Remembering 1980: Are the Polls Missing Something?

  • Share
  • Read Later
AP

U.S. President Jimmy Carter, left, and Republican presidential candidate Ronald Reagan face their panelists during their televised debate at the Cleveland Convention Center in Cleveland, Ohio, on Oct. 28, 1980.

In November 1980, the great TIME correspondent and editor John F. Stacks (a mentor to such stars as John Dickerson and many others) won the unenviable task of analyzing how and why every single public pollster (including ours) missed the Reagan landslide earlier that month. Wrote Stacks:

For weeks before the presidential election, the gurus of public opinion polling were nearly unanimous in their findings. In survey after survey, they agreed that the coming choice between President Jimmy Carter and Challenger Ronald Reagan was “too close to call.” A few points at most, they said, separated the two major contenders.

But when the votes were counted, the former California Governor had defeated Carter by a margin of 51% to 41% in the popular vote–a rout for a U.S. presidential race. In the electoral college, the Reagan victory was a 10-to-1 avalanche that left the President holding only six states and the District of Columbia.

After being so right for so long about presidential elections–the pollsters’ findings had closely agreed with the voting results for most of the past 30 years–how could the surveys have been so wrong? The question is far more than technical. The spreading use of polls by the press and television has an important, if unmeasurable, effect on how voters perceive the candidates and the campaign, creating a kind of synergistic effect: the more a candidate rises in the polls, the more voters seem to take him seriously.

With such responsibilities thrust on them, the pollsters have a lot to answer for, and they know it. Their problems with the Carter-Reagan race have touched off the most skeptical examination of public opinion polling since 1948, when the surveyers made Thomas Dewey a sure winner over Harry Truman. In response, the experts have been explaining, qualifying, clarifying–and rationalizing. Simultaneously, they are privately embroiled in as much backbiting, mudslinging and mutual criticism as the tight-knit little profession has ever known. The public and private pollsters are criticizing their competition’s judgment, methodology, reliability and even honesty.

At the heart of the controversy is the fact that no published survey detected the Reagan landslide before it actually happened. Three weeks before the election, for example, TIME’S polling firm, Yankelovich, Skelly and White, produced a survey of 1,632 registered voters showing the race almost dead even, as did a private survey by Caddell. Two weeks later, a survey by CBS News and the New York Times showed about the same situation.

Now, there has hardly been a presidential campaign adviser since 1980 who, finding his candidate either down or uncomfortably close to the competitor, has not cited the Carter-Reagan pre-election polls as evidence they are in much better shape than they appear to be. For the most part, the polls have returned to their pre-1980 accuracy.

But the current debate between the Obama and Romney camps over this year’s voter turnout–whether more or less people will come out to vote than in 2008–makes Stacks’ thorough after-action report particularly relevant (subscribe to read the whole thing).

On the one hand, Romney backers point to Gallup’s projections for a shrinking electorate to claim the polls have way overstated Obama’s advantage. The tightest summary of that position, with some links, is here. Gallup is very blunt: “U.S. Voter Turnout Will Likely Fall Short of 2004, 2008.”

On the other hand, Obama aides David Plouffe and Jim Messina have been saying for months that their single-minded goal is an expansion of the pro-Obama electorate and that their very well-organized ground game will deliver it on Tuesday. Here‘s Messina on that, including the blunt assertion that “Voter turnout will reach an all-time high in this election.”

The Gallup and Messina scenarios can be reconciled (sort of) if turnout is down in red states but up in battleground states. But there’s also the possibility that one side is very right and the other is very wrong and that we’ll be looking for answers come Nov. 7th.

Stacks, unfortunately, isn’t around to provide them–he died Sept. 4–but there are plenty of his acolytes on the job.

46 comments
RandyBrown1
RandyBrown1

Polling conducted by the mostly liberal polling outfits often if not most often turn out results that are highly tainted by wishful thinking.  This has been obvious for decades.  

JubalBiggs
JubalBiggs

You could also mention 2004 with the dramatically under-reported Bush performance in Ohio. Or how about 2010, when most pollsters thought a relatively mild wave in favor of the GOP would maybe give them a couple seats, but the overwhelming narrative was "an evenly divided nation". 2010 wasn't anything like 'evenly divided', Congressional Democrats got slaughtered. Why is it that these mistakes only occur when reporting good years for Republicans? Would it have anything to do with this?

"The spreading use of polls by the press and television has an important, if unmeasurable, effect on how voters perceive the candidates and the campaign, creating a kind of synergistic effect: the more a candidate rises in the polls, the more voters seem to take him seriously."

Why are current swing state polls sampling based on an expected Democratic turnout in excess of their 2008 advantage? Do they just believe everything that comes from he Obama campaign as God's Own Truth, or are they intentionally trying to keep their guy in the fight against the fundamentals? Why do we somehow poll latinos far more than evangelicals? Why, why why... Why do we believe polls anyway?

gordo29
gordo29

We haven't had such a clear choice in direction in such a closerace since 1980, so the article makes sense.  Having lived throughthat, I can tell you the level of angst this year dwarfs 1980.  We werebeaten down by the terrible economy, disasters overseas and simple doom andgloom, but I don't remember such fear and loathing between the sides.Carter, though a very poor president, did not manage to alienate everyone hedisagreed with, nor did he try...but he made very poor decisions and alientatedmany in his own party with his anti-Washington stance.  But he was a niceman who wore a sweater, etc.  In 1980 we had a challenger who was agovernor and former actor vs a failed president who was likable.  This timewe have governor and businessman vs a failed president who has beenunbelievably partisan, but who is likable for half the electorate.There is one important result that this country needs beyond anything -whoever wins MUST become a leader for all the people.  Romney has alreadydone that many times over.  Obama never has, but he will need to ASAP if hewins.Should be a crazy night!

mrnathan.ges
mrnathan.ges

One problem facing the country is this: If low-interest voters don't turn out and Romney wins, a lot of Democrats are going to blame it on alleged voter suppression. If turnout is high and Obama wins, a lot of Republicans are going to blame it on alleged voter fraud. Neither side remotely trusts the other. Partisan Democrats sincerely believe Republicans will fire-hose crowds of black people whenever they can get away with it, and partisan Republicans sincerely believe Democrats will cast false votes to inflate their numbers in big cities.I don't have a dog in this fight, because I think the evidence is inconclusive. None, one or both of the allegations could be right. But in any case, we're going to have to heal this rift, and restore faith in democratic integrity, before we can properly assess whose turnout models actually reflect the desires of the citizens.

MartR2012
MartR2012

Are the Polls missing something? Yes they are. They are not asking about all the candidates on the ballot. I was asked who would I likely vote for, Romney or Obama. I said neither. They ended the call right then and there. 

Yes. There IS something missing in the polling. If they are not thorough then the results are not reliable.

JoeBrandimore
JoeBrandimore

I have been thinking myself how much this election reminds me of 1980. As I recall that year (my first election), Carter didn't really have much support. There was a anti-Reagan vote for sure, as the media had charactertured Reagan as some sort of "bomb Russia" war monger.

As the debates especially unfolded, people found the media's "message" to be way off base - that while Reagan for sure had strong anti-communist principles he was also far from crazy. He became an acceptable - even preferable alternative to Carter.

Reagan also had strong supporters on his side and not just "anti-Carter" votes.

Contrast 1980 with this year. Once again, we had a media "message" about Romney as a latter day Marie Antioinette type of plutocrat blown up in the debates. Romney either won or lost, but clearly he established himself as a viable alternative to Obama.

Taken along with the fact there seems to be a palpable enthusiasm gap towards Romney and I have a weird feeling we are going to have a blowout next Tuesday.

I could be wrong, but even among the Obama voters I know I don't sense much desire for a 2nd term for Obama.

Leftcoastrocky
Leftcoastrocky

Reagan held a lead from mid-September onward and had a two or three point lead heading into the debates. Private polling conducted for the Reagan and Carter campaigns showed the same thing. Thus, Reagan’s 10 point victory is NOT a model for a come-from-behind victory

TyPollard
TyPollard

Question: What were the personal favorability ratings of Reagan compared to Romney?

I'm guessing they are different.

charlieromeobravo
charlieromeobravo

"Are the Polls Missing Something?"

Is the media bored with 2 days of relatively no campaign news so they feel the need to write blog posts like this? 

nhautamaki
nhautamaki

Lost in all that analysis is the fact that the media has a vested interest in pollsters concluding that races are 'too close to call--but stay tuned in for more exciting developments on this neck-and-neck race to the finish line!'

The media is supposed to be politically unbiased and to be honest I do think that most journalists (not pundits, mind you, but journalists) take that goal very seriously and get as close as humanly possible a lot of the time.  But politically unbiased is not unbiased, and it's virtually impossible to be financially unbiased.  If the media narrative was 'Obama is walking away with this election and there's virtually nothing Romney can do to stop it'--ie the truth--people might not follow the news as obsessively and ad revenue might drop.

ahandout
ahandout

All the momentum is in Romney's favor.  That we know for sure.  Prediction:  Romney wins by 5 points.

S_Deemer
S_Deemer

Anybody who says they know who will win next week is lying. That said, polling and computer modelling have evolved a lot since 1980 (partially in response to the miscall of the 1980 election), and, based on their record and their methodologies, I'm more inclined to believe the odds offered by Nate Silver and Sam Wang, although they could be wrong. We'll know in 7 days.

DonQuixotic
DonQuixotic

Why is there a perpetual need on the media's part to second guess everything that is done in this election for the sake of appearing non-biased.  

"We have all this polling data, but is it accurate?"

"Mitt Romney lies a lot, but does Obama lie too?"

Give me a break.  You're not going to win over any Fox viewers and your efforts just come off as insincere.  If you have no certainty over anything, at all, that will happen in this election then I have to ask; why write anything?

pfcarini
pfcarini

@mrnathan.ges   Comparing Republicans fire hosing black people with Democrat voter fraud, then stating one or both allegations could be right is absurd. The former is hyperbolic rhetoric while the latter has indeed happened on several occasions based on those convicted of such.

Chuck1973
Chuck1973

@Leftcoastrocky You point to a blog that offers not one sliver of evidence. The writer points to "internal polls" but offers not one bit of proof. Yet we have hard proof that both Gallup and the NY Times published polls just days before the election of 1980 showing Carter in the lead. And I love how liberals always use words like "exploded" because they think it will give their argument more weight. And then other liberals point to that argument as proof that they are right. It's too funny. Liberals are so simple minded. If you all weren't so mean and filled with hate, I might actually think you all were cute.

jred1979
jred1979

@sacredh Nate Silver's model is based on one BIG assumption:  That the polls are correct.  The point of the article above is that the polls are not always correct.

Most of the polls I have looked at assume that voter turnout will follow roughly the same lines as 2008.  Republicans claim that the turnout among minorities and young voters will be lower this year than in 2008.  The Obama campaign claims that minority turnout will be higher, particularly among Latinos.

Nate Silver's model, while mathematically solid, is only as good as the information which go into the model.  In statistics, this is referred to as "Garbage In, Garbage Out".  The model is as good as the polls used to create it, and the polls are only as good as the turnout assumption used to weight likely voters.  If the Obama campaign is correct and the Latino vote is underpolled, then Silver is a pessimist.  If the Republicans are correct and the demographic breakdown is closer to 2004 or 2000, then Silver is in for a bad night.

sacredh
sacredh

Math and scientific oberservation is propaganda. Shakes fist at thunder gods.

ahandout
ahandout

@sacredh No, that's just more leftist BS.  You need to learn the difference.  In 5 days you will.

nflfoghorn
nflfoghorn

@sacredh Nate admittedly didn't run the latest OH poll numbers through his model yet.  I suspect this gap to be even larger Thursday morn.

mantisdragon91
mantisdragon91

@JoeBrandimore Two quick problems with you analysis:

1) Obama is viewed much more favorably than Carter was, especially in light of how he handled the aftermath of Sandy.

2) Romney doesn't have the character and charisma of a dead Reagan, much less a live one.

MrObvious
MrObvious

@JoeBrandimore 

only difference of course is that Reagan didn't spend his entire campaign flip flopping between severely and 'I'm dah moderate'

I could be wrong, but even among the Obama voters I know I don't sense much desire for a 2nd term for Obama.

Romney was laughed out of CPAC for his flip flopping - whatever enthusiasm Romney gets it feels like it came with drinking Drano first.

TyPollard
TyPollard

@JoeBrandimore 

If the Tea Party existed in 1980 as an albatross around Reagan's neck I doubt he would have won. The Republican party of today is the problem.

Chuck1973
Chuck1973

@Leftcoastrocky Again yuo are repeating things that aren't based in any fact. Just because one liberal whacko writes it doesn't make it a fact. Fact: Carter led Reagan in a NY Times poll and a Gallup poll in late October. Sorry, I know you liberals hate facts but the rest of us like to ground ourselves in reality.

PaulDirks
PaulDirks

@charlieromeobravo Actually "the accuracy of polls" is quite the hot topic at the moment and it deserves to be. The discrepancy between National polls and the individual State polls is something that needs explaining. 

nflfoghorn
nflfoghorn

This is not an election that's decided by popular vote.

PaulDirks
PaulDirks

@ahandout Unless of course you actually pay attention to data in which case you know that his 'momentum' lasted exactly 1 and a half weeks and has leveled off and is currently backsliding slightly.

MrObvious
MrObvious

@ahandout 

Momentum - ain't that one of those words where it describes something moving forward?

Yet polls show a reversal. 

mantisdragon91
mantisdragon91

@ahandout If all the momentum is in Romney's favor why are his ads getting more desperate and deceptive?

Sue_N
Sue_N

@DonQuixotic Imagine if they had put all this effort into covering (and explaining) actual issues.

snidelywhiplash
snidelywhiplash

@Chuck1973 @Leftcoastrocky I would point out that Gallup's last Presidental poll (before the storm) had Romney up by 5.  If they were unreliable in 1980... (In point of fact, Gallup isn't all that reliable - in 2008 they were pretty far down the pack in reliability.)

prestonesque
prestonesque

@sacredh The same "math and scientific oberservation" that predicted Republicans wouldn't gain more than 60 House seats in 2010.

Silver's supporters have now begun saying that even if Obama loses, the model still isn't wrong. It's being set up so that it can't fail, which also means it can't succeed. Not convincing.

Chuck1973
Chuck1973

@MrObvious @JoeBrandimore Tell you what. Give me some examples of Romney flip-flopping. Because I can give you examples for Obama - he was against gay marriage and then for it. He voted for the Iraq war and came out against it. He stated running up the national debt was unpatriotic and then ran it up more than other President in history. He promised that he would wouldn't run again if he didn't improve the economy and then he indeed did run again. And he promised to lower the unemployment rate below 5.5% and it actually rose under his watch. Okay, sorry, that last one is probably more a lack of knowing what to do then a flip-flop but the others are true examples. Now, it's your turn: Let's see if you can come-up with factual examples of Romney flip-flopping or are you just swallowing another lie told my the Obama mean machine?

sunlitweb
sunlitweb

@mantisdragon91 @ahandout For sure! Romney is struggling for air when Chrysler has to come out and spank him like a little kid for lying about their company. I think after the election Romney needs to drive a Jeep to Italy and apologize to some folks.

MrObvious
MrObvious

@Sue_N

And get off the fainting couch?

Media today spend way to much guessing and less actual reporting. Such as talking to voters; finding out and report on issues. We get more 'Will the voter ID law decide the outcome of the election?' instead of 'Voter suppression and how it effects the election'.

snidelywhiplash
snidelywhiplash

@prestonesque @sacredh You know how I know you don't understand how statistical modeling works, particularly the FiveThirtyEight model?

Silver's model said in 2010 that the *greatest likelihood* was the the GOP would pick up 54-55 seats.  They picked up 63.  That 63 number was within the confidence interval for the model, which Silver was VERY upfront in stating was showing a considerable amount of uncertainty (i.e. a wide range of possible outcomes).  The fact that the model hit a 87% success rate is pretty impressive.

It's also worth pointing out that Silver's model correctly picked 36/37 governor races in 2010.

Chuck1973
Chuck1973

@Hollywooddeed @mantisdragon91 @ahandout Look at the polls. Dig into the data. Unless the poll is weighed toward 6-8 pts more Dems turning out to vote, Romney is leading. Of course, you don't want to hear facts. You just want to close your eyes and your ears to the truth and call people names. That's all you have because you can't defend - 49 million on food stamps, the economy growing slower each year for the past 3 years, unemployment at 7.9%, and 23 millions Americans out of work. These are all facts. But go ahead, call Romney a monster and call me a name. Because that's all you can do.