Here’s Why You Shouldn’t Trust Polls

In our modern world, public polls are still assumed to be accurate representations of people’s sentiments on current issues.

As in any other election cycle, polls have been utilized — almost exclusively — to determine and understand the current state of the 2016 U.S. Presidential race. These polls tell both the public, as well as the candidates, how each contender is fairing.

At least, they do that in theory.

Not only do these polls have implications for shifting public sentiment, but based on their standing in them, a candidate may choose to shift strategies or even drop out of the race before a single ballot is cast.

But basing a campaign strategy on poll results is dangerous and unwise — largely because these polls are often inaccurate.

Some polls predicted Mitt Romney to barely defeat Barack Obama in the 2012 Presidential Race.

Polls in Kentucky predicted Mitch McConnell to lose to Alison Lundergan Grimes in 2014, but McConnell went on to win by close to 20 points.

Outside of the U.S., polls inaccurately predicted the results of the most recent Israeli and British elections.

President Romney? Yeah, not so much. Photo Credit: James Currie/ Flickr (CC By 2.0)

President Romney? Yeah, not so much. Photo Credit: James Currie/ Flickr (CC By 2.0)

These erroneous results have lead analysts to question the industry and whether it has kept up with a “rapidly transforming, highly-mobile electorate.”

Not only have polls become less accurate, but their results are often oversimplified and sensationalized.

Polls conducted with 300 or so participants are said to describe a group as large as “the Democrats” or “the Republicans.”

If our polls are becoming less accurate, the fact that their results are sensationalized to increase a partisan divide is extremely worrisome.

Michael Traugott, a University of Michigan political science professor told US World and News Report, that polls “give the public an independent voice that’s not generally present” in other news coverage.

Recent polls are coming back with false results due to a lower number of responses as well as a concentrated group of respondents.

Polls are targeting people’s home phones (in addition to cell phones), which due to our culture, are now being answered less than before.

Not only are they being answered less, but more than 40% of American adults, such as myself, no longer have landlines.

Those with landlines are often older and poorer.

86% percent of people over 65 still have landlines. That number drops to 44% for those between the ages of 25-29.

Further adding to the woes of pollsters is the 1991 Telephone Consumer Protection Act, which bans autodialing to cell phones.

This law applies to public-opinion polling and market research, thereby limiting their reach. One would assume that increasingly inaccurate polls which reflect an extremely small slice of the population wouldn’t be utilized to represent the sentiments of whole groups.

Unfortunately, this is not that case, as these inaccurate, non-representative polls are utilized to represent the views of “Democrats” or “Republicans.”

Wilson Perkins Allen Opinion Research recently stated that “44 Percent of Democrats Support Taking Refugees from a Fictional Country.”

This poll was issued in response to a left-leaning Public Policy Polling result which found that 30% of Republican voters would support bombing Agrabah, a fictional country/city from the Disney film Aladdin.

The WPA poll, reporting to represent the entirety of “Democrats,” polled 384 registered Democrats, and 1,132 total voters.

Utilizing 384 people to purport to represent the entire Democratic Party is absurd and wildly disingenuous.

Utilizing 384 people to purport to represent the entire Democratic Party is absurd and wildly disingenuous.

To be clear, it is not just the WPA nor solely Right-leaning polling sources which perpetuate this nonsense.

The original Public Policy Polling survey polled 532 Republicans to represent the entirety of the party.

Public Policy Polling even went as far as to post the results on their Twitter, stating “30% of Republican primary voters nationally say they support bombing Agrabah. Agrabah is the country from Aladdin,” along with the mocking hashtag of #NotTheOnion.

In an attempt to demean those they disagree with politically, PPP claimed to represent all “Republicans” with 532 respondents. Both left and right-wing pollsters and media groups utilize these misrepresentative pols to decry the other.

There is no question that polling has a future in our political process, but we must do better. Polling 400-500 people and labeling the results as representing the entirety of a political party, race, or religion is both demeaning and reckless.

More importantly, they are inaccurate, and promote false narratives. 44% of all Democrats certainly to not support taking refugees from a fictional country. 30% of Republicans also certainly do not support bombing the same fictional country.

This sensationalism shows Democrats to be soft and overly welcoming while it decries Republicans as warmongers. Instead of finding issues to agree on, this type of sensationalism further divides the two parties.

We should try to figure out a better way to do it.

RISE NEWS is a grassroots journalism news organization that is working to change the way young people become informed and engaged in public affairs. Anyone can write for you us as long as you are fiercely interested in making the world a better place. 

Cover Photo Credit: Gage Skidmore/ Flickr (CC By 2.0)

What Do You Think?


About the Author
Charles Diringer Dunst is a Government major at Hamilton College, and is scheduled to graduate in 2018. A New York City native, Dunst can be followed on Twitter @CDDUNST.
Scroll to top