Source: GoFish Digital
Almost everyone who wants Joe Biden to stop campaigning for president points to political polls. He was one of the reasons for them. I did that. Supporters of Kamala Harris’ potential successor cited political polls. Recent polls have fueled optimism among Harris supporters and caused panic and a change in strategy within Donald Trump’s campaign.
Opinion polls are influencing some very important decisions. Admittedly, I wish I could have trusted them more. And also in the way the media reports it.
Polls remain the best way to quickly and valuablely assess public opinion, and they are almost always accurate. But in 2016, that was wrong. Although national polls generally reflected Hillary Clinton’s lead in the vote, many state polls failed to reflect Trump’s level of support in some key states, which led to Trump losing the electoral vote. The possibility of winning was hidden. That’s why that night surprised you, me, and most of the world.
While 2020 polls showed Biden as the correct winner, they severely underestimated the strength of Trump and other Republican candidates.
“The polling industry is broken and should be blown up,” Politico wrote in November 2020. Also, a New York Post headline from that month said, “The pollsters were wrong again. Why do we listen to the pollsters?”
In the 2022 midterm elections, pollsters were criticized for predicting a Republican advance that didn’t happen. Meanwhile, polling website FiveThirtyEight.com, which also rates numerous polling organizations, declared the 2022 polling cycle to be the most accurate since 1998.
Opinion polls serve two different purposes: assessing public sentiment at a given moment and predicting election outcomes. James Stovall, a former UA journalism professor who co-directed Southern Polling in the 1980s and 1990s (with whom I occasionally collaborated on political polling on behalf of the Birmingham News at the time), said: “Well-constructed” polls can be effective, he said. It serves both purposes.
But not all polls meet that standard, he added, and polling data can be easily misinterpreted by the media and the public. “With so many polls being conducted and reported, it is not easy to say that they are always ‘accurate,'” he wrote in an email.
He pointed out that all polls have some degree of statistical inaccuracy. It happens for various reasons. For one thing, the best polls use truly random samples of the population (to increase the likelihood that the demographics of the sample match those of the population as a whole); It has become more difficult to obtain adequate random samples from the population.”
With the spread of mobile phones, it has become harder to contact people, and people have become even more reluctant to answer the phone due to telemarketing. The last time the young man answered the phone was… never. And when researchers get people to respond, they become less and less willing to talk. Some current poll managers say this is especially a problem among Republicans because many Republicans consider polls to be part of the news media and don’t like the news media. Some people talk. “Survey research is always difficult because of response rates,” Stovall said.
News organizations report on polls endlessly and sometimes commission polling companies to conduct polls. When introducing a poll, news articles should include basic information to help the public assess the value of the poll. sample size. When and how respondents were surveyed. margin of error. And the way you phrase your question can (but shouldn’t) steer your answer in a certain direction. “Unfortunately, many reporters don’t understand these factors and their importance,” Stovall said.
Margin of error is a big issue. This means recognizing that the results of a poll based on a sample may not exactly match the results of polling the entire population. If the poll shows Ms. Harris leading by 2 points and the poll has a margin of error of plus or minus 3 points, Mr. Trump may actually have a 1 point lead. Or Mr. Harris leads by five. News articles generally do not adequately explain the margin of error. Or even mention it.
According to a March 2023 article published on FiveThirtyEight.com, candidates who lead in the polls by less than 3 points only have a 55% chance of winning. “In other words, a race that’s within three points in the polls is a little bit better than a toss-up. This is what we’ve been shouting from the rooftops for years.” There will be a conflict.)
Unexpectedly low or high Election Day turnout by certain demographic groups can also mislead predictive poll results. So is the simple fact that people change their minds. Sometimes this happens due to events, and sometimes it’s just a last-minute change of heart.
Consistent headline coverage of poll results is a staple of the media every election cycle. We can only hope that such “horse race” coverage does not draw too much attention to the candidates’ policies or personalities, and that everyone understands the limitations and variables of polling. And I said that even though I knew “electability” was a huge problem for Democrats this year.
Stovall believes poll news stories are a popular part of campaign coverage. “That seems to be what viewers of political reporting want to know. And it certainly seems to be what political reporters and editors want to report.”
Tom Arenberg is a lecturer in news media at UA.
Tom Arenberg is a lecturer in news media at the University of Alabama. He worked for the Birmingham News and Alabama Media Group for 30 years. He originally published this commentary as a post on his blog, The Arenblog.
About News is a BirminghamWatch feature that publishes commentary from people who teach the craft and think about the value and performance of journalism today. BirminghamWatch also publishes About News articles on Facebook and invites readers to join the conversation there.
