Political polls: they try to predict the outcome of the races, but can we trust them? -WESH 2 Orlando
The year was 1948. New York Republican Governor Thomas Dewey was widely favored in the polls and by most of America’s leading newspapers to beat incumbent President Harry S. Truman. he guns and misses, with Truman having the last laugh. The photo of him holding the Tribune with its headline “Dewey Defeats Truman” above the fold with his Cheshire cat grin is political legend. Fast forward to the 2016 presidential showdown. Polls had Democrat Hillary Clinton leading Republican Donald Trump. In fact, former Secretary of State Clinton won the “popular vote” by nearly three million votes. But Trump managed the surprise with more votes in the electoral college. Analysis of that election result reveals that pollsters seriously underestimated Trump’s popularity among less educated voters without a college degree. Those voters often do not respond to polling queries. The Hill published a poll in 2018 that showed that a majority of Americans, 52%, don’t think the polls are accurate. Most of the voters WESH 2 News spoke to agree. Eric Borelli told us, “I don’t know if it’s really accurate data.” “You don’t know what to believe,” agreed Nedra Fordham. “We pay attention to this every day. You know, most normal people don’t,” Michael Binder said with a smile. Binder tells WESH 2 that while most people don’t understand polling, as head of the Public Opinion Research Laboratory (PORL) at the University of North Florida, he says you can certainly trust him. In PORL, students organize political poll questions and then collect demographic information about the people they call, such as age, race, ethnicity, voting habits, political party, or independent status. That helps shape a complete sample for a survey. The information is obtained from voter registration records, which are public. UNF publishes the results. Sure, there is skepticism. Only three to five voters out of 100 will speak to the student callers. Call supervisor Rachel McDonald told us, “We tell them who we are, what we do, completely unbiased. We are not going anywhere in particular.” “We are not trying to sell you anything. We’re just trying to help people get information,” added student caller John Cashin with a laugh. which voters who provide email addresses when they register can receive links via email to answer questions. The UNF Voting Team He admits that voters sometimes ask if the questions are designed with a conservative or liberal bias, but the head of this lab, Binder, insists that they downplay the medium.”I think the confidence is mixed, I would say. But as far as accuracy, I think we’re better than ever,” he said. So what should voters be most aware of when looking at polls? Find out if respondents are “registered” or “likely” voters. Likely voters have a consistent history of casting votes. Is the poll specific to your state? National polls often contain fewer poll samples from where you live. And “is there a trend” showing changes in a race from the same pollster? For example, in last year’s Senate race between incumbent Marco Rubio and Val Demings, a February Mason-Dixon poll showed Rubio leading 49% to 42%.In September, the same polling firm had Rubio 47% and Demings 41% Not much difference No bias Rubio won re-election with 58% of the vote Bottom line: Voters decide elections Polls are just a tool to track political campaigns. WESH 2 also spoke to pollsters who work exclusively with Republicans or Democrats. Those firms are paid by candidates and parties to collect voter information to guide their campaigns. If you are contacted, you should ask how your survey responses will be used before providing answers to your questions. Top Headlines: Listen to the news and weather headlines with the WESH 2 podcast. Carlee Russell’s family says she was “not in good shape” upon returning home.
The year was 1948. The Republican governor of New York, Thomas Dewey, was widely favored in the polls and by most of the major newspapers in the United States to beat incumbent President Harry S. Truman.
After the polls closed on Election Day, some newspapers, including the Chicago Tribune, rushed in and got it wrong, with Truman having the last laugh. The photo of him holding the Tribune with its headline “Dewey Defeats Truman” above the fold with his Cheshire cat grin is political legend.
Fast forward to the 2016 presidential showdown. Polls had Democrat Hillary Clinton leading Republican Donald Trump. In fact, former Secretary of State Clinton won the “popular vote” by nearly three million votes. But Trump managed the surprise with more votes in the electoral college.
Analysis of that election result reveals that pollsters seriously underestimated Trump’s popularity among less educated voters without a college degree. Those voters often don’t respond to polls.
The Hill published a poll in 2018 that showed that a majority of Americans, 52%, don’t think the polls are accurate.
Most of the voters WESH 2 News spoke to agree.
Eric Borelli told us, “I don’t know if it’s really accurate data.”
“You don’t know what to believe,” agreed Nedra Fordham.
“We pay attention to this every day. You know, most normal people don’t,” Michael Binder said with a smile.
Binder tells WESH 2 that while most people don’t understand polling, as director of the Public Opinion Research Laboratory (PORL) at the University of North Florida, he says it certainly does. you can trust them.
In PORL, students organize political poll questions and then collect demographic information about the people they call, such as age, race, ethnicity, voting habits, political party, or independent status. That helps shape a complete sample for a survey. The information is obtained from voter registration records, which are public. UNF publishes the results.
Sure, there is skepticism. Only three to five voters out of 100 will speak to the student callers.
Call supervisor Rachel McDonald told us, “We tell them who we are, what we do, completely unbiased. We’re not going anywhere in particular.”
“We are not trying to sell you anything. We’re just trying to help people get information,” added student John Cashin with a laugh.
The most common and most expensive survey is done over the phone, just like what you see here. Online polls are becoming more popular, where voters who provide email addresses when they register can receive links via email to answer questions.
The UNF polling team admits that voters sometimes ask if the questions are designed with a conservative or liberal bias, but the head of this lab, Binder, insists they’re playing it down the middle.
“I think the confidence is mixed, I would say. But as far as accuracy, I think we’re better than ever,” she said.
So what should voters be most aware of when viewing the polls?
Find out if the respondents are “registered” or “likely” voters. Likely voters have a consistent history of casting votes. Is the survey specific to your state? National surveys often contain fewer survey samples from where you live. And “is there a trend” showing changes in a career from the same polling company?
For example, in last year’s Senate race between incumbent Marco Rubio and Val Demings, a February Mason-Dixon poll showed Rubio leading with 49% to 42%. In September, the same polling firm had Rubio at 47% and Demings at 41%. Not much difference. Without bias Rubio won re-election with 58% of the vote.
Simply put: voters decide elections. Polls are just a tool to track political campaigns.
WESH 2 also spoke to pollsters who work exclusively with Republicans or Democrats. Those firms are paid by candidates and parties to collect voter information to guide their campaigns. If you are contacted, you should ask how your survey responses will be used before providing answers to your questions.
Main headlines:
:quality(70)/cloudfront-eu-central-1.images.arcpublishing.com/thenational/XVFCBW2HSTJIU5H5FZSNQOJB6M.jpg)
:quality(70)/cloudfront-us-east-1.images.arcpublishing.com/tronc/457ORBYE2RBRROCBT7E6H4UQFE.jpg)
