New Poll Shows Trump and Isakson Leading in Georgia
Landmark Communications just released new Georgia polling that shows Republican Donald Trump leading Democrat Hillary Clinton by 5.5 percentage points, and Republican Johnny Isakson leading Democrat Jim Barksdale by 13.6 points. The poll was taken Tuesday and Wednesday using 1,400 likely Georgia voters. The margin of error is 2.7%
IN the presidential race, Trump had 47.8%, while Clinton had 42.3%. Libertarian Gary Johnson had 4.3%, and 5.6% were undecided. Trump was favored by each age group, with the strongest support from those 65 and over. Trump pulls the support of 12.3% of African Americans, while Hillary Clinton is favored by 9.7% of Republicans. Independents are almost evenly split between Trump and Clinton, at 34% for Trump and 35% for Clinton.
The Senate race shows Isakson at just over 50%, at 50.4%. Barksdale has 36.8% support, while Libertarian Allen Buckley has 5.0. 7.8% remain undecided. That Isakson is over 50% is important. If he can stay above 50%, that will avoid a January runoff. Isakson has over 50% support in every age group except for 18-39 year olds where he has 45.6% support. Isakson has the support of 16% of African Americans, and is in the lead among independents, 41.9% to 27.1%.
“The biggest difference between Trump and Isakson’s numbers is among female voters,” said Jordan Fuchs, Landmark Vice President. “Senator Isakson leads among women by one percentage point (44-43%), while Trump is losing among women voters by nine percentage points (40-49%). It appears that the Barksdale campaign has grinded to a halt, while Senator Isakson’s campaign has solidified his vote. If this trend continues, Senator Isakson will win reelection by a double digit margin.”
Behold, the crosstabs.
[pdfviewer]https://www.georgiapol.com/wp-content/uploads/2016/10/Landmark-Poll-Georgia-Statewide-Oct-11-12-2016.pdf[/pdfviewer]
About Author
Jon Richards
Add a Comment
You must be logged in to post a comment.
The electorate will be over 50% Republican? OK, sure Landmark.
What do you think the % should be? I’m guessing Landmark might be a bit low on the GOP side. Georgia is not in play, but I hope dems keep wasting time and money chasing that unicorn.
50 percent gop seems high to me too but I’m not a pro.
Regardless, Landmark’s numbers are in line with previous polls in the past 3-4 weeks, according to rcp.
http://www.realclearpolitics.com/epolls/2016/president/ga/georgia_trump_vs_clinton_vs_johnson-5968.html
No pro here either, but basing it on 2012 where Romney got 53% of 3.9m ballots cast. My guess this time is a decline in dem voters with an equal or larger number of GOP votes thus the percentage might be a tick higher.
David, interestingly enough, I also thought that seemed high at first. However, I went back to our previous polls from this fall, and three weeks ago on Sept 22 we released a poll that had 46% of voters describing themselves as Republicans, and 37% of voters describing themselves as Democrats.
So today, the number for GOP rose from 46 to 50%, an upward change of 4%. The Democratic number stayed nearly identical, today at 37% (statistically unchanged).
What’s interesting is looking at the self-described Independents. In our Sept. 22 poll, Trump was winning Independents by 6 points, 39%-33%…but today Trump is losing independents by 1 point.
Why would that be? Because most Independents aren’t really Independents…many are actually just ticked off Republicans who won’t use the term “Republican.”
Independent voters in Georgia are for the most part really just disaffected Republicans. All polling firms show them to almost always vote for Republican candidates.
The number of people giving “Independent” responses as their answer went *down* four (4) percentage points of the electorate — and self-described Republicans gained 4% of the electorate. Trump’s support among so-called “Independents” declined 6 points from Sept 22 to October 12th – and it’s really because some Independents are starting to call themselves “Republican” as the election nears.
i think it’s spot on that many or most self-identified Independents are really just ppl who lean Republican, or for a few, Libertarian. what i am not accepting easily is the 12.3% African-American vote for Trump. That seems like an outlier relative to every other poll and state i’ve come across.
Thx bethebalance. Believe me, we are very cognizant of the issue you raise, and I addressed it to edatlanta in a reply below (so won’t repost here). Please check that.
That said, I think there is a group of African American voters who do want to do something politically different, and it shows up in polling in larger numbers in the months preceding elections.
However, it historically closes down to single digits as election day approaches. That is the value of intense GOTV efforts and persuasive advertising. Most polling firms just poll a few African American voters and “weight” what they find to nearly a third of the sample, which completely throws off numbers in a bad direction. We don’t do that – we actually make a concerted effort to get the correct number of AA voters to participate in our polls because we genuinely want to know their opinions, not just use weighting as an easy (and cheaper) way to report data. Thanks.
i appreciate your engagement here on this blog, and your rigorous approach to data collection and analysis. knowing the trends that you do, as they approach the election, i’m sure you’re tempted to make a predictive model of the actual election numbers…?
Also, compare to the last Presidential election in 2012. I went back to review it because it’s a fair question. In October 2012 (October 25th, 2012, to be exact), our poll of Georgia showed that 49% of Georgians called themselves Republican. So today’s number (also in October) of 50% really isn’t that different than 2012: only a percentage point difference, and certainly within margin of error.
In that October 25th poll, 49% called themselves Republican, and we had Romney at 53% of the vote. His final number on Dlection Day? 53%.
You’ve got some good results in the past, but it still seems odd to me to have this high a Self ID Republican unless you’re somehow really, really pushing people to ID a party. Are you weighting your sample to any particular vision of the electorate (age/race/gender?) Georgia didn’t get exit polled in 2012, but if you look at similar states in 2012, (with the “Indies are Shy Rs” theory), you don’t get IDs like that.
Alabama:
Romney 61-38
Party ID: R 43 D 34 I 24
Shy R Factor (Indies go R 75 D 23)
Missouri:
Romney 54-44
Party ID: R 35 D 37 I 24
Shy R Factor (Indies go R 59 D 35)
Or, for that matter, look at Georgia in 2008, when we did get exit polls:
McCain 52-47
Party ID: D 38 R 35 I 28
Shy R Factor (Indies go R 57 D 40)
Even if it’s not as close as in ’08 in Georgia, it seems weird to have party ID shoot up that much this year. And it really defies belief that somehow, the last three weeks have made people MORE likely to want to ID themselves as Republicans as opposed to disaffected indies.
Thanks David. We asked the Party ID question with the exact same phrasing we’ve been using for many years. So no, we didn’t push them in any direction or weight by it, nor did we weight by some other factor that would be different this time. It was the same weighting as our other head-to-head polls this year.
Hey Mark, I always appreciate the fact that you come here and answer questions after you guys release a poll. Do you have a break down of Metro v. outside of metro-Atlanta? It would be interesting to see that difference.
Thanks T.E. We didn’t do a Metro vs outside of Metro. But I’ll go ahead and run that and post it here. Give me a few hours to get some work off my plate and I’ll post it here by 5.
The Eiger, my data mgr had to leave early today so we didn’t get the “inside vs outside metro atlanta” numbers as of yet, but I’ll still post them here when he returns to the office-
-Mark
Thanks!
I think I inadvertently posted this response on another post, though you did see it there. However, I am again posting the information here, as well, for posterity sake. See my copy/paste of the information below:
—————————————–
“I can’t post a pdf or a jpg here, so I have to type the answers shorthand. Apologize for informality. Landmark poll of Georgia poll from 10.11.16, Metro Atlanta vs non-Metro Atlanta.
…………………Metro Atlanta
………………….YES………NO
Trump………36.2……….56.9
Clinton……..52.9……….33.9
Johnson…….6.1…………3.2
Undecided…4.8…………6.0”
12.3% of the African-American vote going to Trump? lawlawlawlawl.
There are always outliers, and liars (for example I take a guilty pleasure in messing with the robots), but keep in mind that David Duke was averaging 14.5% of the La. AA vote:
https://www.washingtonpost.com/news/the-fix/wp/2016/08/05/david-dukes-getting-more-support-from-black-voters-in-his-race-than-donald-trump-is-in-his/
Regardless, I do trust Landmark than others to report the facts as they get them.
Thanks Will Durant. In celebration of your Mr Peabody icon, and the heretofore poor reception of the existing third-party candidates, I offer this:
https://s-media-cache-ak0.pinimg.com/236x/f7/14/32/f7143251d50ecc11dc76c49e0042a9f1.jpg
Yeah, this year I’m leaning more towards Rocky as he was the cynic. As noted by Charlie earlier in the week this is the year, if any in the modern era that a third-party coulda’ been a contenduh but the Libertarians in particular blew it.
Thanks, edatlanta. Always glad to brighten your day. 😉 It’s not uncommon for some African American voters to consider voting Republican in the months and weeks leading up to elections. Most pollsters find this. It is also caused in some cases by pollsters not conducting enough actual interviews among black voters — instead relying on weighting, which really doesn’t work well in Georgia since Georgia has such a large percentage of minority voters. We don’t do that — we actually call and get the number of interviews with black voters that we need (didn’t just “weight up”).
But that said, I agree with your sentiment that ultimately more African American voters will vote Democratic. But in the interim, remember that we have to report what people tell us, not just insert our opinion of how we think they will vote.
Actually, other polling firms are finding about the same: Realclearpolitics.com (link below).
• JMC Analytics has 10% of black voters for Trump
• Quinnipiac Univ has nonwhite voters at 14% for Trump (they didn’t break out black voters per se, just “minorities”, but GA has a low number of *non*-black minorities as active voters, so it’s likely close to that.
• Opinion Savvy had 22% of black voters for Trump on 9/15/16, their last report that I see on RCP for Georgia
Ahhhh, James Edgar Comey, the latest Clinton stooge. Truly pathetic!
http://spectator.org/former-u-s-attorney-agents-see-fbi-chief-comey-as-a-dirty-cop/