 813 |
| 5/3-7/20 | -7 | 44 | 51 | 2 | 2 | 811 |
| 6/14-18/20 | -12 | 42 | 54 | 2 | 2 | 805 |
| 8/4-9/20 | -13 | 42 | 55 | 2 | 1 | 801 |
| 8/30-9/3/20 | -12 | 42 | 54 | 2 | 1 | 802 |
Table 10: Favorable or unfavorable view of Biden, January–September 2020
| Poll Dates | Net | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 1/8-12/20 | -5 | 41 | 46 | 8 | 4 | 800 |
| 2/19-23/20 | -19 | 34 | 53 | 8 | 4 | 1000 |
| 3/24-29/20 | -11 | 39 | 50 | 7 | 4 | 813 |
| 5/3-7/20 | -4 | 42 | 46 | 7 | 4 | 811 |
| 6/14-18/20 | -2 | 44 | 46 | 7 | 2 | 805 |
| 8/4-9/20 | -5 | 43 | 48 | 7 | 2 | 801 |
| 8/30-9/3/20 | -2 | 45 | 47 | 6 | 1 | 802 |
The vice-presidential candidates are less familiar to respondents, with 15 percent unable to give a favorable or unfavorable opinion of Republican Vice President Mike Pence and 24 percent unable to give an opinion of Democratic Sen. Kamala Harris.
Table 11 shows favorable and unfavorable ratings for Pence. The Marquette Law School Poll has asked about Pence only twice since 2017. Table 12 shows favorability ratings for Harris since August 2019, when she was a candidate in the Democratic presidential primaries.
Table 11: Favorable or unfavorable view of Pence, 2017–2020
| Poll Dates | Net | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 6/22-25/17 | 2 | 37 | 35 | 26 | 3 | 800 |
| 8/30-9/3/20 | -7 | 39 | 46 | 12 | 3 | 802 |
Table 12: Favorable or unfavorable view of Harris, August 2019–September 2020
| Poll Dates | Net | Favorable | Unfavorable | Haven’t heard enough | Don’t know | n |
| 8/25-29/19 | -14 | 20 | 34 | 35 | 10 | 800 |
| 10/13-17/19 | -8 | 24 | 32 | 33 | 11 | 799 |
| 8/30-9/3/20 | 1 | 38 | 37 | 20 | 4 | 802 |
Trump and Biden: Do they care about people like you?
Tables 13 and 14 show the perception that Trump and Biden “cares about people like me,” measured in June, and again in September. There has been little movement in how either candidate is viewed following the party conventions in August, with a majority saying “cares about people like me” does not describe Trump and a slight plurality saying the phrase does describe Biden.
Table 13: Does “cares about people like me” describe Trump, June–September 2020
| Poll Dates | Describes | Does not describe | Don’t know | n |
| 6/14-18/20 | 39 | 57 | 3 | 805 |
| 8/30-9/3/20 | 41 | 56 | 3 | 802 |
Table 14: Does “cares about people like me” describe Biden, June–September 2020
| Poll Dates | Describes him | Does not describe him | Don’t know | n |
| 6/14-18/20 | 46 | 42 | 10 | 805 |
| 8/30-9/3/20 | 48 | 45 | 5 | 802 |
Trump job approval
Approval of how Trump is handling his job as president in polls since May is shown in Table 15. As of September, 44 percent approve and 54 percent disapprove, the same as his approval rating in August.
Table 15: Approve or disapprove of Trump’s handling of his job as president, May–September 2020
| Poll Dates | Approve | Disapprove | Don’t know | n |
| 5/3-7/20 | 47 | 49 | 3 | 811 |
| 6/14-18/20 | 45 | 51 | 3 | 805 |
| 8/4-9/20 | 44 | 54 | 2 | 801 |
| 8/30-9/3/20 | 44 | 54 | 2 | 802 |
Trump’s handling of the economy
Handling of the economy remains Trump’s strongest area of approval. The September poll shows a one-point increase in approval and a two-point decline in disapproval, as shown in Table 16.
Table 16: Approve or disapprove of Trump’s handling of the economy, May–September 2020
| Poll Dates | Approve | Disapprove | Don’t know | n |
| 5/3-7/20 | 54 | 40 | 4 | 811 |
| 6/14-18/20 | 50 | 46 | 3 | 805 |
| 8/4-9/20 | 51 | 46 | 3 | 801 |
| 8/30-9/3/20 | 52 | 44 | 2 | 802 |
Trump’s handling of the coronavirus pandemic
After an initial approval rating in March of over 50 percent for Trump’s handling of the coronavirus pandemic, approval has fallen to 41 percent in September. That figure improved one point from August. Disapproval stands at 56 percent in September, two points less than a month earlier. The full trend is shown in Table 17.
Table 17: Approve or disapprove of Trump’s handling of the coronavirus pandemic, March–September 2020
| Poll Dates | Approve | Disapprove | Don’t know | n |
| 3/24-29/20 | 51 | 46 | 2 | 813 |
| 5/3-7/20 | 44 | 51 | 3 | 811 |
| 6/14-18/20 | 44 | 52 | 3 | 805 |
| 8/4-9/20 | 40 | 58 | 2 | 801 |
| 8/30-9/3/20 | 41 | 56 | 2 | 802 |
Approval of Evers’ job performance and handling of coronavirus
After six months of elevated approval ratings, approval of Gov. Tony Evers’ handling of his job fell in September, back to pre-coronavirus levels. In September, 51 percent approve and 43 percent disapprove of Evers’ job performance. In August, 57 percent approved and 37 percent disapproved. The trend in overall approval of Evers in 2020 is shown in Table 18.
Table 18: Evers’ job approval, January–September 2020
| Poll Dates | Approve | Disapprove |
| 1/8-12/20 | 51 | 40 |
| 2/19-23/20 | 51 | 38 |
| 3/24-29/20 | 65 | 29 |
| 5/3-7/20 | 59 | 33 |
| 6/14-18/20 | 54 | 38 |
| 8/4-9/20 | 57 | 37 |
| 8/30-9/3/20 | 51 | 43 |
Evaluation of Evers’ handling of the coronavirus issue is 57 percent approve and 38 percent disapprove. In August, 61 percent approved and 35 percent disapproved. The trend in approval is shown in Table 19.
Table 19: Approval of Evers’ handling of the coronavirus outbreak, March–September 2020
| Poll Dates | Approve | Disapprove |
| 3/24-29/20 | 76 | 17 |
| 5/3-7/20 | 64 | 32 |
| 6/14-18/20 | 58 | 37 |
| 8/4-9/20 | 61 | 35 |
| 8/30-9/3/20 | 57 | 38 |
Attitudes concerning COVID
Table 20 shows the level of worry about being personally affected by the coronavirus outbreak since March. After the percentage “very worried” rose in August, it declined in September, while those not at all worried did not change.
Table 20: How worried are you about being affected by coronavirus, March–September 2020
| Poll Dates | Very worried | Somewhat worried | Not very worried | Not worried at all | Already had COVID-19 (VOL) | Don’t know | n |
| 3/24-29/20 | 30 | 40 | 18 | 11 | 0 | 0 | 813 |
| 5/3-7/20 | 25 | 35 | 20 | 19 | 0 | 0 | 811 |
| 6/14-18/20 | 19 | 36 | 21 | 24 | 1 | 1 | 805 |
| 8/4-9/20 | 27 | 36 | 17 | 19 | 1 | 0 | 801 |
| 8/30-9/3/20 | 21 | 39 | 19 | 19 | 1 | 0 | 802 |
Feelings about reopening schools
There has been a decline since June in the percentage comfortable with reopening schools, including a small decline from early August to September. Those uncomfortable with reopening is above 50 percent for the first time, as shown in Table 21.
Table 21: Comfortable or uncomfortable with reopening schools, June–September 2020
| Poll Dates | Comfortable doing this | Uncomfortable doing this | Don’t know | n |
| 6/14-18/20 | 54 | 38 | 8 | 805 |
| 8/4-9/20 | 45 | 48 | 6 | 801 |
| 8/30-9/3/20 | 43 | 51 | 6 | 802 |
Those with school-age children have become more uncomfortable with reopening schools. In early August, among those with such children, 53 percent were comfortable and 45 percent uncomfortable. But, by September, 44 percent were comfortable and 54 percent were uncomfortable. Those without school-age children did not change their feelings about reopening schools. These results are shown in Table 22.
Table 22: Comfortable or uncomfortable about reopening schools, by school-age children in home, August–September 2020
| School-age children? | Poll Dates | Comfortable doing this | Uncomfortable doing this | Don’t know | n |
| No school-age children | 8/4-9/20 | 42 | 50 | 7 | 554 |
| No school-age children | 8/30-9/3/20 | 42 | 49 | 8 | 548 |
| Has school-age children | 8/4-9/20 | 53 | 45 | 2 | 242 |
| Has school-age children | 8/30-9/3/20 | 44 | 54 | 2 | 243 |
Acceptance of a COVID vaccine when available
While the world pharmaceutical companies have raced for a vaccine against COVID, some people say they are not likely to be vaccinated. Thirty-five percent in the September poll say they would definitely get vaccinated and another 29 percent would probably get the vaccine. However, 15 percent say they would probably not get vaccinated, and 18 percent would definitely not get the vaccine.
Table 23 shows likelihood of getting the vaccine by age, and Table 24 shows it by partisanship. Older people are more likely to get the vaccine than those 30 to 59, though those under 30 are also a bit more likely to say they will definitely get vaccinated.
Among partisans, Republicans are about equally divided between definitely or likely to get vaccinated and definitely or likely not to get the vaccine. Democrats and independents are much more receptive to the vaccine.
Table 23: Get COVID vaccine, by age, September 2020
| Age | Definitely get | Probably get | Probably not get | Definitely not get | Don’t know | Refused | n |
| 18-29 | 35 | 26 | 19 | 20 | 0 | 0 | 119 |
| 30-44 | 30 | 27 | 16 | 24 | 1 | 1 | 177 |
| 45-59 | 28 | 29 | 19 | 20 | 2 | 2 | 194 |
| 60+ | 44 | 31 | 9 | 11 | 4 | 0 | 308 |
Table 24: Get COVID vaccine, by party identification, September 2020
| Party ID | Definitely get | Probably get | Probably not get | Definitely not get | Don’t know | Refused | n |
| Republican | 31 | 23 | 19 | 24 | 2 | 0 | 238 |
| Lean Republican | 28 | 27 | 17 | 26 | 1 | 0 | 119 |
| Independent | 22 | 34 | 23 | 13 | 5 | 3 | 67 |
| Lean Democrat | 41 | 36 | 12 | 8 | 2 | 1 | 141 |
| Democrat | 45 | 31 | 9 | 13 | 3 | 0 | 224 |
Choice of ballot type in November
There has been a decline in the percentage of people planning to vote by absentee ballot by mail, together with a rise in those planning to vote in person on Election Day. Table 25 shows the trend since May.
Table 25: Ballot type, May–September 2020
| Poll Dates | Election day, in person | Early, in person | Absentee by mail | Probably/might not vote | Don’t know | n |
| 5/3-7/20 | 39 | 11 | 43 | 2 | 4 | 811 |
| 8/4-9/20 | 46 | 12 | 35 | 3 | 3 | 801 |
| 8/30-9/3/20 | 50 | 14 | 32 | 2 | 2 | 802 |
A majority of Republicans say they plan to vote in person on election day, as opposed to absentee by mail (or early, in person), while most Democrats are likely to prefer absentee by mail, as opposed to any other particular option, as shown in Table 26.
Table 26: Ballot type by party identification, September 2020
| Party ID | Election day, in person | Early, in person | Absentee by mail | Probably/might not vote | Don’t know | n |
| Republican | 69 | 11 | 18 | 1 | 1 | 357 |
| Independent | 41 | 18 | 29 | 8 | 3 | 67 |
| Democrat | 34 | 16 | 47 | 2 | 1 | 365 |
There have been changes in the preferred type of ballot since May, with a smaller percentage in each partisan category choosing absentee by mail, though large partisan differences persist, as shown in Table 27.
May was the high-water mark for people saying they would vote absentee by mail among all partisan categories. Democrats have become less likely to choose absentee by mail in each subsequent poll, and more likely to say they will either vote in person on election day or during early voting, although, as stated, absentee by mail is their single most popular choice. Intentions of Republicans and independents have been relatively stable in August and September.
Table 27: Ballot type by party identification, by poll date
| Party ID | Poll Dates | Election day, in person | Early, in person | Absentee by mail | Probably/might not vote | Don’t know | n |
| Republican | 5/3-7/20 | 59 | 13 | 25 | 1 | 2 | 353 |
| Republican | 8/4-9/20 | 67 | 12 | 15 | 3 | 3 | 359 |
| Republican | 8/30-9/3/20 | 69 | 11 | 18 | 1 | 1 | 357 |
| Independent | 5/3-7/20 | 30 | 15 | 43 | 5 | 6 | 74 |
| Independent | 8/4-9/20 | 39 | 15 | 27 | 13 | 5 | 71 |
| Independent | 8/30-9/3/20 | 41 | 18 | 29 | 8 | 3 | 67 |
| Democrat | 5/3-7/20 | 21 | 9 | 62 | 3 | 5 | 367 |
| Democrat | 8/4-9/20 | 27 | 12 | 55 | 2 | 3 | 363 |
| Democrat | 8/30-9/3/20 | 34 | 16 | 47 | 2 | 1 | 365 |
The shift in partisan preferences for each ballot type has also reduced, but far from eliminated, expected differences in vote by ballot type, as shown in Table 28 among likely voters. The election day in-person vote still favors Trump, but by less than in May or August. Meanwhile, the early in-person ballots have shifted from favoring Trump in May to favoring Biden in September. The absentee-by-mail ballots are heavily for Biden, but by less than in August, and slightly less than in May.
Table 28: Vote by ballot type by poll date, among likely voters, May–September 2020
| Ballot type | Poll Dates | Biden | Trump | Other/DK/Ref | n |
| Election day, in person | 5/3-7/20 | 26 | 68 | 6 | 256 |
| Election day, in person | 8/4-9/20 | 26 | 67 | 7 | 332 |
| Election day, in person | 8/30-9/3/20 | 33 | 58 | 9 | 356 |
| Early, in person | 5/3-7/20 | 36 | 53 | 11 | 76 |
| Early, in person | 8/4-9/20 | 50 | 45 | 5 | 80 |
| Early, in person | 8/30-9/3/20 | 53 | 35 | 12 | 97 |
| Absentee by mail | 5/3-7/20 | 72 | 23 | 5 | 299 |
| Absentee by mail | 8/4-9/20 | 81 | 14 | 5 | 241 |
| Absentee by mail | 8/30-9/3/20 | 68 | 22 | 10 | 222 |
Potential sources of change in the vote
Three sources of potential change in the election outlook are (1) undecided voters who might disproportionately favor a candidate later; (2) the possibility that those less likely to vote may in fact decide to vote; and (3) new voters coming into the electorate who might vote differently than those who have been registered before.
We look at these possibilities below.
- How do the undecided lean?
Among all registered voters interviewed since June, 11 percent have said they would vote for someone other than Trump or Biden, are undecided, or declined to give a choice. Pooling all three polls to include enough cases for analysis, we find that when asked, “Would you say you lean toward Biden or toward Trump?” 26 percent chose Biden and 19 percent chose Trump. In September, when Jorgensen, the Libertarian candidate, was included, 4 percent chose her.
Table 29 shows the vote among likely voters by poll wave since June, including those who are undecided, but lean to a candidate. The effect of including the “leaned” vote is small, leaving the September margin between Biden and Trump unchanged.
Table 29: Vote among likely voters including “leaned” vote choice, June–September 2020
| Poll Dates | Joe Biden | Donald Trump | Jo Jorgensen | None/other (VOL) | Refused | n |
| 6/14-18/20 | 52 | 44 | NA | 2 | 1 | 686 |
| 8/4-9/20 | 50 | 46 | NA | 2 | 1 | 675 |
| 8/30-9/3/20 | 48 | 44 | 4 | 0 | 2 | 688 |
- What about less likely voters?
There is always the potential for people to vote who have not in the past, either because of greater motivation this election or by new registrations to vote. How do these “less likely” voters compare to their “likely” or previously registered counterparts?
Table 30 compares likely voters (those who say they are certain to vote) with those who are not as likely to vote (those who say they are less than certain to vote). Polls from May through September are combined to provide sufficient sample size for analysis.
A much higher percentage of those less likely to vote (than of likely voters) say they are undecided or prefer someone other than Biden or Trump. Among the less likely to vote, Biden has a 9-point advantage over Trump, compared to a 5-point advantage among likely voters in this pooled May–September sample.
Table 30: Vote comparing likely with less-likely voters, May–September 2020
| Likely/Unlikely voter | Biden | Trump | Other/DK/Ref | n |
| Likely | 49 | 44 | 7 | 2699 |
| Less likely | 40 | 31 | 29 | 520 |
Table 31 shows the vote by the full range of certainty of voting as reported by the respondent, again pooling all polls from May through September. The percentage undecided rises as the chance of voting goes down.
Biden maintains a lead over Trump with those who say they are not certain to vote but are very likely to do so. However, the vote is tied simply among those who say they have only a 50-50 chance of voting. Biden’s advantage returns among those who say they will not vote.
Table 31: Vote by certainty of voting, May–September 2020
| Certainty of voting | Biden | Trump | Other/DK/Ref | n |
| Absolutely certain | 49 | 44 | 7 | 2699 |
| Very likely | 45 | 35 | 20 | 302 |
| 50-50 | 30 | 30 | 40 | 147 |
| Will not vote | 38 | 14 | 47 | 57 |
- New registrants
The Marquette Law School Poll samples people who say they are currently registered to vote and those who say they are not registered but plan to register by election day, terming them all as “registered voters.” This latter group provides a look at potential “new voters.” Once more, polls from May through September are pooled to provide adequate sample size.
Table 32 shows little difference in vote preferences between those who say they will register and those who are already registered, with Biden leading by 5 points among those who say they will register and by 6 points among those who are already registered.
Table 32: Vote comparing new registrants with previously registered voters, May–September 2020
| Registered or will register | Biden | Trump | Other/DK/Ref | n |
| Registered | 48 | 42 | 10 | 2989 |
| Will register | 44 | 39 | 17 | 230 |
Views of the economy, past and future
Views of the direction of the economy have turned sharply down since February with many more people saying the economy has gotten worse over the past year. There was an upturn in September.
Respondents have a strongly positive outlook for the economy over the next 12 months, and a more positive outlook in September than in August. Tables 33 and 34 show the recent trends in these measures.
Table 33: Change in economy over past 12 months, January–September 2020
| Poll Dates | Gotten better | Gotten worse | Stayed the same | Don’t know | Net |
| 1/8-12/20 | 48 | 17 | 33 | 2 | 31 |
| 2/19-23/20 | 47 | 15 | 36 | 2 | 32 |
| 3/24-29/20 | 41 | 31 | 25 | 3 | 10 |
| 5/3-7/20 | 28 | 46 | 20 | 4 | -18 |
| 6/14-18/20 | 27 | 50 | 19 | 4 | -23 |
| 8/4-9/20 | 22 | 56 | 19 | 3 | -34 |
| 8/30-9/3/20 | 25 | 51 | 19 | 3 | -26 |
Table 34: Outlook for the economy over the next 12 months, January–September 2020
| Poll Dates | Get better | Get worse | Stay the same | Don’t know | Net |
| 1/8-12/20 | 33 | 23 | 37 | 6 | 10 |
| 2/19-23/20 | 36 | 21 | 37 | 7 | 15 |
| 3/24-29/20 | 44 | 34 | 13 | 8 | 10 |
| 5/3-7/20 | 45 | 31 | 16 | 7 | 14 |
| 6/14-18/20 | 50 | 19 | 24 | 6 | 31 |
| 8/4-9/20 | 45 | 23 | 21 | 11 | 21 |
| 8/30-9/3/20 | 48 | 18 | 21 | 13 | 30 |
Views of state officials
Tables 35-37 present the recent favorability ratings of elected officials in Wisconsin and the percentage of respondents who haven’t heard enough or say they don’t know.
Table 35: Gov. Tony Evers recent favorability trend, January–September 2020
| Poll Dates | Favorable | Unfavorable | Haven’t heard enough | Don’t know |
| 1/8-12/20 | 45 | 37 | 14 | 3 |
| 2/19-23/20 | 43 | 40 | 12 | 4 |
| 3/24-29/20 | 54 | 28 | 11 | 6 |
| 5/3-7/20 | 50 | 36 | 7 | 5 |
| 6/14-18/20 | 54 | 37 | 7 | 2 |
| 8/4-9/20 | 52 | 35 | 9 | 4 |
| 8/30-9/3/20 | 47 | 41 | 9 | 2 |
Table 36: Sen. Tammy Baldwin recent favorability trend, January–September 2020
| Poll Dates | Favorable | Unfavorable | Haven’t heard enough | Don’t know |
| 1/8-12/20 | 44 | 40 | 13 | 2 |
| 2/19-23/20 | 43 | 40 | 13 | 3 |
| 3/24-29/20 | 40 | 39 | 16 | 4 |
| 5/3-7/20 | 45 | 37 | 14 | 3 |
| 6/14-18/20 | 40 | 38 | 19 | 3 |
| 8/4-9/20 | 43 | 36 | 17 | 3 |
| 8/30-9/3/20 | 42 | 35 | 19 | 3 |
Table 37: Sen. Ron Johnson recent favorability trend, January-September 2020
| Poll Dates | Favorable | Unfavorable | Haven’t heard enough | Don’t know |
| 1/8-12/20 | 39 | 29 | 28 | 3 |
| 2/19-23/20 | 37 | 34 | 24 | 5 |
| 3/24-29/20 | 35 | 32 | 29 | 4 |
| 5/3-7/20 | 38 | 34 | 23 | 5 |
| 6/14-18/20 | 35 | 32 | 29 | 3 |
| 8/4-9/20 | 33 | 35 | 27 | 4 |
| 8/30-9/3/20 | 32 | 36 | 28 | 5 |
About the Marquette Law School Poll
The Marquette Law School Poll is the most extensive statewide polling project in Wisconsin history. This poll interviewed 802 registered Wisconsin voters by landline or cell phone from Aug. 30 to Sept. 3, 2020. The margin of error is +/- 4 percentage points for the full sample. There are 688 likely voters, with a margin of error of +/- 4.3 percentage points.
The partisan makeup of the sample, including those who lean to a party, is 44 percent Republican, 45 percent Democratic and 8 percent independent. The partisan makeup of the sample, excluding those who lean to a party, is 30 percent Republican, 28 percent Democratic and 41 percent independent.
Since January 2017, the long-term partisan balance, including those who lean to a party, in the Marquette poll has been 45 percent Republican and 45 percent Democratic, with 9 percent independent. Partisanship exuding those who lean has been 30 percent Republican and 29 percent Democratic, with 40 percent independent.
The entire questionnaire, methodology statement, full results and breakdowns by demographic groups are available at law.marquette.edu/poll/results-and-data.
