News Analysis: Why The Polls Were Wrong

Pedro Wolfe, Reporter

Hang on for a minute...we're trying to find some more stories you might like.


Email This Story






Both Democrats and Republicans alike were shocked this past election day when Donald Trump was announced as the new president elect, despite every poll predicting that Clinton would win by a large margin. In the past, voting polls have made relatively accurate prediction of the final results, but this year, something obviously went wrong, leaving the American public to wonder just what happened behind the scenes

 

While polls before election day may have had different results, they all shared the same statistic that Clinton was more likely to win. At the end of October, The New York Times predicted that Clinton had a 90% chance of victory versus Trump’s 10%, while Real Clear Politics predicted a far more moderate 47.7% for Clinton and 44.8% for Trump. In the week preceding the election, polls rarely shifted, and even the day of, the results were largely the same. Real Clear Politics changed their prediction slightly to 46.8% for Clinton and 44.3% for Trump, CBS reported 45%-41% for Clinton and Trump respectively, and FiveThirtyEight, a statistics analysis publication, predicted 71.4%-28.9% in favor of Clinton. Even the New York Times, who had a wildly inflated prediction of 90%-10% back in October, had only changed their prediction to 85%-15% the day of the election.

 

The predictions themselves serve as the easiest explanation for the turn of events on election day, as many feel that voters were lulled into a false sense of security for their candidate. News coverage may have left Democrats feeling content that Clinton was going to win, keeping them at home while Republicans stormed the polls to support their underdog candidate. This certainly had some effect on the results, but the exit polls from election day show that the explanations can go far deeper.

 

Considering Trump’s polarizing comments on various ethnic groups, race was expected to play a big role on election day. The Washington Post produced an extensive poll that broke down the support from different demographics, and concluded the day before the election that 51% of voting whites favored Trump, compared to a measly 17% for hispanics and 3% for blacks. However, the exit polls ended up netting far different results. The New York Times reported that in actuality, 58% of whites voted for Trump, and even more surprisingly, 29% of Hispanics and 8% of blacks ended up voting for him as well. It might not seem like a massive difference, but the margin of error reflects numerous possible trends which are all equally disturbing.

On the one hand, it is possible that many voters lied about who they would cast their vote for on election day. Considering that both Trump and Clinton received public backlash for their respective scandals and personalities, certain voters may have felt pressured to publicly support one or the other despite their true feelings. This holds especially true for minorities, who created anti-Trump movements nationwide in response to comments he made regarding Mexicans, Muslims, and Blacks.

 

Another possible explanation is bias, or rather that news sources simply assumed certain demographics would vote for one candidate or the other. In 2012, President Obama was supported by 90% of non-white voters and 95% of black voters, and publications may have assumed similar results would follow for Hillary Clinton, the first ever female presidential nominee. In the pre-election demographic breakdown, published by The Washington Post, it was predicted that 34% of women would vote for Trump and 49% for Clinton, but on election day, the results proved to be 42% for Trump and 54% for Clinton. Once again, it is possible that voters either lied or their responses were lied about, and unfortunately, we will never know what happened.

 

Even still, more possibilities remain. Our own Professor Kovac says, “the United States has a rule that you cannot call a cell phone with an automatic calling device, and so a lot of people with cell phones aren’t contacted and our demographic of people with landlines skews the poll results.”

 

This law, which was passed in 1991 as the Telephone Consumer Protection Act, was clearly created to protect cell phone users from persistent telemarketers, but in the process it also hides a massive number of potential voters from representation in the polls. With a growing number of people shifting towards only using cell phones, these laws will need to be repealed if the pollsters hope to keep calling voters, but unfortunately, the effects of these outdated policies have already shown themselves in the polls.
We can only wait and see how polling will be conducted once the next election season comes around in 2020. Regardless of your political views, it is obvious that our polling methods are in dire need of change, but just exactly what we need to do differently is still unclear. Hopefully in four years time, the presidential election will prove less divisive, but if our current trends continue then the next election could have just as surprising results in 2016.

Print Friendly, PDF & Email

Leave a Comment

If you want a picture to show with your comment, go get a gravatar.




*

Navigate Right
Navigate Left
  • News Analysis: Why The Polls Were Wrong

    Editorial

    Kanye West … More Like Kanye Best

  • News Analysis: Why The Polls Were Wrong

    Editorial

    Confessions of a Nemo Hater

  • News Analysis: Why The Polls Were Wrong

    Feature

    Gender Roles

  • News Analysis: Why The Polls Were Wrong

    Editorial

    Point/Counterpoint: Comey’s Letter DID Cost Clinton The Election

  • News Analysis: Why The Polls Were Wrong

    Editorial

    Point/Counterpoint: Comey’s Letter Did NOT Cost Clinton The Election

  • News Analysis: Why The Polls Were Wrong

    Editorial

    SAT Memes: Self Expression or Perjury?

  • News Analysis: Why The Polls Were Wrong

    Showcase

    CSS Rugby Does It Again!

  • Arts and Entertainment

    Brining the Heights to Central Park

  • News Analysis: Why The Polls Were Wrong

    Feature

    New CSS Publication: Coffeestains

  • Editorial

    Why I Don’t Watch The Voice Anymore

News Analysis: Why The Polls Were Wrong