The answer for the wide variance in outcomes may be due to the different methodologies used by the polling firms.
An understanding of those different methodologies is an important reference point for readers interested in understanding where the race in this key battleground state stands with less than four weeks remaining until Election Day, November 3.
Maintaining those 29 electoral college votes in 2020 is a key building block of President Trump’s reelection strategy in 2020. In 2016, he won the presidency with 304 electoral college votes to Hillary Clinton’s 227 electoral college votes, with seven faithless electors (two defecting from Trump and five defecting from Clinton). If President Trump were to win all the states he won in 2016, but lose Florida, his electoral college total, barring faithless electors, would drop from 306 to 277, just seven more than the 270 majority needed to win the presidency.
USA Today/Suffolk University poll of 500 likely voters was conducted via live telephone interviews between October 1 and October 4 and has a 4.4 percent margin of error. The partisan breakdown of poll respondents was 37 percent Democrat, 36 percent Republican, and 23 percent independent.
The Change Research/CNBC poll of Florida was a subset of a battleground state poll of 2,688 likely voters conducted between October 2 and October 4 in six states: Arizona, Florida, Michigan, North Carolina, Pennsylvania, and Wisconsin. In this larger poll of six states, Joe Biden leads. The margin of error in the poll that includes respondents from all six of these battleground states is 1.8 percent.
The Change Research/CNBC poll does not breakdown either the sample size or margin of error for the Florida subset of its battleground state poll, nor does it provide a breakdown of poll respondents by party affiliation.
Here’s how Change Research
describes their polling methodology:
We recruit new participants online for each and every poll, so Change Research polls are not affected by the dwindling response rates of phone polling. We use proprietary and patent-pending approaches to recruit fresh participants, and we don’t rely on online panels of habitual survey takers.
We collect survey responses by publishing targeted online solicitations via advertisements on websites and social media platforms. By finding a representative set of respondents via web and social media to take a poll, we are able to cast a net that is wider than phone polls. We reach millennials and seniors, rural and urban dwellers, and members of every gender, race, creed, and political persuasion
The University of North Florida poll of 3,142 likely voters
was, “administered through email via Qualtrics, an online survey platform,” between October 1 and October 4 and has a 1.8 percent margin of error. The poll’s cross tabs do not provide a partisan breakdown of the sample survey, but 42 percent of respondents who recall voting in 2016 say they voted for Hillary Clinton, while 42 percent say they voted for Donald Trump.
USA Today/Suffolk University poll and the University of North Florida poll provided extensive cross tab data on their poll results.
Key results for the presidential ballot test by demographic group from the
USA Today/Suffolk University poll include the following:
Men support Trump over Biden by a 47 percent to 42 percent margin, while women support Biden over Trump by a 49 percent to 44 percent margin.
White voters support Trump over Biden by a 57 percent to 41 percent, Hispanic voters support Biden over Trump by a 46 percent to 42 percent margin, and black voters support Biden over Trump by an 81 percent to three percent margin.
Trump leads Biden in only one age group, those between 35 and 49, among which 50 percent support Trump and 42 percent support Biden.
Biden leads Trump in all other age groups: 42 percent to 37 percent among those between age 18 and age 34, 48 percent to 47 percent among those between age 50 and age 64, and 49 percent to 46 percent among those over age 65.
Notable in these results is the relatively small margin of difference in presidential preference by gender as well as age group.
Perhaps most surprisingly, Biden has a narrow four-point lead over Trump among Hispanics, but a huge 81 percent to three percent lead over Trump among black voters, dramatically lower results than Trump’s eight percent among black voters in 2016 and a number of recent polls that show Trump winning double-digit support among black voters, ranging all the way up to 20 percent.
Key results for the presidential ballot test by demographic group from the University of North Florida poll include the following:
Men support Trump over Biden by a 51 percent to 45 percent margin, while women support Biden over Trump by a wide margin, 56 percent to 39 percent.
White voters support Trump over Biden by a 54 percent to 43 percent, Hispanic voters support each candidate equally with 47 percent support, and black voters support Biden over Trump by a 93 percent to seven percent margin.
Trump leads Biden in the two oldest age groups: 50 percent to 47 percent among those aged 65 years and older and 56 percent to 39 percent among voters 55 to 64.
Biden leads Trump in the four youngest age groups: 69 percent to 23 percent among those age 18 to 24, 53 percent to 42 percent among those age 25 to 34, 61 percent to 36 percent among those age 35 to 44, and 49 percent to 48 percent among those age 45 to 54.
USA Today/Suffolk University poll, the University of North Florida poll shows a much wider gender gap, with Biden winning women by 17 percent and Trump winning men by six percent.
The University of North Florida poll also shows a much wider gap by age group than the
USA Today/Suffolk University poll.
The two polls have similar results in showing very low support for Trump among black voters (three percent in the
USA Today/Suffolk University poll and seven percent in the University of North Florida poll), but very high support for Trump among Hispanic voters — tied with Biden in the University of North Florida poll and trailing him by just four percent in the USA Today/Suffolk University poll.