Polling in America has changed.
Many pollsters are no longer making a good-faith determination of public opinion on any particular issue or politician. Instead, polls are being commissioned and deployed strategically to influence public opinion, the outcome of upcoming elections, the direction of public policies, and all sorts of other things.
None of this is good of course.
In spite of the fact that polls are breaking increasingly favorably for Democrats, such does not necessarily translate into success at the ballot box. Without reliable polls, Democrats can’t even be certain that their strategic polling is having any effect.
In 2016, pollsters humiliated themselves by swearing blind that Donald Trump had no chance of becoming president. Based on polls, the best odds any mainstream media outlet was willing to give him were 15%. Even betting oddsmakers, not ones to be distracted by little things like political preferences and media narratives, lost a mint betting against Donald Trump based on the unrivaled strength of polling numbers against him.
In 2018, pollsters didn’t fare much better. Democrats managed to take back the House in 2018, yes, but the opposing party generally is more successful in the mid-term elections as a rule. Voters like checks on political power in Washington.
Still, Democrats didn’t make the kinds of gains they were hoping to make in 2018, especially considering the enormous pressure the media was then bringing to bear on Trump for the Russia and Ukraine scandals.
According to pollsters in 2020, even pre-Covid, Donald Trump was supposed to lose in a landslide. Considering how unremittingly negative media coverage of the Trump presidency was, and considering what the polls were reflecting, he should have had no chance at all.
That he came as close as he did in 2020, in spite of everything, including Covid-19, should have frightened vulnerable Democrats badly enough to stop taking polls at face value.
But aside from lamenting that Republicans, and especially Trump-voting Republicans are harder for pollsters to reach than ever and are now actively resistant to being polled, Democrats don’t seem much troubled by the phenomenon.
This is quite a surprise, considering how equally the country is divided at the moment. The ability of one political party to peel voters away from the other is the way elections will be won in the near future.
Fixing the problems in polling isn’t a small matter either, even if Democrats seemed inclined to try, which they don’t.
From 2016 on, while plenty of media outlets lamented the problems in polling, no one seemed to have much of an idea on how to improve it. Indeed, some of the most experienced pollsters wondered aloud after 2018 if polling was dead forever.
The end of the landline, the vulnerability of online surveys, the lack of a reliable random contacting method; these and other factors have made it all but impossible to get a truly randomized sample of people.
Never mind that polling has, always has had, and always will have, a major, fundamental, fatal flaw at its heart.
A poll isn’t a sample of people; it is a sample of people willing to be polled.
Not everyone is. Republicans have become increasingly harder for pollsters to reach, for various reasons, including the fact that blue collar workers don’t spend as much time online as other demographics.
People with strong opinions, one way or another, are more likely to agree to be polled. This often creates a polling illusion that only strong opinions exist on the subject. The vast majority of people, who may not have very strong opinions about the subject at all, if any, aren’t interested in being polled.
Even experimenting with paying people for their opinions has been met with limited success. It merely creates a poll of people willing to fill out a survey for money.
After this last election, media outlets began reflecting in earnest. All too soon, however, this honest reflection is likely to yield to pressure to get Democrats elected.
In one short year, it will be June of 2022; primetime for the 2022 election cycle which will be a referendum on the first years of Joe Biden’s presidency and determine which party will control the House of Representatives from 2022–2024.
Campaign season hasn’t so much already started as it never really ended in the first place. In spite of President Biden’s thin victory on the national level, down the ballot, things aren’t looking terribly rosy for Democratic chances.
At a state and local level, Republicans cleaned up in 2020. They did so by, among other things, winning every single one of the 27 races pollsters called as tossups.
There have been other bellwethers; Democrats lost a seat in California in 2020. A recent race in Texas in a heavily Hispanic district went heavily for Republicans.
It is a shift Republicans like Sen. Marco Rubio hope to make permanent in 2022; and a shift Democrats are desperate to reverse.
Polling, when used correctly, has revealed some serious problems with Democratic Party messaging to certain demographic groups.
Unless the party, and the mainstream media, is willing to make an honest assessment of the divergent political and social attitudes which prevail in groups politicians are too eager to see as homogenous, there might be some very unpleasant surprises for Democrats in the elections to come.
(contributing writer, Brooke Bell)