The poll began by asking, “Of all the candidates currently running for president, can you please name the one person you would most like to win the election?” The question offered no specific candidate names. The poll’s sample of registered voters responded:
- Carson (R): 19 percent
- Hillary Clinton (D): 16 percent
- Donald Trump (R): 14 percent
- Bernie Sanders (D): 5 percent
“Carson is leading his closest Republican rival, businessman Donald Trump, by a significant margin, and both Carson and Trump have better favorability ratings than Democratic frontrunner Hillary Clinton,” said Jason Reineke, associate director of the statewide poll at Middle Tennessee State University.
“Notably, though, about a third of state voters can’t or won’t say whom they would prefer. That’s to be expected, given that even primary voting remains a long way off.”
Carson and Trump both surpassed Clinton in follow-up questions separately gauging support for each candidate becoming president, with 51 percent supporting Carson, 35 percent for Trump, and 25 percent for Clinton.
The poll randomly surveyed 603 registered voters statewide Oct. 25-27 and has a margin of error of 4 percentage points.
Breakdowns by party
A retired neurosurgeon who has risen to GOP frontrunner in recent weeks, Carson was the top choice among self-identified Republicans, with 33 percent of responses, followed by “Don’t know” with 29 percent and Trump with 22 percent.
Clinton led among Democrats, with 44 percent naming her, followed by “Don’t know” with 25 percent and Sanders with 16 percent.
Among independents who did not identify with either of the two major political parties, the most frequent response was “Don’t know” (38 percent.) Carson was the choice of 16 percent of independents, while Clinton and Trump were named by 12 percent each.
Support for Carson surged, though, among independents who identified as ideologically conservative, with 25 percent of these respondents picking him. Trump garnered 15 percent of this group, while 33 percent were undecided. Margins of error are larger for subgroups of respondents like political partisans.
More oppose than favor most top candidates
Carson was the only candidate who remained in positive territory in “net favorability” — +28 — when the percentages opposing or strongly opposing his presidency were subtracted from the percentages favoring or strongly favoring his presidency.
On the Democratic side, Clinton had the lowest net favorability score of all candidates at -37, while Sanders was comparable at -33.
But favorability deficits were not limited to Democratic candidates. Among the Republicans, former Florida Gov. Jeb Bush scored -26; Trump came in at -12, U.S. Sen. Ted Cruz scored a comparable -11, while fellow U.S. Sen. Marco Rubio scored -7 points.
Some respondents did not give an opinion one way or the other, or refused to answer the question.
“Some of these are frontrunner effects,” said Ken Blake, director of the poll. “Since Carson is in the lead, people who support him are going to be reluctant to say they could accept anybody else. There are also, of course, partisan effects. Republicans have an advantage, and it would be surprising to see net favorability for any Democratic candidate in a red state like Tennessee.”
Reineke added that it’s also important to note that voters in Tennessee are more decided about some candidates than others.
“People are more opinionated about candidates like Clinton and Trump — 89 and 82 percent have an opinion about them one way or the other, respectively,” Reineke noted.
“But only 61 percent each stated an opinion one way or the other about Cruz and Rubio. So, Cruz and Rubio have an arguably easier task of creating a favorable impression among undecided voters, while Clinton and Trump would have to convert voters from opposition to favorability to make up the difference.”
Interviews for the poll were conducted by Issues & Answers Network Inc., which completed 603 telephone surveys among a random sample of registered Tennessee voters aged 18 and over.
Data was collected using Tennessee statewide voter registration sample with a mix of 60 percent landline and 40 percent cell phones. The average interview length was nine minutes.
Quotas by gender and geographic region were implemented to ensure the sampled respondents were representative of Tennessee’s adult population. U.S. Census Bureau data were used to determine the gender distribution each of Tennessee’s Grand Divisions: East, Middle, and West. Data was weighted on age to ensure that it was representative of Tennessee registered voters
The survey’s margin of error is +/- 4 for the entire sample percentage points, meaning that we are 95 percent confident that the actual result lies within 4 percentage points (in either direction) of the result our sample produced. Subgroups have wider margins of error.
Thanks to MTSU for the release. There will be more coming out next week about other topics tied to the presidential races.