The polls had suggested the BIG RED WAVE might become a BLOODBATH!!
Okay, but exactly WHOSE blood were they REFERRING to?
[PLEASE NOTE: You may assume that a bold emphasis inside a blockquote is probably mine.]
IT’S IMPORTANT TO CHECK OUT WHAT YOUR NEIGHBORS THINK FIRST ... before you cast your ballot, just to scope out which way the wind is blowing. After all, who the hell wants to be caught voting for a loser? (George Caleb Bingham, about 1846, via Wikipedia / Public Domain)
BACK IN EARLY 1970, WHEN I TOLD A FRIEND I WOULD BE LEAVING NBC TO TAKE A JOB ACROSS THE STREET AT THE ASSOCIATED PRESS, HE LAUGHED AND SAID, “I HOPE YOUR CLOSET IS FULL OF WHITE SHIRTS AND DARK TIES!!”
I asked what he meant, and he told me all the male employees at AP had to wear slacks and white shirts and dark ties. It wasn’t a dress code, so to speak, they all just chose to dress that way.
But fortunately, he was wrong. By the time I started, there was nopressure for everyone to dress alike, and in fact, I was sitting at my desk one day and happened to be wearing a white shirt and navy tie, which I would wear on occasion, when a fellow worker came over and whispered, “You know, Rick, you don’t have to wear those to work.”
Yes, I said, I realize that, and then he asked, so why are you wearingthem, and I said because I just felt like dressing this way today, to which he asked, “But as you look around the room, don’t you notice that everyone else here is a non-conformist?”
Aha! A delightful misunderstanding over the precise meaning of words!!
I laughed and, hoping to alert him to how funny that sounded, I turned and asked, “Joe, did I just hear you suggest that I, too, should dress just like everybody else — that is, like all the non-conformists in the room?”
His wide-eyed answer: “Well, yeah, but that’s only because it’s true!!”
It finally occurred to me after this recent election that the real reason people don’t seem to like Joe Biden is
... that the word had gotten around somehow that he’s very unpopular, and people just sense that other people tend not to like people who are unpopular.
But after seeing recent election results, I discovered maybe that’s just not all that true, especially after reading this in CNN’s newsletter:
Joe Biden is a stronger leader than even many Democrats thought
When Biden was asked by a reporter Wednesday what he would do differently in the remaining two years of his term, he answered: "Nothing."
That might be understandable, given Biden's track record. "The midterms make clear that Biden is a much stronger president than he is often given credit for," wrote historian Julian Zelizer. "He has been underestimated and criticized despite having a formidable first two years. The midterms should make Republicans nervous as they think about 2024."
Biden not only defeated an incumbent president, but was "able to move a formidable legislative agenda through Congress, overcoming fierce Republican opposition and even winning a few GOP votes along the way. The American Rescue Plan, the Bipartisan Infrastructure Law and the Inflation Reduction Act stand up as a historic trifecta -- a legislative track record arguably more significant than any that we have seen since President Lyndon Johnson's Great Society."
But then, there was also this:
Donald Trump is a weaker leader than many Republicans thought
The midterms were "supposed to be a shellacking of President Joe Biden, but it was Trump who got the thumping," Frida Ghitis wrote. ...
"Despite his awful showing, Trump plans to declare his candidacy soon. Most Democrats find the prospect hard to stomach, but most Republicans would also like him to just focus on his golf game," Ghitis noted.
So things turned out not to be what we had all assumed!
WHAT WERE WE THINKING, AND WHY? AND THEN, WHY WERE WE SO SURPRISED?
Maybe it had to do with polling, which is what we’re looking into this week.
IT WASN’T ALWAYS THE WAY IT IS TODAY: ... See that solid black line? That’s voter turnout in presidential election years since the nation’s founding. See the dotted one? That traces turnout in the midterm years. Back in our beginning, more voters voted during midterm elections, but not so much during the presidential years, or at least not until the 1820s and 1830s when American democracy started to kick in, with more people without property being allowed to vote. Also, Americans stopped thinking so much of Congress as running the country and started paying more attention to the man in the White House, starting with Andrew Jackson. (Courtesy Pew Research)
“Voter turnout always drops off for midterm elections, but why?”
That was asked and (maybe, sort of) answered by Drew Desilver back in 2014, on pewresearch.org:
Who turns out to vote and why is of much more than academic interest. In an era of increasingly polarized politics, campaign strategists must decide how much effort to put into persuading independent-minded voters to come out and support their candidate without antagonizing their party’s core supporters, who are more likely to vote anyway. ...
History break: ... in the early decades of the republic, midterm elections typically drew more voters than presidential contests. Back then, most states only gave voting rights to property owners, and Congress — not the presidency — tended to be the federal government’s main power center and focus of electoral campaigns.
Those conditions changed in the 1820s during the Second Party System, when most states repealed property qualifications, interest in politics soared as politicians increasingly appealed to ordinary people, and the parties directed much of their energy on capturing the White House after the disputed 1824 election (which John Quincy Adams won even though Andrew Jackson received the most votes).
By 1840, turnout among the white, male electorate topped 80%; the total number of votes cast that year was 60% higher than in 1836. (Blacks received the right to vote in 1870 with the Fifteenth Amendment, women in 1920 with the Nineteenth Amendment.)
Though political scientists long have noted the midterm dropoff, they don’t agree on precisely what it means. In an influential 1987 article, James E. Campbell theorized that “the surge of interest and information in presidential elections” typically works to the advantage of one party or the other; that party’s partisans become more likely to vote, while those of the disadvantaged party are more likely to stay home during presidential elections. Independents, “lacking a standing partisan commitment…should divide disproportionately in favor of the advantaged party.”
Midterm elections lack that “wow” factor, according to Campbell, and turnout among both partisans and independents return to more normal levels and patterns.
Okay, maybe, but I’m not sure I buy it.
I suspect it has more to do with the obvious — that our president is, to us, as Britain’s monarch has been to the UK, a visible representation of the nation, in the flesh, but one that citizens could help choose every four years.
THERE IS, OF COURSE, THAT WHOLE QUESTION OF ELECTION FORECASTING. HOW’D THOSE THINGS EVER GET STARTED?
“BE BOLD, BE WISE” ... I guess whoever wrote that didn’t know exactly how to spell, but it’s the thought that matters ... and it’s hard not to agree with the advice. (Tony Fischer via Wikipedia / CC BY 2.0)
George Horace Gallup (November 18, 1901 – July 26, 1984) was an American pioneer of survey sampling techniques and inventor of the Gallup poll, a successful statistical method of survey sampling for measuring public opinion. ...
As a teen, George Jr., known then as "Ted", would deliver milk and used his salary to start a newspaper at the high school, where he also played football. ...
In 1932, Gallup did some polling for his mother-in-law, Ola Babcock Miller, a candidate who was a long shot from winning a position as Iowa Secretary of State.
With the Democratic landslide of that year, she won a stunning victory, furthering Gallup's interest in politics.
In 1936, his new organization achieved national recognition by correctly predicting, from the replies of only 50,000 respondents, that Franklin Roosevelt would defeat Alf Landon in the U.S. Presidential election. This was in direct contradiction to the widely-respected Literary Digest magazine whose poll based on over two million returned questionnaires predicted that Landon would be the winner.
Not only did Gallup get the election right, he correctly predicted the results of the Literary Digest poll, as well as using a random sample smaller than theirs but chosen to match it.
Twelve years later, his organization had its moment of greatest ignominy, when it predicted that Thomas Dewey would defeat Harry S. Truman in the 1948 election, by between 5% and 15%; Truman won the election by 4.5%. Gallup believed the error was mostly due to his decision to end polling three weeks before Election Day, thus failing to account for Truman's comeback.
OKAY, SO WHAT ARE THE MOST RECENT JOB APPROVAL RATINGS FOR JOE BIDEN?
FiveThirtyEight: (observed Day 666; 11/16/22)
Polls of likely or registered voters
Approve: 43.2% Disapprove: 53.0%
All polls
Approve: 41.6% Disapprove: 53.3%
Polls of adults
Approve: 40.2% Disapprove: 53.3%
Polling dates: Oct 3-20, 2022 (Before the midterms)
Latest job approval rating: 40%
Latest job disapproval rating: 56%
Latest job approval by party ID:
GOP 4% - IND 39% - DEM 85%
Biden, low ratings? Hah! That 85% approval by Democrats doesn’t look all that shabby after one just cancels out the 4% of them pesky Republicans!
But do the rankings of presidents really even matter?
Abraham Lincoln, Franklin D. Roosevelt, and George Washington are most often listed as the three highest-rated presidents among historians. More recent presidents such as Ronald Reagan and Bill Clinton are often rated among the greatest in public opinion polls, but generally do not rank as highly among presidential scholars and historians. ...
Historian Alan Brinkley stated that "there are presidents who could be considered both failures and great or near great (for example, Nixon)". Historian and political scientist James MacGregor Burns observed of Nixon: "How can one evaluate such an idiosyncratic president, so brilliant and so morally lacking?”
This article, “Pollsters fear they’re blowing it again in 2022”, by Steven Shepard, was on Politico’s website in late September, reflecting worries ahead of the upcoming midterms vote.
Since Donald Trump’s unexpected 2016 victory, pre-election polls have consistently understated support for Republican candidates, compared to the votes ultimately cast.
Once again, polls over the past two months are showing Democrats running stronger than once expected in a number of critical midterm races. It’s left some wondering whether the rosy results are setting the stage for another potential polling failure that dashes Democratic hopes of retaining control of Congress — and vindicates the GOP’s assertion that the polls are unfairly biased against them.
Were polls broken in 2016 when they predicted that Hillary Clinton would win? Here’s a political scientist, interviewed by Bill Boyarsky back in the fall of 2019:
“No, I think that’s ridiculous,” said UCLA political scientist Lynn Vavreck.
“Hillary Clinton won the popular vote. The polls showed she was going to win the popular vote. They were closer in 2016 than they were in 2012 in the actual popular vote election outcome, which is what most of these polls are measuring.”
Boyarsky, a former City Editor of the L.A. Times, points to one of the things that went wrong in 2016:
[Most] organizations polling the states failed to catch a key factor: Older white men with high school educations or less supported Trump in the Midwestern battleground states, where polls showed that Clinton was favored — but Trump won narrowly. Many analysts felt this was the pollsters’ biggest mistake of 2016.
“Education was strongly correlated with the presidential vote in key states: That is, voters with higher education levels were more likely to vote for Clinton,” said the American Association for Public Opinion Research. “Yet some pollsters — especially state-level pollsters — did not adjust for education in their weighting, even though college graduates were over-represented in their surveys...”
Wait!! So what’s that “adjust for...” stuff? You mean pollsters would have had more accurate guesses had they only “tweaked” a few things? Is that ethical?
Courtney Kennedy, a VP at Pew Research, explained it in “Key things to know about election polling in the United States” back in August of 2020:
All good polling relies on statistical adjustment called “weighting” to make sure that samples align with the broader population on key characteristics. Historically, public opinion researchers have relied on the ability to adjust their datasets using a core set of demographics to correct imbalances between the survey sample and the population.
There is a growing realization among survey researchers that weighting a poll on just a few variables like age, race and sex is insufficient for getting accurate results. Some groups of people – such as older adults and college graduates – are more likely to take surveys, which can lead to errors that are too sizable for a simple three- or four-variable adjustment to work well. Pew Research Center studies in 2016 and 2018 found that adjusting on more variables produces more accurate results.
A number of pollsters take this lesson to heart. The high-caliber Gallup and New York Times/Siena College polls adjust on eight and 10 variables, respectively. Pew Research Center polls adjust on 12 variables. In a perfect world, it wouldn’t be necessary to have that much intervention by the pollster – but the real world of survey research is not perfect.
Failing to adjust for survey respondents’ education level is a disqualifying shortfall in present-day battleground and national polls. For a long time in U.S. politics, education level was not consistently correlated with partisan choice, but that is changing, especially among white voters. As a result, it’s increasingly important for poll samples to accurately reflect the composition of the electorate when it comes to educational attainment.
Since people with higher levels of formal education are more likely to participate in surveys and to self-identify as Democrats, the potential exists for polls to overrepresent Democrats.
But this problem can easily be corrected through adjustment, or weighting, so the sample matches the population. The need for battleground state polls to adjust for education was among the most important takeaways from the polling misses in 2016.
I suppose all the “weighting” and “adjusting for” and “correcting for” might be okay as long as you’re just trying to get a “match” of the population,
… and not just cooking the books by always trying to fight the next war by fighting the last one -- which, it seems to me, kind of defeats the point of forecasting.
For more war stories from the history of political polling, Erin Overbey authored a profile of the travails of Gallup over the years in New Yorker back in 2012:
On August 17th [of 1960], as Nixon and Kennedy were preparing to enter their final three months of campaigning, Gallup — which was at that point the preëminent polling service in the country — published a poll announcing that fifty per cent of Americans favored Nixon, forty-four per cent favored Kennedy, and six per cent were undecided.
Two weeks later, on August 31st, Gallup published a new poll, showing Kennedy and Nixon in a virtual tie, each with forty-seven per cent of the vote.
Many pundits were flabbergasted. How could a race that had consistently been so far apart now be tied? And how valid were these numbers to begin with?
Shortly thereafter, Joe Alsop followed up in the same magazine:
In 1944 (when soldiers overseas could not be polled), Gallup underestimated the Democratic vote by 2.3 per cent of the national total, or 1,105,000 votes.
In 1948, the year Thomas E. Dewey failed to beat Harry S. Truman, Gallup underestimated the Democratic vote by 5.4 per cent of the national total, or 2,637,000 votes.
In 1952, Gallup went just about as far in the other direction, overestimating the Democratic vote by 4.4 per cent of an increased national total, and thus cheating the Republicans of 2,714,000 votes.
And in 1956 he again underestimated the Democratic vote, that time by 1.7 per cent of the national total, or 1,054,000 votes.
No, wait! There’s apparently a logical explanation!!
At some point in every campaign, a Presidential aspirant or a member of his team “becomes convinced that some sort of dark design lies behind” these numeric discrepancies. But as Alsop points out, many of Gallup’s numbers, and Kennedy’s lower percentage in the earlier poll, could easily be explained by the organization’s painstaking process of reducing the number of non-voters and “undecideds” in their poll samples.
Take the example of the August 17th poll, which showed a huge, six-point advantage for Nixon. Alsop wrote that the first step of the Gallup organization’s process was usually a reduction of the total sample.
Gallup aimed to make its predictions more accurate by throwing out a large proportion of the ballots — typically about twenty per cent. Some respondents, Gallup believed, simply won’t vote, and so their responses shouldn’t be counted in the poll.
The motivations for this move are non-partisan. But the result, Alsop notes, invariably reduces the Democratic percentage, and that of undecided voters. As many pollsters know, “non-voters are predominantly found in the lower-income groups, which are predominantly Democratic.” Fail to correct for this, and your poll will give a consistently inaccurate edge to the Democrats.
I’ve tried to make sense of statistician Nate Silver’s stated methodology, but failed. (Still, truth be known, because I flunked fifth grade arithmetic, I had to take that whole semester over again.) Yet I couldn’t fail to notice that Silver’s FiveThirtyEight gang made their fame after
... an election forecasting system he developed successfully predicted the outcomes in 49 of the 50 states in the 2008 U.S. Presidential election.
In the 2012 United States presidential election, the forecasting system correctly predicted the winner of all 50 states and the District of Columbia.
But although political forecasting still does look pretty sketchy, I guess there must be something to it.
BUT YOU KNOW WHAT I FIND A BIT DISQUIETING IS HEARING CNN’S (extremely smart) JOHN KING, DOING HIS SPIEL WHILE STANDING AT THE “MAGIC BOARD” ON ELECTION NIGHT, AND HAVING TO REMIND MYSELF WHENEVER I HEAR HIM SAY SOMETHING LIKE,
… (just paraphrasing here), “Now, if I’m Senator Shmendenic right now, I want to see if there aren’t any more votes to be found in southern Honkytonk County…”, that, at that point, we know all us political junkies are just playing a game of Monopoly, and that, since the polls closed hours ago, there’s nobody in Honkytonk County doing anything to pull Senator Shmendenic’s sorry-ass career out of the dumpster right now.
But then, I’m also wondering if, by any slim chance, former president Marshmallow-Brain might be listening to King say this, and think, “God Damnit!! If somebody’s there right now, doing this for Shmendenic, why can’t they do it for me?”
BUT HERE’S THE THING ...
CAN SOMEONE BE THE “ONLY” CONFORMIST IN A ROOM FULL OF “NON-CONFORMISTS”? CAN THE REAL REASON SOME GUY ISN’T LIKED BE THAT HE JUST ISN’T “POPULAR”?
And can the answer to both of those be, “no” and “no”?
(That answer is yes.)
Which is to say that, if people who hate so much on a president who’s actually doing a fairly good job of trying to do the right thing, simply because they heard somewhere in Tweetybirdland, that everybody else does,
… and that’s why they think the country is going in the wrong direction? Then maybe the Trumper Troopers are right about democracy after all … that maybe all the non-nitwits in America need to give in and move to Costa Rica.
Although I might be looking at this wrong. We didn’t really lose, and nor did those people really win! They got cocky, too sure of themselves, and got taught a lesson about what Americans are really thinking … assuming, that is, that they can see it.
Maybe what we need to do is just out-and-out abolish electoral polling, or at least the polling that forecasts some candidate’s chances of winning an election?
After all, if our showing up to vote depends on how well we think they’re doing, rather than on whether or not we want them to win, then we could be leaving the decision up to chance than to the will of the people.
Might not we be better off if we had no idea how it’s going? For one thing, that might make watching election night on TV more exciting!
But no, making political polling illegal, of course, would never work.
Too many people -- the campaigns, for one example, and bookies, for another -- absolutely need to see how things are going, and without someone supplying this information, it would go underground, with a cottage-industry of back-alley speakeasies would proliferating to serve a population of campaign staff and journalists and college professors (and, of course, actual bookies), but similar to the bootlegged swill in the 1920s, the quality of the product might be like something that was created in a bathtub, or worse, some internet chat room.
Besides, what always goes unmentioned as an actual legitimate reason for voting is to chose someone that other people also find acceptable, and how can we do that if we don’t know what other people are thinking?
The unmentioned secret sauce as to how self-governing works is for all the individuals in the group to decide how to make a group decision work. We should note the failure of the process whenever we hear somebody say, which happens way too often,
“I only watch TV news that gives me nothing but the straight facts, not opinions, and that lets me make up my own mind! Why do I need anybody telling me what to believe?”
But there actually is a real answer to that, and it comes in the form of another question:
Why would anyone want to live in a land ruled by voters with bananas-for-brains who are too arrogant to listen to anyone’s opinions other than their own?
