The ending was anticlimactic, but at least it’s over. With Joe Biden’s decisive win on Saturday morning thanks to Pennsylvania’s 20 electoral votes, the 77 million — and counting — people who voted for him can breathe a collective sigh of relief and set their clocks for Inauguration Day 2021. However, as much as we’d like to move on from such a stressful week, it’s worth examining what made it so stressful — specifically, polling. Expectations and reality were, once again, vastly inconsistent.
To start, while many key states were predicted to tip in Joe Biden’s favor, the actual results ended up favoring the Trump administration for a second presidential term. We had the same problem in 2016, where everyone believed in Hillary Clinton’s lead in the polls, but the electoral results ended up clearly favoring the opposition. Even NC State was not immune to this overtrust in polls, with many NC State students not expecting the results and 2016’s Technician underrepresenting conservatives. Simply put, the polls were even more off for Democrats in those four swing states this year than they were in 2016, contributing to the four nail-biting days we sat and waited for someone, anyone, to declare a winner.
In the aftermath of 2016, there seemed to be clear reasons why the election didn’t tip in Hillary’s favor when it felt undeniable. According to Pew Research Center, pollsters underestimated Trump’s resonating support, especially among white, uneducated voters. Another reason was the prominence of “shy Trumpers,” Trump supporters who were quiet in their backing of the Republican candidate due to Trump’s brazen positions that polarized the race.
Pollsters may have also identified the wrong likely voters, severely overestimating turnout in some key districts and underestimating in others. Regardless, the upshot was that there was plenty of room to grow before the 2020 election, and pollsters adjusted their methods accordingly. This year was going to be different, they claimed.
Except, it wasn’t. As I explained earlier, the polling differences were substantially worse for Democrats in 2020, leading to election-night blues and a spark of hope for Trump supporters. Even though virtually every pollster made substantial changes for the presidential election, at a time when the stakes couldn’t be higher, the polls fell short again. Joe Biden won Wisconsin by less than a point, not 17, and Florida tipped nearly three points in President Trump’s favor when FiveThirtyEight predicted Biden to pull ahead by almost the same margin.
Even in North Carolina, Democrat Cal Cunningham was favored to beat Thom Tillis in the race for the Senate, despite a sexting scandal, and the state House of Representatives and Senate were locked in a fierce battle that seemed to be leaning blue, thanks to redrawn district lines and fierce volunteer support. Neither of those predictions came to pass, and voters from both parties were surprised. Why were the polls so inconsistent, despite more meticulous methodology and increased awareness?
We may not be able to tell for a while, but let’s remember that this election season was a chaotic one at best and catastrophic at worst. The coronavirus pandemic led to a slew of mail-in ballots from mostly Democratic precincts, lending itself to excruciatingly slow results from the states that still matter the most — here’s looking at you, Georgia — but enough states tipped in Joe Biden’s favor by Saturday morning.
There may have been a number of unforeseen “shy Trumpers” in this election as well. It wouldn’t be surprising, given President Trump’s catastrophic handling of the coronavirus pandemic and an incredibly polarizing election. Thanks to Trump’s weaponization of journalism, his supporters tend to be less enthusiastic about supporting whatever Trump slams as “fake news.” Because of this, news organizations — Technician included — can rarely integrate conservative opinions into their content unless that’s the general tone of their organization.
As a result, conservatives may feel underrepresented and less keen to contribute to journalistic endeavors, such as a massive survey to gauge political support. People are also increasingly less likely to pick up the phone or text back a pollster — I can personally attest to this, as I must have received at least a hundred phone calls and texts from unknown numbers in the weeks leading up to the election, the bulk of which I ignored without a second thought.
All of these discrepancies aren’t to say that there weren’t reasons to celebrate after looking at the polls. 50-52% of young people aged 18-29 voted in the 2020 election, almost a 10-point increase from 2016. We also filled in for at-risk poll workers, volunteering in the weeks leading up to the election. No matter who the 2024 candidates will be, it’s crucial that we bring that energy to upcoming elections and fight complacency as hard as possible. However, beyond our efforts, pollsters should tweak their methodology once again heading into future elections. Either that, or maybe we should consult a Magic 8-Ball.