Why the Polls in 2020 US Elections were Wrong again
Right before 2020 US Election day, all of the polls were pointing towards a landslide win for Joe Biden. It was as if Biden’s win was all but certain in the 2020 US Elections. However, the smile on the faces of the Democrats quickly faded away when the initial results came in, as Trump swept away Ohio, Florida, and Texas. Now, that didn’t last long of course, and Biden did end up winning the election. But, the outcome was a lot closer than anyone had anticipated. So, the question that arises is, why were the polls wrong again? Or more importantly, can we even rely on them?
How wrong were the polls?
Now, we have to remember that a lot of the polls for the 2020 US Elections were accurate, or at least, within the margin of error. But, the polls were way off in many states as well. For instance, Florida, Wisconsin, Iowa, Ohio, and a few others. Polls predicted that Florida and Iowa were marginally leaning towards Biden and that Wisconsin had a huge margin of 6.4%. And Ohio was marginally 0.6% in favor of Trump.
However, the results were quite different in these states. Trump won Florida convincingly by 3%, Ohio by 8%, and Iowa by more than 8% as well. And the result in Wisconsin was too close, with Biden winning the state by less than 1%. Now, these four states could have swung the entire 2020 US Election anywhere, because of the Electoral college and how the 270 electoral seats required to win can easily change hands thanks to a few thousand voters in these swing states. And the polls got them horribly wrong, far beyond the margin of error. So, what did go wrong with the polls again?
Moreover, it was claimed that polls would be more accurate this year, as now they were taking into account the education of the voter as well, which hadn’t been done before in US elections. Then why, despite an improvement in the collection of data, did the polls go wrong?
Here’s why the polls were off again for the 2020 US Elections
So, here are a few problems with polling, that have been there for many years but have just been exacerbated by the polarisation of American society, among other factors:
- Firstly, it is difficult to ascertain the likelihood of a person to vote. Polls cannot predict whether a voter will stand in line on election day if it’s raining or if any factor in his personal life impeded him from voting.
- Secondly, the COVID-19 pandemic made things very uncertain as well. People’s emotions towards the disease
- This is arguably the most divided that America has been since the Civil War. And that makes categorizing undecided voters all the more difficult.
- Furthermore, the “Lamestream media” narrative has not only affected digital media, but it has also projected on the polls. According to an expert political poller Rob Daves, it has become harder to reach voters, as they do not fill out surveys from mainstream polls anymore because of the distrust against the mainstream media. Moreover, Trump supporters or voters who lean further to the right are more reluctant to openly admit their political stance. Which brings us to our main reason why the polls failed in the 2020 US Elections and are continuously failing:
The polls failed because data is flawed
Finally, the major reason that these polls have failed in the 2016 and 2020 US Elections is the failure in the intersection of data and politics. An over-reliance on data for determining the political view of a nation with a population of 300 million and opinions differing from person to person is by its nature flawed.
Data ignores emotion, the capacity to lie, internal biases that are far beyond just education, age groups, or counties, which makes these numbers inherently unreliable. Data itself can never be neutral. It depends on the surveyor, the methodology, and the person being surveyed. And they, by no means, are neutral in their stance on politics. They have biases, which cannot be determined on any survey, regardless of the methodology. Therefore, we may never be able to get accurate polls, until or unless we are somehow able to mimic the human mind. It should only be used as a yardstick, not the final word. Otherwise, it only works to give voters a false sense of hope, or discomfort, which plays a role on election day, much like it did in the 2020 US elections.