When “Fake News” Becomes Propaganda
If you believe that Jon Ossoff is ahead of Karen Handel by 7 points, you probably live in a reality where Jason Carter is a US Senator from Georgia and Michele Nunn is Governor. Wherever that reality is, I hope it’s pleasant, in addition to being fact-free.
Those of us who are not window-licking morons know different.
The “poll” from Abt Associates and published by the AJC, was conducted poorly and without any detectable professional rigor. As a subscriber to the AJC, I am offended they even published it, but I have only sympathy for Greg Bluestein, Aaron Gould Sheinin and Kristina Torres, who did their best to report on the steaming pile of debacle. It’s not easy to polish poo, but they did a fairly good job.
Let’s start with the organization that conducted the poll, Abt Associates, which describes itself as “…a mission-driven, global leader in research, evaluation and program implementation in the fields of health, social and environmental policy, and international development.” Missing from the company’s description of itself is the word “pollster.” (The link goes to their website, so check for yourselves.) Using a social justice warrior research organization to conduct a political poll is a poor decision on the AJC’s part. Calling it a “poll” is journalistic malpractice.
When you dive into the crosstabs and take a hard look at the respondents, you’ll find that 60% of them were contacted by cell phone. That’s nearly double the rate that any competent pollster would accept cell phone only voters skew younger, and lean more Democrat than the average voter. I hire pollsters for my clients, and would refuse to pay for a poll based on 60% cellphone responses.
Getting 60% of your responses via cell phone leads to a cohort that self-identifies as 29% Democrat, 28% Republican, and 33% Independent, and voted for Hillary Clinton over Donald Trump 55% to 45%. There is simply no way this represents reality in the 6th. Even Democrats admit the district leans Republican by 5% or so, and Jon Ossoff has run his entire campaign to the middle -spending millions of liberal donors’ money pretending to be a budget hawk hoping to root out “wasteful government spending.” And Donald Trump, like it or not, WON the district by nearly 2 points –he didn’t lose it by 10.
According to Abt’s Magic 8-Ball memo, among the 74 respondents who have already voted, Ossoff leads Handel by 31 points. The early voting data, (which is publicly available even to Abt Associates) flatly contradicts this. Republicans and leaners are turning out at a higher percentage than they did on April 18 –and there were more Republican than Democrat ballots cast that day.
The reality is that this race is as close as a sock to an ankle, which is in no way to try to spin Abt’s pathetic work into good news for Karen Handel. It’s a coin-toss. Landmark Communications (the official pollster for WSB-TV) published a real poll last week that said just that. The surprise is that a nothingburger candidate is in a dead heat with a proven, experienced conservative with a demonstrated record of public service –in an R+8 district that elected Newt Gingrich, Johnny Isakson and Tom Price.
That’s newsworthy, but the survey published this weekend by the AJC is a pile of dog squeeze that doesn’t qualify as a poll, not even a “skewed poll.” They should release the raw date to show it’s not utter garbage –and reporting it takes “fake news” to the level of out-and-out propaganda.
The AJC should be ashamed of itself.
About Author
Mike Hassinger
Add a Comment
You must be logged in to post a comment.
Aaron’s so ashamed, he’s leaving. :>)
But wouldn’t Carter be Governor and Nunn be Senator in said alternate universe?
It’s all propaganda, Mike. Typical Democratic voter suppression efforts and outright lies. Who can forget President Lurch in the 2004 exit polls? Hmmmm?
I’m from the sixth, from a reliably republican household, and we stopped using a landline in the late 90s. I’d think that there’s a higher adoption of cell phones than the national mean in the 6th given how affluent it is.
What year will cell phone use stop being a proxy for your political affiliation?
I wouldn’t think the method of contact is anywhere near as important to the weighting of responses as party affiliation, age, race, or voting history, which should all be asked and known directly.
“The surprise is that a nothingburger candidate is in a deadheat with a proven, experienced conservative…”
But is it really a surprise?
As is well known, Trump only carried this district by a point (rounded to the nearest whole number) over Clinton…a mere one point. “Of all 241 Republican-held (Congressional) districts, Hillary Clinton improved on Barack Obama’s 2012 performance more in GA-6 than she did in any other (district) except for TX-7, held by Rep. John Culberson (R).”—University of Virginia’s Center for Politics website, February 2, 2017. So in large part, this race has become a referendum on Trump—and I doubt Trump’s numbers are any better in CD 6 than they were seven months ago, when he was elected president.
Secondly—and I have pointed this out before—the 6th CD is not really a “conservative” district, if we mean akin to say, Austin Scott’s 8th CD in middle and south Georgia or Doug Collins’s Gainesville-based 9th CD. The “conservative” 6th District voted 77% for Sunday retail alcohol sales in 2012—even higher than the 74% approval from John Lewis’s ultra-Democratic district just to the south of CD 6. In 1996, when he sought the GOP US Senate nomination—running pro-choice on abortion—Isakson got 60% in what is now the 6th CD (though he lost statewide in the runoff to Guy Millner). Do we have reason to think that 21 years later, the 6th CD is more conservative on that issue than today? Ossoff no doubt is aware that CD 6 is not a hotbed of social conservatism—why else would he be hitting Handel so hard on the issue of Planned Parenthood? One might argue the 6th CD is more a “moderate” district overall—fiscally conservative but more moderate to liberal on social issues.
Third, Handel’s campaign has left much to be desired…heck, even Dick Williams of the Georgia Gang yesterday—hardly a liberal voice—was lamenting the state of Handel’s campaign. About 99% of her ads are negative (is that a sign of desperation?)—I can’t think of a single positive ad I saw for her with just a couple hours in front of the tube yesterday. Heck, after seeing her ads, you’re almost scared to make a shopping run to Lenox—you know, the Jihadists in our shopping malls and terrorists in the streets of Buckhead. Ossoff at least has a mixture of positive and negative ads. I can’t even recall a single Handel ad where she directly addresses the voters (as opposed to some third-party voice).
Adding on all these good points: I wouldn’t hold fast to the November results as being a guarantee of a Trump +2 electorate. Special elections and midterms with lower turnout rates plus opposition enthusiasm can result in electorates that voted very differently than they did in a broad general election. 2010 is the comparison: You went from an electorate Obama had won by 8 points to one that he’d lost. It’s certainly possible that next Tuesday you have a pool of voters, that if they’d been the only one voting last November, Trump would have lost the district by any number of points.
Well. I guess Mike feels Handel needs the push. No surprise there, she seems to need all the help she can get, although my money is still on a close race with Handel probably the favorite.
I don’t understand his remark about 60% cell phone rate being “nearly double the rate that any competent pollster” would use. Either he has some caveats in mind or I’m reading things wrong or he believes some heavy polling outfits are incompetent.
From the first two links I checked: Pew increased their cellphone call rate from 65 to 75% last year and Wapo/ABC used 65% as of late 2015. What percent did Landmark use last week?
Do polls that show a lead above the margin of error actually help the leader? Doesn’t second place try harder or does the old marketing bandwagon effect count for more?
I get it that this race is considered nationally to be an indicator of our Trumpian future but I’m finding it more to be an indictment of the Citizens United decision. Possibly $50 million for 0.0023 of a vote in the House and that is not even for a full term at that.
Fake news and propaganda are the same thing. However, what you’re talking about is just poor poll sampling. And the issue with that has to do more with the sample not being representative of the district than with the cell phone usage. Land line respondents skew older and more conservative, and something like 95% of the population has a cell phone.
Let’s not use the derogatory term of the day to dismiss findings we don’t like just because it’s simple to do.
All this begs the question- how critical or investigative must a published piece of news be in order to be not “fake”? Are all puff pieces fake news? What if the piece just presents an existing albeit limited perspective? “Real” does not necessarily mean fair and balanced.
I guess SurveyUSA which also had Ossoff by 7 is bad too…just that they have an A+ rating from 538…and they do polls
One thing that would be helpful to see in all these polls is the regional breakdown, by county. All I hear are the overall totals or percentages. In the April special, 45% of the total votes came from Fulton, 32% from the Cobb portion of the district and 23% from DeKalb. Ossoff in Round 1 got 41% in the Cobb portion of the district, 48% in Fulton and 59% in DeKalb. If he is only polling in the 30s in the Cobb portion, he won’t win, and ditto if in Fulton, easily the largest county in the district, he only polls low to mid 40s. Obviously he has to win DeKalb ()and he will), but that alone is insufficient for victory.
Another question is whether there will be falloff from November? In the previous 2 special congressional elections, while the Republican candidate won, his margin of victory was far less than Trump’s had been, like a two-thirds drop in the margin between Trump and the respective candidate. If there is even a minimal dropoff in CD 6, say from Trump +1 to Ossoff +2 or 3, well…you know the outcome…
Hmmmm. Maybe the fact that a nothingburger candidate is in a dead heat with Handel in an R+8 district that elected Newt Gingrich, Johnny Isakson and Tom Price suggests, strongly, that Handel is not a proven, experienced conservative with a demonstrated record of public service.
Much wiser fingers have typed words that I agree with in this thread, so I won’t add anything other than…
Pass the Salt.
I hear salt helps cut the bitterness…so might want to pass it back
These comments on the recent AJC poll are inaccurate and misleading. First, contrary to the author’s claim, Abt Associates is a highly respected survey research firm that has conducted numerous public opinion surveys for government agencies as well as media outlets. Second, the methodology employed in the AJC poll is in line with the highest standards of contemporary public opinion polling. In fact, the AJC poll is the only recent poll in Georgia to adhere to what is considered by survey researchers to be the “gold standard” for conducting such polls: using live interviewers to call both landlines and cell phones. The 60% proportion of calls to cell phones is also right in line with current polling standards and reflects the reality that a majority of voters in Georgia and the United States are now in households without landline phones. Third, the results of the AJC poll are entirely consistent with the results of other recent polls in the 6th district race. According to the realclearpolitics website, recent polls in this race have shown an Ossoff lead ranging from 1 point to 7 points with an average lead of almost 5 points. We won’t know until election night whether the polls in this race were accurate, but it is simply wrong to single out the AJC poll for special criticism because the author does not like the poll’s results.
It’s not the results I don’t like, it’s the methodology. Quibble all you want, this race is a dead heat, as today’s results show. They sampled likely voters, not registered voters, and they don’t say if they appended it to a voter file, which is what anyone who runs campaigns for a living would do. That’s the difference between what I call “working” polls -those conducted for political candidates- and “academic” polls which are run by institutions.
And if Abt wants to be a pollster, they should say as much on their frickin’ website.
So if you were working for a candidate or an organization with an agenda, and you could craft a poll such that it’s results supported that candidate or agenda, you would do it too. Heck you probably have done it.
Like the old adage that any competant prosecutor can indict a ham sandwich, any competant pollster will skew their results to create the best case scenario for whoevet is paying them. It’s all literally smoke and mirrors. We’ll know next Tuesday!
frickin’ website…. the description by a supporter of the candidate that as ACP describes it, has a “starter website”.
Emotion can cloud judgement. The desperation is palpable.
Republicans love to say Nothingburger