In a presidential election year, news broadcasts—including internet news sites—are inundated with the ‘latest’ presidential polling results. Biden is up. Trump is up. Every time we see one, we’re either buoyed or distressed depending on your candidate. Thankfully, we can always take heart with the ever-present addendum, ‘margin of error plus-or-minus three to five percent.’ There’s always a chance.
And while polls, presidential or otherwise, are usually on or, at least, close to the mark for accuracy, there have been some foul ball predictions over the years.
Ever hear about President Thomas Dewey? Or more recently, President Hillary Clinton? Didn’t think so. Even the most highly regarded and respected polling can now and then do an epic faceplant. In 2016, we saw it—in spades!
Up until Election Day 2016, a Huffington Post poll stated confidently that Hillary Clinton had a 98 percent chance of beating Donald Trump. Even Trump, as has been written, thought he’d lose. Oops! But the same poll redeemed itself a few years later when it correctly picked Joe Biden over the incumbent.
Because the whole country can’t be polled in a presidential election, a sample size of voters, somewhere between 1,000 to 1,500, is queried. Years of refining the science has proven this sample size to be an accurate cross section. Not a hundred percent, as we have seen. But, generally accurate.
Metropolitan State University-Denver’s Rob Pruehs discusses polls regularly in his classes. “I tell them (students), there’s always going to be some errors in polling.” They have limitations. They are, he said, “only as good as their degree of sampling.”
Timing is another consideration, said Pruehs. Right now, the economy is on the upswing. But that could change and change people’s minds between now and November. Other things, world events, wars or as President Carter learned with Americans held hostage, can also flip-flop conventional thinking quickly.
“We see good economic news,” said Pruehs. And that bodes well for Biden. Something else that may play well for Biden, is “the chaos and perhaps even criminality of the ex-president.” But criminal exoneration could also upend conventional thinking on Trump. Indeed, things can turn suddenly.
It may be too easy to write off polls, especially those that go against one’s own personal leanings. Pruehs suggests the best way to weigh a decision is to not focus on just one poll. “You’ll get a better sense if you look a number of good polls. ”Also, consider the individual polling source.
538, owned by ABC News and created by highly respected pollster and former New York Times reporter Nate Silver, is an organization that examines opinion polls. It analyzes everything from politics to economics, and sometimes even sports. It regularly ranks dozens of polling organizations for their accuracy.
Recent rankings had The New York Times, ABC News/ Washington Post and Marquette University Law as the gold standards in polling. Others in the top ten include Suffolk University, Marist College and Emerson College. Each earned three stars, the highest rating in the survey. (In case you’re wondering; McLaughlin & Associates earned only one half of a star. A number of other polling services got none.)
Polls can be accurate barometers of final outcomes. But it is always wise to know a bit about a poll before making any final decision. It’s smart to see the manner in which questions are asked, said Pruehs. It’s also a good idea to see who conducted the polling.
Good questions to ask include: Who funded the poll? Who was actually polled? What was the sample size and sampling methods? What was actually asked of people contacted? In an era of changing demographics, it would be wise to inquire about that, too.
Also, never forget that single variable that throws poll results completely off: the fickleness of the American public.
As the election of 1935 drew close and the country was in the throes of The Depression “relief and recovery” were the hottest issue of the day, Gallup was also conducting its first ever presidential poll.
Its poll showed the majority of Americans were skeptical about the government’s plan for turning things around. Six in ten Americans thought the government was spending too much on the aforementioned ‘relief and recovery.’ But the poll also showed Democrats more in favor of government programs than Republicans, 90 percent of whom opposed Roosevelt’s solution and leadership. That time, Gallup got it right, FDR won and polling has only continued to grow.
Interestingly, while technology on the polling side has changed dramatically, so too have those contacted for the polls. Not that long ago poll calls were made exclusively to households with a land line. A poll today targeting only land lines would be almost automatically tossed out given the reality that land lines are disappearing by the day, even the hour. Spanish language polling has also found a place in election cycles.
Between now and November, expect to see and hear polls predicting a reelected incumbent or a former president pulling out victory. It’s your pick.
Will an upsurging economy propel President Biden to a second term? Will a metastasizing Middle East conflict turn him into a 21st century Jimmy Carter, a losing incumbent? Or will the legal system now churning like riptide swallow the expresident? Can he stall the various court cases long enough to outlast the calendar? Finally, could he find himself behind bars?
Or, as has been asked by scores of pundits and pollsters, will time and all that comes with it, catch up with the two current front runners? If that’s the case, who then might be the ‘bolt-out-of-the-blue’ next presidential election contenders?
