The major development in polling about religion since mid-decade when Baylor 2006 appeared was the 2010 Pew Forum Poll on Religious Knowledge and the same organization’s survey of religious change in America.

Pew 2010 (The Religious Landscape Survey) had a few surprises: (1) that 1/4 of American adults have left the faith in which they were raised; (2) that the number of people who claim to be unaffiliated with any particular faith now stands at 16%, 1.6% of whom are atheists; 2.4 agnostic; and 12.1 uninterested in the question but “Nothing in particular”; (4) that for the first time in history America is on the verge of becoming a Protestant-minority country, with barely 51% of Americans being members of protestant denominations.

While Catholicism has experienced the greatest net losses (about a 7% drop), immigration from predominantly Catholic countries has kept Catholicism  at about 24% of the religious population.

Other survey highlights: Men are more likely than women to declare no religious affiliation (24% against 13%); Muslims and Mormons have the largest families; and 50% of Hindus, 34% of Jews and 25% of Buddhists have received postgraduate education.  The religion with the lowest retention rate, at 37%, is the Jehovah’s Witnesses.

When it comes to religious literacy, Americans are “challenged” in a number of ways:

Fewer than half (47%) knew that the Dalai Lama is Buddhist, only 27% knew that Islam is the dominant faith in Indonesia, the country with the world’s largest Muslim population.  Less than 25% knew that it is permissible to use the Bible as literature in a classroom, and (the shocker) 45% of Catholics surveyed could not answer a question correctly about their Church’s belief about the Eucharist (i.e., the doctrine of the “real presence“).

In the rankings, atheists took away the trophy with 20.9 of 32 questions answered correctly, followed by Jews and Mormons.  Catholics and Mainline Protestants were tied with a whopping 50% (do I see a big red F?) of correct responses.

Balanced against the fact that America remains the largest and busiest religious restaurant in the world, you’d expect more people would want to know a little more about the menu.  As always, however, America is full of surprises!

Following is a review of Baylor 2006, which originally appeared in Free Inquiry magazine,

I am an enormous fan of religion polls. I used to imagine pollsters as sleuths in trench coats, pulling palm-sized ring notebooks from their pockets and asking distracted bystanders whether God gives meaning and purpose to their lives-or something else, like maybe a good Caesar salad. We all know that in the Land of the Free, God gives meaning and purpose to around 90 percent of American lives, or at least that is the percentage advanced by the Baylor Religion Survey (September 2006) as the number who say they believe in God.

Recent polls conducted (infrequently) in the United Kingdom suggest that only 23 percent of Britain’s brood believe in God, and only 7 percent believe that the Bible is the word of God. That makes the American religious scene endlessly fascinating and completely confusing. So put away your ideas about pollsters in trench coats (I know I have) and ask yourself, “Who is asking the questions?” and “What do the answers really tell us about God in America?”

Flash back to the unanointed first pollster of American religion, Alexis de Tocqueville, whose long essay, Democracy in America, is increasingly cited by religious conservatives as proof that religion, of a certain sort, has always been good for America. Tocqueville argued in his 1835 work that the first political institution of American democracy is religion. His thesis, to paraphrase liberally, went something like this: the premises of secular materialism do not sustain democracy but undermine it, while the premises of Judaism and Christianity lead to democracy, uplift it, carry it over its inherent weaknesses, and sustain it.

Impressive as this might sound, it is difficult to find another political thesis with so much wrong about it. To be blunt, de Tocqueville could not anticipate the rise of religious movements so wedded to ignorance and out of step with the fundamental principles of both aristocracy and enlightened (Jeffersonian) democracy that they would undercut government in a way that even the rawest and most unrefined democratic passions never could. Nonetheless, the French observer joins two ideas that have remained central to all political estimates of American religion: in the absence of high culture, which, according to him and other nineteenth-century European travelers, America lacked and would always lack, religion performs a “civilizing function.” It reduces passion. It imposes by doctrine or moral fiat what might be achieved by reason in more enlightened parts of the world.

Tocqueville,first critical observer of American religion

And how are we doing religiously almost two centuries after de Tocqueville put down his pen?

The Baylor Survey provides some interesting and perplexing answers. First, it comes from Baylor, the flagship of Southern Baptist theological conservatism, and was funded by the Templeton Foundation, whose stated purpose is “to encourage a fresh appreciation of the critical importance-for all peoples and cultures-of the moral and spiritual dimensions of life.” Embracing that pious assumption greedily, the research group consisted of seven team members, including Rodney Stark, a Baylor faculty member, whose studies of American religion in the 1960s provided a model for what the team claims “is the most extensive and sensitive study of religion ever conducted”–by which they mean 1,721 respondents, mixed-mode sampling (telephone and self-administered mail surveys), and plans for “additional waves . . . with rotating topical modules every other year.”

Indeed, the study is so “extensive” that cherry-picking topics is the only way to do it the injustice it deserves–so let me focus on the most specious parts of the effort, leaving aside those that are merely uninformative.

According to Baylor, 86.5 percent of evangelical Protestants have “no doubt that God exists” (compare: 74.8 percent of Catholics and 42.9 percent of Jews). The real interest here is the sizeable number (13.5 percent) of evangelicals who appear to entertain doubts about God’s existence–not normally a trait one associates with Bible believers and a discovery that poses serious questions about the doubt-index for related doctrines. For example, 94.4 percent of evangelicals believe that Jesus is the son of God, 11 percent higher than those who have no doubts about God’s existence (compare that with 84.9 percent of Catholics and a mysterious–or deaf–9.6 percent of Jews).

The prospect that Jesus was a fictional character, by the way, is appealing to 13.7 percent of those described as “unaffiliated” but to almost no one else.

Almost 50 percent (47.8) of evangelicals believe the Bible is literally true. Presumably, no one defined the term literally or the number might be different. Literally outside the evangelical tradition has been defined as “theologically,” “spiritually,” “verbally,” “historically,” and “indubitably,” and, without narrowing those choices, it is difficult to know how to gauge the relevance of this response. But there is one interesting aside: despite the emphasis on Bible study invoked in the afterglow of Vatican II, only 11.8 percent of Catholics see the Bible as “literally” true. Catholics, apparently, see only bingo and what the pope says as literally true. Some 82.3 percent of those classified as “unaffiliated” see it as a collection of ancient history and legend.

Confused by the surprisingly large number of evangelicals who do not seem to accept the doctrine of plenary literal inspiration (50 percent), or have no idea what the question meant, I turned to the issue of religious labels–which, in fact, should be placed before the beliefs-survey, as it defines the terms used in the assessment. But confusion is again at hand. First, we are told, in a footnote, that respondents were only required to answer “Yes” or “No” to each label, that categories were not mutually exclusive, and that therefore the reported percentages do not add up to 100 percent.

Fair enough, say I, we live in a world where many things are less than 100%.  But statistical confusion still reigns: 47.2 percent of respondents described themselves as “Bible-believing” when (a) only 17.6 percent described themselves as theologically conservative and (b) only 14.9 percent described themselves as evangelical. In the first place, Bible believing is a term with currency unique to religious conservatives (not many of those 11 percent of Catholic biblical literalists would choose it, for example) and the menu of choices beyond it-“born-again,” “moral majority,” “seeker,” “religious right,” “Fundamentalist,” “Charismatic,” “Pentecostal”–seems pulled from the same Baptist hat. If the point of the menu was to express the deep structure of Protestant religious conservatism, its effect is to sideline other forms of Christian and religious commitment that do not fit the assumptions of the surveyors: “Would you describe yourself as Born-again, Bible-believing, or a little of both?”

What Flavor God: Distant or Critical?

Baylor assures us that “religious affiliation does not exist in isolation from belief and behavior.” It is not clear whether this should be translated as cause and effect or as bacon and eggs. In a way, that is the least problematical aspect of this section of the survey. The more troublesome issue is the vivisection of God into four “types”: authoritarian (angers easily, punishes harshly); benevolent (angers slowly, forgives easily); critical (God kind of is and kind of isn’t interested in the world); or distant (like the deus otiosus of philosophy, God is either asleep at the wheel or on permanent sabbatical and basically happy with the way things turned out). This quadruplex deity (trinity plus one?) is expressed in various social and political behaviors: conservatives like an authoritarian God; Catholics and mainline Protestants tend toward the “distant” view. People along the eastern coast of the United States tend to believe in a critical God, and southerners in an authoritarian God. Among those who choose none of the above, atheists are said to account for 5.2 percent of the sample. (Of the 10.8 percent of respondents who claimed no religious affiliation, 40 percent are atheists).

Baylor’s September 2006 survey actually dampens the percentage of unaffiliated from a previous benchmark of 14 percent to just under 11 percent, offering this caveat: “Researchers have previously over-counted the religiously unaffiliated by 10 million Americans and may have overlooked as many Americans who are actually affiliated with Evangelical congregations and denominations.” One refrains from attributing this undercount, if real, to the slipshod way of identifying evangelicals, already discussed. But wait: why the correlation between “affiliation” and “Evangelical affiliation”? To put it more directly, why not ask the whole sample about the authority of the pope or the assumption of the Virgin? Answer: because Bible belief is considered “normative,” while these other beliefs are thought of as “particular.” Effect: Christianity is defined in terms of the beliefs of the Protestant majority, with the Bible at the center, a recipe for disaster in achieving a deep profile of religious belief in a complex religious (and irreligious) society.

The tacit assumption throughout is that American religion is primarily about deviations from the norm of white, conservative-evangelical Protestantism. The upshot of this, reinforced methodologically, is that unbelief can be treated as an aberration of the norm, rather like the under-analyzed 4 percent “Other” category that includes a dog’s breakfast consisting of Mormons, Hindus, Muslims, Greek Orthodox, and Unitarians.

The Parson Thwackum Factor

We were taught in high school that every good book review should include not a rehashing of the plot but an assessment of the work as a whole. Here is my assessment of Baylor 2006.

The study is deficient in uncountable ways but chiefly at a level that affects its claim to be “the most extensive survey of religion ever conducted.” Its evangelical bias is not just implied in the mechanics of the survey–the choice of topics to be surveyed and the framing of questions–but in the often-preposterous techniques surveyors employed to relate God to social and religious behaviors.

Is it surprising, for example, that 90 percent of those who believe in a fascist God want prayer in schools, while only 47 percent of those who believe in a “distant” god want the same thing? Indeed, anomalies in Table 10 alone (“The Four Gods and the Role of Government“) raise significant issues about whether the respondents understood which God on the ballot to vote for.  If the study is all but useless, it is still useful in one way: it calls attention to the recurrent inability of all such surveys to do justice to unbelieving and pluralist-religious minorities. There is something mildly cloying and a little revealing about the following setup on page 8 of the survey. To quote:

“Barely one in ten Americans is NOT [sic] affiliated with a congregation, denomination or other religious group. . . . Fewer than five percent of the US population claims a faith outside the Judeo-Christian mainstream. . . . [But] fully a third of Americans, roughly 100 million people, are Evangelical Protestants by affiliation.”

As Case Western professor Brent Plate mused after looking at Baylor 2006, “If I were a Jain and received this survey to complete, I might not be very interested in filling in the boxes and returning it.” How much less an atheist or secular humanist!

Baylor’s researchers share the view of Fielding’s Parson Thwackum, who responds to a question about his definition of religion as follows: “When I mention religion, I mean the Christian religion; and not only the Christian religion, but the Protestant religion; and not only the Protestant religion, but the Church of England.” Just substitute “Baptist Church” at the close of the last sentence: You get the idea.