Does polling matter?

February 27th, 2018

Everyone needs a fun fact to use as an icebreaker at a social gathering. My go-to is the story of how I had pizza with Walter Cronkite when I was in college at the University of Texas in Austin in the late 1980s. If you don’t recognize the name, Cronkite was a well-known journalist from the 1930s through the 1980s and anchorman for the CBS Evening News with Walter Cronkite for 19 years starting in 1962.

A small group of communication students were invited to spend time with Cronkite during a three-day symposium that focused on the future of journalism. He shared his experiences as a journalist and his post-retirement concerns about where reporting was headed. He wasn’t optimistic. One of his concerns involved opinion polling, especially the “man on the street” interview. Cronkite denounced the practice of turning to an uninformed citizen for facts and information when an expert would serve this role better.

Cronkite also worried about the overuse of opinion polls around election time, which was increasingly turning pre-election news coverage into horse races (who’s ahead now?) rather than substantive reporting (what is the candidate’s platform?). During his tenure as a national anchorman, Cronkite was known as “the most trusted man in America.” Ironically, it was an opinion poll that gave him that label.

Are polls trustworthy?

Public opinion polls are unlikely to go away. They remain popular with politicians, business leaders, journalists and even churches who want to measure what the public thinks and what it desires from its leaders. Researchers at the Brookings Institution, a Washington, D.C.-based public policy nonprofit involved in opinion research, admit that several conundrums surround opinion polling. In a 2003 article on titled “Polling & Public Opinion: The Good, the Bad, and the Ugly,” writers E. J. Dionne and Thomas E. Mann note that many people cite opinion polls, especially when it works in favor of their cause or belief. “But people are deeply skeptical of polls, especially when opinion moves in the ‘wrong’ direction,” they write.

Those who question polls have legitimate concerns, such as these: Are the questions in polls worded in a manipulative way? Are the right questions being asked? Did they ask the right group of people? “Some of the doubts are wrapped up in a mistrust of the political parties, marketers, and media giants that pay for the polls,” Dionne and Mann point out.

As we near the midterm election season in November 2018, poll reporting will increase, along with distrust of polls — and with good reason. Polls in a number of swing states predicted that Hillary Clinton would defeat Donald Trump in the 2016 presidential election. We even have polling data to show that most Americans distrust polls. A poll taken in March 2017 by McClatchy/Marist found that only 37 percent of registered voters have a great or good amount of trust in public opinion polling.

Professional pollsters, for the most part, defend their 2016 election predictions. According to a January 2018 PolitiFact article on (Tampa Bay Online), they point out national polls were correct that Clinton would win the popular vote. What they failed to predict correctly were state voting trends. Polling was less accurate in states such as Michigan, Pennsylvania and Wisconsin, and those were the states Trump narrowly carried on his way to an Electoral College win.

Polling experts also discovered in retrospect that they didn’t weight their results properly for educational attainment. If they had adjusted their predictions based on the level of a respondent’s education, results would have been more accurate. Researchers also discovered that polls that use traditional methods, such as a live pollster gathering information by telephone, are more accurate than polls that use nontraditional methods such as online surveys, politics fact-checker PolitiFact reported. Probabilistic forecasts, a new tool that aggregates multiple polls into one statistical model, are also causing some concern. According to a study by the Pew Research Center, these forecasts don’t just show which candidate is ahead at any given moment but also estimate their probability of winning the election. The media, especially cable news, used these forecasts widely in 2016. On Election Day, the odds of Clinton winning the presidential election ranged anywhere from 70 to 99 percent. According to Pew’s study, this can give voters the impression that a race isn’t as competitive as it seems when presented in percentage form and may lead to lower voter turnout.

Polling and American politics

As a college student, George Gallup was fascinated by numbers and measurements. When he edited the Daily Iowan at the State University of Iowa, he started the practice of actually asking the readers what they thought of the paper instead of just guessing, according to a November 2016 Time article.

Gallup turned his interest in polling into a Ph.D. and a career, starting the American Institute of Public Opinion. In 1936, Gallup predicted that Literary Digest would wrongly predict Alf Landon as the next president. Gallup was right: FDR remained president, and the Gallup Poll increased in popularity.

In 1948, it was Gallup’s turn to get the presidential election wrong, predicting a loss for Harry S. Truman. A Time magazine cover story written earlier that year said there was no point in arguing whether polls were good or bad — they were here to stay. The story also wisely declared that polls “can find out what the people, who rule a democracy, think and want. But a democracy also needs leadership by men who must frequently tell the people why a popular notion — no matter how widely held — can be wrong.”

Faith and polling

As demographics in the United States change and the religious landscape begins to look more and more unfamiliar, churches have turned to polling data to understand the culture they’re trying to reach. Traditional denominations, including United Methodists, are finally coming to terms with the fact that church participation is declining and that the churches of the future will look very different than the churches our parents and grandparents attended. As more and more people identify as “nones” and as more people see Christianity as irrelevant, new expressions of faith in community are needed.

Resources from organizations like the Barna Group can assist faith leaders in understanding religious trends. Founded in 1984, Barna claims it is “a go-to source for insights about faith and culture, leadership and vocation, and generations.” For example, Barna can help Christian leaders understand how Americans view the Bible. From 1991 to 2009, 45–46 percent of American adults told Barna they read the Bible at least once a week. However, by 2016, that number had dropped to one-third of Americans.

If we can refrain from hand-wringing and anxiety, this data can be valuable information. It can spur faith leaders to encourage biblical literacy, and evangelism efforts can focus on explaining biblical concepts in more modern terms. It can motivate churches to offer more classes on the Bible rather than assume people are familiar with its contents. For example, when I teach, I no longer assume people even know where the books of the Bible are. Starting with the basics, such as the table of contents, can give people confidence to go further in using Scripture as a spiritual resource.

The Good Neighbor Settlement House in Brownsville, Texas, recently used surveys to ask homeless clients what they really need instead of assuming, a February 2018 article in the Brownsville Herald says. Feedback from last year’s survey included requests for medical assistance, so Good Neighbor now runs a clinic. Assessment volunteers come long distances to serve Christ in this way. One team that helps every year is from Casper, Wyoming, 1,500 miles from Brownsville.

“So those who are last will be first. And those who are first will be last,” said Jesus in Matthew 20:16. As he proclaimed this upside-down kingdom of God, Jesus had little use for popular opinion. What he did do was listen to the hurting. Just as Christ listened with compassion, we can use information to understand a broken world better, increasing our effectiveness for ministry.

Be sure to check out FaithLink, a weekly downloadable discussion guide for classes and small groups.

comments powered by Disqus