WASHINGTON TIMES: DAVID BOSSIE: Where’s the accountability in the polling industry?
Following all the polls in the run-up to Election Day has become something of a national pastime in recent years. Keeping up with the various races — who’s up and who’s down on any given day — gives us a snapshot of what might happen on Election Day.
This year, however, we’re seeing a constant and unprecedented deluge of polls in races all over the country. I believe this oversaturation is continuing a dangerous practice that we witnessed in 2016 and 2020. The sheer volume of polls being released and amplified by the biased liberal mainstream media — many that turn out to be dead wrong after the fact without any meaningful accountability — has created a new phenomenon in American politics. It’s the dark art of conducting surveys and releasing data in a strategic manner during our troublesome new monthslong voting process. It’s not done just to inform the public, but also to influence the outcome of elections.
For example, in the 2020 U.S. Senate race in Maine, the last eight public polls showed Sen. Susan Collins trailing by an average of 5.5 percentage points. On Election Day, however, Mrs. Collins was reelected by a margin of 8.6% — a 14-point discrepancy between the polls and what actually happened.
Similarly, in the Senate campaign in South Carolina the same year, four of the last six polls released to the public showed the race between Sen. Lindsay Graham and his Democratic challenger either tied or had Mr. Graham leading by a single percentage point. But at the end of the day, Mr. Graham easily prevailed 54.5% to 44.2% — essentially a landslide by today’s standards.
And who could ever forget the infamous ABC News/Washington Post poll released just over a week before Election Day 2020 that found President Donald Trump trailing candidate Joe Biden by a whopping 17 points in Wisconsin? As it turned out, Mr. Biden carried the Badger State by a razor-thin 20,000 votes out of over 3.2 million votes cast — a scant margin of 49.6% to 48.9%.
It’s difficult to quantify the precise impact of all the polls that are released and the advantage gained by releasing the data. For instance, did Sens. Collins and Graham actually win by a larger margin than expected because the polls created a sense of urgency for their supporters to get out and vote? Or did the inaccurate polling make their opponent’s campaigns artificially stronger?
That said, in the case of the Trump-Biden Wisconsin poll it’s simply a matter of common sense. If voters wake up a week before Election Day and read that their candidate is trailing by 17 points, it could very well cause them to “stay home.” I said it at the time and I’ll say it again: It was irresponsible to release this poll. To me, it was an overt political maneuver designed to suppress Mr. Trump’s vote in the critical battleground state of Wisconsin.
Once again this year, there’s mounting evidence that polls are being released to affect the upcoming election results. Questionable data swirling around on the critically important congressional generic ballot question is always ripe for speculation. In 2020, for example, the final generic ballot average of polls showed a +7 advantage for House Democrats, even though Republicans ended up picking up 14 seats in Congress. And as we approach Election Day this time around, I remain skeptical of some of the methodologies that are being used.
Five recent polls have come out showing the Republicans averaging nearly a 5% advantage on the generic ballot question. If a D+7 advantage achieved a 14-seat again for the GOP in 2020, what does an R+5 edge translate into for Republicans in 2022? This question is giving the left heartburn, which is why the recent AP-NORC survey was like manna from heaven for Democrats. Compared with the other surveys, the recent AP poll showed a 41% to 34% Democratic advantage on their generic congressional test — nearly a 12-point swing from the others.
Upon further examination, it’s possible that these results were achieved because the sample consisted of 1,121 “adults,” as opposed to “registered voters” or “likely voters.” Either way, the poll is now out there making the rounds on the internet and social media, where the methodology isn’t much of a concern — essentially amounting to a large in-kind contribution for desperate Democrats.
These polls and others are being released in specific races that amount to election interference. For example, in the House race in Michigan’s 7th Congressional District, a poll was put out recently showing incumbent Elissa Slotkin leading her Republican challenger, state Sen. Tom Barrett, by a wide margin of 56% to 38%. Considering the fact that Michigan’s 7th is now an R+2 district after redistricting in a year when the political winds are blowing in the GOP’s direction, I find these numbers hard to believe.
In addition, other polls are showing that Republicans have a 4-point statewide generic ballot advantage in Michigan and Mr. Biden’s approval rating is underwater in the state, 46% to 52%. To me, this has all the markings of a race that could easily flip to the Republican column this year. We’ll just have to wait and see if the aforementioned poll succeeds in suppressing GOP turnout in this district —I sure hope not.
After the smoke clears on Election Day 2022, it’s become obvious to me that there needs to be an honest conversation about the lack of accountability in the media and polling industry for using survey data to blatantly influence the outcome of elections.
• David N. Bossie is president of Citizens United and he served as deputy campaign manager for Donald J. Trump for President.