Commentary
How Many Polls Does It Take To Screw In A Lightbulb?
- by Joe Mandese @mp_joemandese, November 16, 2020
If the biggest winner of the 2020 U.S. presidential election was American democracy, the biggest loser was political polling -- and by extension -- survey-based consumer research.
Since presidential campaigns arguably are the highest-stakes form of marketing, then polling is also the highest-stakes form of marketing research. And while caveats …
8 comments about "How Many Polls Does It Take To Screw In A Lightbulb?".
Read more comments >
Next story loading 
Joe Mandese is the Editor in Chief of MediaPost. You can reach Joe at joe@mediapost.com.
Joe, as you probably know, I offered some of the same points about the political polls in a recent post on MP. However, what is still perplexing is the fact that the national polls came pretty close to predicting the vote---as opposed to the electoral college outcome--and this was also true in 2016. However, the state polls---in so-called "swing states" were off to a disturbing degree. This suggests that lying and refusing to cooperate---by Trumpists---was more pronounced in those areas because the voters knew how important their votes might be. I woder how accurate the same kinds of polls were in states where there was no doubt about the winner---like New York, California, Oklahoma or Idaho? If these were highly accurate, one might conclude that this was mainly a "swing state" issue. If not, it indicates that serious rethinking of their methodologies by the pollsters should be a top priority.
How many people will not admit they are racist or support fascism when they vote for someone who is ? They are in the mix.
There is a "C" answer. This is people are burned out over the past 10 years of the number of spam telephone calls. Everyone hates them and are taking measures to block calls. Things are so back that even legit callers are having problems getting through. In short, polls have been pegged amoung the worst or the worst no matter the good intentions.
When pollsters overestimate the outcomes of two consecutive Presidential elections by such a huge margin, one wonders whether polling is a form of vote suppression. I wonder how many undecided voters stayed home when they were convinced by the polls that the candidate to whom they leaned simply could not win. It is fitting that the same suppression that elected Trump also may have proved his undoing. Mission accomplished.
Douglas, if the pollsters did it intentionally, I'd agree with you that it's a form of voter suppression---but I doubt that that's the case as they would be drummed out of the business if it became known that they slanted their studies in this manner. For example, the Fox News Channel is an ardent supporter of Trump, yet many of its polls showed him in a worse light than those sponsored by "liberal media". Wouldn't you expect the opposite ---if they were engaged in voter suppression? More likely we are seeing an unusual situation, caused by the hatred and fear that has been generated in a divided country playing out---to the chagrin of mostly honest pollsters.
There is one other issue with pollsters and their callers. Most work on a hourly basis. However some work on hourly and bonus for completed survey. Last is all commission surveys. Hourly is best but the numbers and easliy be corrupted.
It's been a long time since I've come upon a "shy" Trump voter.
@Josh Chasin: That is the best line ever!