Americans exaggerate how often they go to Church in answers to surveys, a new study finds. Catholics and Mainline Protestants exaggerated more than Evangelical Protestants and black Protestants.
Since religious service attendance is considered positive behavior, Americans tend to overestimate how often they attend services when answering questions from a live interviewer. This is called a "social desirability bias."
When answering questions for an anonymous internet survey, the social desirability bias does not have the same effect. So, Public Religion Research Institute compared the results of phone surveys to those of internet surveys.
Among all Americans, the study found a five percentage point difference for frequent church attendance. Thirty-six percent of Americans answered that they attend religious services weekly or more when asked by a live interviewer, compared to 31 percent who answered the same for an online questionnaire.
Among infrequent church goers, the difference was even greater, 13 percentage points. Only 30 percent of phone survey respondents answered that seldom or never attend religious services, compared to 43 percent of internet survey respondents.
The study found some differences among denominations, age groups, and regions of the country. Catholics, Mainline Protestants, young people, Northeasterners and Westerners were more prone to exaggeration than Evangelicals, Black Protestants, old people, and Southerners. The differences were also greater among those who said they seldom or never attend religious services than among those who said they attend weekly or more. No differences were found based upon race, ethnicity or gender.
Among those who said they seldom or never go to church, there was an 18 percentage point difference among Catholics (15 to 33 percent) and 17 percentage point difference among white Mainline Protestants (28 to 45 percent); compared to a 10 percentage point difference among black Protestants (14 to 24 percent) and eight percentage point difference among white Evangelicals (nine to 17 percent).
Oddly, those who are unaffiliated with any religion also demonstrated a social desirability bias toward attending church. When answering the online survey, 91 percent of the unaffiliated said they seldom or never attend religious services. For the phone survey, though, 73 percent said they seldom or never go to church, an 18 percentage point difference that is as high as Catholics.
"Even among Americans who claim no religious affiliation, the social pressure to report at least nominal religious engagement is still quite strong," said Daniel Cox, co-author of the study and PRRI's Director of Research. "Very few people are willing to admit that they never attend religious services, even though many of us don't."
The study, "I Know What You Did Last Sunday: Measuring Social Desirability Bias in Self-Reported Religious Behavior, Belief, and Belonging," was presented Saturday at the American Association of Public Opinion Research national meeting.
The phone survey had a random sample of 2,002 adults. The online survey had a national probability sample of 2,317 adults. Both surveys were conducted in 2013.