Why do Americans claim to be more religious than they are?

Two in five Americans say they regularly attend religious services. Upward of 90 percent of all Americans believe in God, pollsters report, and more than 70 percent have absolutely no doubt that God exists. The patron saint of Christmas, Americans insist, is the emaciated hero on the Cross, not the obese fellow in the overstuffed costume.

There is only one conclusion to draw from these numbers: Americans are significantly more religious than the citizens of other industrialized nations.

Except they are not.

Beyond the polls, social scientists have conducted more rigorous analyses of religious behavior. Rather than ask people how often they attend church, the better studies measure what people actually do. The results are surprising. Americans are hardly more religious than people living in other industrialized countries. Yet they consistently—and more or less uniquely—want others to believe they are more religious than they really are.

Religion in America seems tied up with questions of identity in ways that are not the case in other industrialized countries. When you ask Americans about their religious beliefs, it's like asking them whether they are good people, or asking whether they are patriots. They'll say yes, even if they cheated on their taxes, bilked Medicare for unnecessary services, and evaded the draft. Asking people how often they attend church elicits answers about their identity—who people think they are or feel they ought to be, rather than what they actually believe and do.

The better studies ascertain whether people attend church, not what they feel in their hearts. It's possible that many Americans are deeply religious but don't attend church (even as they claim they do). But if the data raise serious questions about self-reported church attendance, they ought to raise red flags about all aspects of self-reported religiosity. Besides, self-reported church attendance has been held up as proof that America has somehow resisted the secularizing trends that have swept other industrialized nations. What if those numbers are spectacularly wrong?

To the data: There was an obvious clue (in hindsight) that the survey numbers were hugely inflated. Even as pundits theorized about why Americans were so much more religious than Europeans, quiet voices on the ground asked how, if so many Americans were attending services, the pews of so many churches could be deserted.

"If Americans are going to church at the rate they report, the churches would be full on Sunday mornings and denominations would be growing," wrote C. Kirk Hadaway, now director of research at the Episcopal Church. (Hadaway's research has included evangelical congregations, which reported sharp growth in recent decades.)

Hadaway and his colleagues compared actual attendance counts with church members' reports about their attendance in 18 Catholic dioceses across the country and Protestants in a rural Ohio county. * They found that actual "church attendance rates for Protestants and Catholics are approximately one half" of what people reported.

A few years later, another study estimated how often Americans attended church by asking them to minutely document how they spent their time on Sundays. Without revealing that they were interested in religious practices, researchers Stanley Presser and Linda Stinson asked questions along these lines: "I would like to ask you about the things you did yesterday from midnight Saturday to midnight last night. Let's start with midnight Saturday. What were you doing? What time did you finish? Where were you? What did you do next?"