What's Right, Wrong With Today's Polls: Former Pollster

Written by Mayor Mary Marvin:

Offering a change of pace, I decided to write about a topic that has always fascinated me: political polling. As background, I spent the early years of my career working for a D.C. pollster and strategist who only represented moderate Republican clients such as Nelson Rockefeller, Pete du Pont, Lamar Alexander and Bob Dole. (As a sidebar I was on the team that came up with the slogan Re-Pete Du Pont; cringe-worthy, but it worked!) My fascination with polling of late is how inaccurate it has been in so many races so I delved into the Whys.

According to my research, in three of the last four national election cycles, the polls have significantly overestimated the performance of the Democratic candidates. Just two weeks ago in Virginia and New Jersey were stark examples.

The national polls as an aggregate in 2020 were the least accurate in 40 years and state-level polls the worst in two decades. As an example, in New Hampshire, Senator Susan Collins never led in a single poll throughout the campaign, yet won handily.

It appears the crux of the problem is trying to gain a representative cross-section of voters in the electronic age we live in with the almost universal use of cell phones and caller ID. The Pew Research Center estimates that only 6 percent of the people polling firms call actually respond, down from nearly 50 percent in the late 1980s.

With many Americans abandoning landlines, pollsters have been forced to call cell phones, but federal law prohibits them from making automated calls to these phones, making polling more expensive.

Also, the dwindling number of people who still respond to polls are generally older, whiter, more educated and more likely to be female than the voting public.

Pollsters try to correct for this skew but the 2016 presidential election proved it didn’t work. Most polls predicted Hillary Clinton would win the national popular vote by 3 percentage points — it was only one point off, well within the margin of error.

But on the state level, pollsters missed a large last-minute surge in support for Donald Trump in Wisconsin, Michigan and Pennsylvania.

The inaccuracies continued in the 2020 national elections. Biden won with 51.3 percent vs. Trump's 46.8 percent of the popular vote — a healthy margin of victory but most polls predicted a much larger margin and a “Blue Wave” of Democrats in the House which did not materialize. A postmortem report found that pollsters had overstated Biden’s victory margin by 3.9 percent — the industry’s worst performance in 40 years.

Reasons postulated for this unprecedented miscalculation include the new normal brought on by the pandemic. One theory is that the left-of-center voters were more likely to follow social distancing guidelines and stay home, making them more reachable by phone. Trump also repeatedly told his supporters not to trust “fake” polls, perhaps influencing them to ignore the calls.

Also in the case of the South Carolina senatorial election, many voters expressed an interest in a fresh face, yet in the end due to the incredible perks of seniority, realized Lindsey Graham could deliver a lot more money to the state than the newbie, Jaime Harrison. It became an issue of simple economics.

Faulty polls don’t always tilt leftward, however. In 2012, many polls, including Gallup, predicted a narrow victory for Mitt Romney; instead, Barack Obama was re-elected comfortably.

Just last September though, a majority of pollsters predicted that Californians would vote against recalling Gov. Gavin Newsom by a 14-point margin, but he survived by nearly double that.

In an effort to fix what has been an embarrassing streak, pollsters are putting together an amalgam of methods to reach voters including online, text-based and interactive phone response systems in addition to landlines.

Pollsters are also changing what they ask. As example, Pew has added screening questions designed to measure a respondent’s trust in institutions and weigh the mistrusting responses more heavily since they are more likely to be underrepresented.

The jury is out on whether to trust polls. Some experts say that all polls have a margin of error so a two-point “upset” isn’t really a fair evaluation of the pollster.

In addition, because polls are snapshots in time, they can mislead, since events can change how people vote right up to the last minute. A great example of this was FBI Director James Comey’s announcement that he was looking into Hillary Clinton’s emails just 11 days before the 2016 election.

In the end, most political experts believe that despite the significant miscalculations of late, the industry will continue as polling is just too ingrained and attached to the American political process.