1. Less than 20 percent of Americans regularly attend church—half of what the pollsters report.
While Gallup polls and other statisticians have turned in the same percentage—about 40 percent of the population—of average weekend church attendees for the past 70 years, a different sort of research paints quite a disparate picture of how many Americans attend a local church on any given Sunday.
Initially prompted to discover how church plants in America were really doing, Olson, director of church planting for the Evangelical Covenant Church (covchurch.org), began collecting data in the late ’80s, gradually expanding his research to encompass overall attendance trends in the church. In his study, he tracked the annual attendance of more than 200,000 individual Orthodox Christian churches (the accepted U.S. church universe is 330,000). To determine attendance at the remaining 100,000-plus Orthodox Christian churches, he used statistical models, which included multiplying a church’s membership number by the denomination’s membership-to-attendance ratio.
His findings reveal that the actual rate of church attendance from head counts is less than half of the 40 percent the pollsters report. Numbers from actual counts of people in Orthodox Christian churches (Catholic, mainline and evangelical) show that in 2004, 17.7 percent of the population attended a Christian church on any given weekend.
Another study published in 2005 in The Journal for the Scientific Study of Religion by sociologists C. Kirk Hadaway and Penny Long Marler—known for their scholarly research on the church—backs up his findings. Their report reveals that the actual number of people worshipping each week is closer to Olson’s 17.7 percent figure—52 million people instead of the pollster-reported 132 million (40 percent).
“We knew that over the past 30 to 40 years, denominations had increasingly reported a decline in their numbers,” Marler says. “Even a still-growing denomination like the Southern Baptist Convention had reported slowed growth. Most of the mainline denominations were all reporting a net loss over the past 30 years. And at the same time, the Gallup polls had remained stable. It didn’t make sense.”
The Halo Effect
What Hadaway and Marler, along with Mark Chaves, author of the “National Congregations Study,” discovered was at play is what researchers call “the halo effect”—the difference between what people tell pollsters and what people actually do. Americans tend to over-report socially desirable behavior like voting and attending church and under-report socially undesirable behavior like drinking.
Gallup Poll Editor in Chief Frank Newport agrees that the halo effect factors into poll results. During a Gallup telephone survey of a random sampling of about 1,000 Americans nationwide, interviewers ask respondents questions such as, “In the last seven days, did you attend a church service, excluding weddings and funerals?” to determine their church-going habits.
“When people try to reconstruct their own behavior, particularly more frequently occurring on-and-off behavior, it is more difficult, especially in a telephone interview scenario,” Newport says. But he stands behind Gallup’s 40 percent figure: “I’ve been reviewing [U.S. church attendance] carefully,” he says. “No matter how we ask the question to people, we get roughly 40 percent of Americans who present themselves as regular church attendees.” He adds, however, that if you were to freeze the United States on any Sunday morning, you may find fewer than 40 percent of the country’s adults actually in churches.
“Although about 40 percent of Americans are regular church attendees, it doesn’t necessarily mean 40 percent are in church on any given Sunday,” he explains. “The most regular church attendee gets sick or sleeps in. The other reason may be people who tell us they go to church but are worshipping in non-traditional ways, such as small groups, people meeting in gyms or school libraries.”
In another study surveying the growth of U.S. Protestants, Marler and Hadaway discovered that while the majority of people they interviewed don’t belong to a local church, they still identify with their church roots. “Never mind the fact that they attend church less than 12 times a year,” Marler observes. “We estimate that 78 million Protestants are in that place. Ask most pastors what percentage of inactive members they have—they”ll say anything from 40–60 percent.”
Even with a broader definition of church attendance, classifying a regular attendee as someone who shows up at least three out of every eight Sundays, only 23–25 percent of Americans would fit this category. Olson notes that an additional million church attendees would increase the percentage from 17.7 percent to only 18 percent. “You”d have to find 80 million more people that churches forgot to count to get to 40 percent.”
Clearly, a disconnect between what Americans say and what they actually do has created a sense of a resilient church culture when, in fact, it may not exist.