The following article is from Free Inquiry magazine, Volume 19, Number 3.
Public opinion polls in the United States have repeatedly demonstrated that Americans are a highly religious people. One of the most widely cited indicators of this religiosity is the extremely high percentage of Americans who say they believe in God, as measured by the standard Gallup Poll question, "Do you believe in God, or a universal spirit?" According to the Gallup organization's most recent reading (December 1994), 96% of U.S. adults said they believed in God.  Furthermore, this figure appears to be exactly the same as it was 50 years earlier, when the question was first asked November 1944. The lowest level of belief in God, as documented by Gallup, occurred in 1947 and 1978, when it dipped to 94%; the highest in 1953-54, when it reached an almost perfect 99%.  Thus the impression has grown of Americans' uniform and unwavering belief in God, since these figures and others (such as Gallup's data showing 75% of Americans believing in life after death) have been reported time and again in academic journals, professional publications, and in the mass media. 
So pervasive has the impression of this religiosity in our society become that various scholars and folk psychologists have theorized there may be an evolutionary biological basis for religious belief  and that religion must therefore satisfy a universal human need for hope, comfort, and a sense of purpose in facing the inevitability of our mortality.  Such is the power of the religious impulse in American thought.
This impression of the universality of religious belief, however, becomes much less solid when we look at survey data from other developed nations, comparing findings on religious beliefs from two cross-national surveys conducted in 1991 and 1993 for the International Social Survey Program (ISSP), which is currently based at the National Opinion Research Center (NORC), University of Chicago.  The data from these surveys shows that the degree of religious belief is not nearly as widespread, persistent, and universal as it appears from the perspective of the American culture, suggesting theoretical explanations other than a simple biological one.
The figures in tables 1-7 from the 1991 ISSP survey provide a pretty good indication of just how variable religious beliefs can be when viewed with a crossnational lens. The certainty of belief in God, for example, varied substantially across nations. Americans, who appear to be among the most religious people in the sample-along with the Irish, the Polish, and the Filipinos - were twice as likely as Austrians, three times as likely as Norwegians, and nearly seven times as likely as East Germans to claim "I know God exists and I have no doubts about it" (Table 1). Though a majority of Americans (62.8%) felt certain of the existence of God, they expressed a greater degree and variety of agnostic beliefs (and nonbeliefs) with this NORC version of the question than has been captured by the standard Gallup question.  So, while the great majority of Americans say they believe in God, regardless of how the question is worded, there is some doubt about the validity of the frequently cited Gallup figure (95%) that is often taken as the empirical gospel on what Americans believe.
Much the same pattern is evident when we look at beliefs about life after death (Table 2). On this score Americans were more certain of a hereafter than anyone else (55%): twice as sure as people from The Netherlands or Great Britain, five times as confident as the Hungarians, and nine times as convinced of it as the East Germans. But once again, too, Americans expressed more uncertainty about the afterlife with this ISSP version of the question than with the standard Gallup question - "Do you believe there is a life after death?" - to which about 75% of Americans generally say "Yes." This should be still another reminder, if one is needed, that the wording of survey questions about even such core beliefs as the existence of God and the afterlife can make a significant difference in the conclusions a pollster would draw about the nature and distribution of religious belief systems.
The data in (Table 3) tell a similar tale: Americans were among the most likely of peoples to believe that "The Bible is the actual word of God and it is to be taken literally word for word" - three times more likely than the Norwegians and nearly five times more likely than the British. They were also the least likely of any people surveyed to believe that "The Bible is an ancient book of fables, legends, history, and moral precepts recorded by man" (14.6% ) - a belief that was much more common in other nations, such as Israel, Hungary, Russia, and Great Britain. These beliefs, particularly biblical literalism, as we will see, turn out to be among the most useful predictors of the persistence of a "pre-scientific" worldview in American society.
We are also among what some would call the most superstitious people in the developed world, if the figures in (Table 4) are any indication. Americans were more likely than any other people to say they definitely believed in "the devil" (45.4 %): more than twice as likely as the Italians, three times as likely as the Poles, about five times as likely as the West Germans, and more than ten times as likely as the Hungarians.
And on and on the story goes for our beliefs in "hell" (Table 5), "heaven" (Table 6), and "religious miracles" (Table 7). We are either first in the faith, or a very close second - with Ireland and Northern Ireland (especially the latter) being the nations most like ours. 
Indeed, when we average across the rankings in tables 1-7, the United States turns out to be the most religious nation (average ranking = 1.71), followed by Northern Ireland (2.43), the Philippines (3.29), Ireland (4.14), Poland (5.21), Italy (5.86), New Zealand (8.0), Israel (8.29), Austria (10.57), Norway (11.0), Great Britain (11.57), The Netherlands (11.86), West Germany (12.07), Russia (12.71), Slovenia (13.86), Hungary (14.13), and East Germany (16.29), respectively. Notice too that the other nations ranked toward the top have substantial Catholic populations, and those toward the bottom are all former Soviet bloc countries. But even European nations such as West Germany, The Netherlands, Great Britain, and Norway are rather different in their religious beliefs from the United States, though they might seem quite similar as developed countries in other ways (e.g., standard of living). This ranking of developed nations might best be thought of, then, as defining an underlying dimension of religious cultural fundamentalism vs. modern secularism.
It should be more than obvious from these crossnational comparisons that religious beliefs of the type measured in the ISSP vary considerably among developed nations and that there is probably no biological propensity per se to hold such beliefs. While there may well be an evolutionary psychological disposition of Homo sapiens toward anthropomorphic interpretations of the world,  as well as other cognitive biases, such dispositions must necessarily interact with the availability of religious (and nonreligious) interpretations of the world provided by various cultures. For religion, as E.O. Wilson reminds us, is essentially a form of tribalism, whether it takes the form of Christianity, Buddhism, Marxism, or other kinds of "isms."  So the content and prevalence of religious beliefs in America may say more about our national history and culture than any human universals, thus making sociocultural explanations all the more plausible, most of which have been developed by sociologists of religion.
Andrew Greeley,  among other sociologists of religion, has argued that the persistence of the belief in life after death and other religious beliefs and practices over time in the United States are the result of a greater "supply" of religious services in the competitive American "marketplace" which, unlike most European nations, has not had a history of an established church.  While he offers some suggestive, indirect evidence to support his supply side hypothesis from a cohort analysis of belief in an afterlife among various religious and ethnic groups in the United States, he does not indicate any way of directly measuring the variability in the "supply of religious services" across nations or over historical periods. So it is a rather difficult proposition to test, however plausible it might seem and however compelling it might be as compared to the standard "secularization" and "modernization" models.  Such models would predict an inevitable decline in religious beliefs over time in the United States - one that has yet to materialize in any noticeable way, with the possible exception of a decline in literal acceptance of the Bible among older birth cohorts  and a recent drop in church attendance (see below).
It is also difficult to see how the supply side model can explain, for example, the similarity of religious beliefs in Northern Ireland, the Philippines, Ireland, and Poland to that of the United States (or to each other), none of which are anywhere near as religiously heterogeneous and as "competitive" for customers as the United States. Ireland and Poland, for example, are nations characterized essentially by noncompetitive, religious monopolies. The larger explanation, then, must lie in the national history and culture, which generate the huge variances we find in these crossnational tables, variances that swamp by comparison any of the usual differences we typically observe between demographic subgroups (age, race, sex, education, etc.) within nations such as ours. And it is a tall theoretical order, indeed, to think of the relevant dimensions along which the various nations in tables 1-7 differ that might help explain the variance in their religiosity. Penetration of what Freud called the scientific "Weltanschauung," however, is one such theoretical candidate. 
A plausible explanation of these crossnational differences lies in the influence of education on religious beliefs, particularly the effects of scientific literacy in a society. The data from a recent survey of American scientists by Larson and Witham indicates that there is a substantial split between scientists and the general public on beliefs about human evolution when they have been asked the same Gallup Organization question.  Only 5% of American natural and physical scientists, for example, believed in the biblical creationist view, as compared to the typical 45-50% of the U.S. public.  A majority of the scientists (55%) endorsed the Darwinian, naturalist position versus just 10% of the American adult population, though a surprising number of them (40%) took the theistic evolutionist perspective - about the same percentage as in the U.S. public. Thus it seems plausible that much of the difference between American scientists and the general public results from the effects of scientific knowledge about human evolution, in particular, and socialization more generally into the scientific worldview. 
Differences in public knowledge of evolution should also be a fairly good predictor of differences in the religious worldview across nations. The 1991 ISSP did not include any measures of public scientific knowledge. The 1993 ISSP, however, did ask a number of questions about environmental and scientific knowledge in 21 nations around the world, including 16 of the same nations that had participated in the 1991 survey on religion. One of the 12 knowledge items that was asked (in a four-point scale: Definitely true, Probably true, Probably not true, Definitely not true) dealt directly with the subject of human evolution: "In your opinion, how true is this? ... Human beings developed from earlier species of animals."
Table 8  shows that Americans were ranked last: the least knowledgeable of any of the 21 nationalities surveyed about this basic scientific "fact" (note again the similarity of Northern Ireland and Poland to the United States), though there is admittedly a fine line here between belief and knowledge. Furthermore, as might be expected, when we correlate the ranking of nations by correct responses to this item (using the 16 nations common to the 1991 and 1993 ISSPs) with the corresponding rankings for the various religious belief items in tables 1-7, we find a sizable, inverse relationship between knowing the scientific fact of human evolution and beliefs in God.
Responses to the question about evolution can therefore be regarded as a rough indicator of the extent to which the scientific worldview has penetrated a given society and given causal impetus to a decline in the religious worldview. Tom Smith has also argued in his analysis of the evolution question that much of the difference in knowledge of human evolution between Americans and Europeans stems from the strength in recent years of the fundamentalist movement in our society, such that even the normally beneficial effects of higher education on evolutionary knowledge are significantly diluted among those identifying themselves with fundamentalist religious denominations.  Differences in the religious environments in America and Europe, he argues, thus produce differences in knowledge of the simple fact of human evolution. As one of the leaders of the "scientific creationists," Henry M. Morris, has described the struggle in America: "There are only two possible world views - evolutionism or creationism."  So the strength of the fundamentalist movement (and the new religious Right) may explain not only the stability of American beliefs about human evolution observed in the Gallup Poll over the past couple decades,  but also the apparent persistence of the religious worldview in American society more generally, thus offsetting the rising percentage of college-educated adults in the American population.
Greeley has also furnished evidence on the growth of fundamentalist denominations and the decline of mainline Protestants in the United States throughout this period.  But devising a direct test of this influence-of-fundamentalism hypothesis is more easily said than done as all the appropriate measures of relevant variables are often not contained in the same surveys that have been conducted over the past several decades. Furthermore, Inglehart has made a persuasive, contrary case that fundamentalism has actually declined in advanced industrial societies such as ours and that it represents a counter-reaction by religious minorities whose values and way of life are threatened by the secular trends in modern cultures. Among other evidence, he points to the decline in church attendance and a decrease in the percentages saying that "God is important in their lives," from 1981 to 1990, in the United States and in other developed nations.  The most recent data from George Gallup's Princeton Religious Research Center (March 1997) would to seem to confirm Inglehart's predictions, showing that weekly church attendance in the United States (as of 1996) has dropped to the lowest point in nearly six decades: 38% (the all-time highs were in 1955 and 1958 when it reached 49%). So we may, if Inglehart is on the right theoretical track, just now be witnessing the incipient decline of the religious worldview in America, one that will become more and more evident in the surveys conducted in the coming decade of the new century.
To explain the persistence of the religious worldview in America we also need to consider the possibility that much of its public expression, with survey interviews being one such social situation, is driven by what Noelle-Neumann argues is the individual's fear of isolation that sets in motion a "spiral of silence."  As a people we are frequently reminded in the mass media and elsewhere that we are a "nation under God,"  that the vast majority of us (95% or so) believe in God - as Gallup tells us periodically - and that we are one of the most religious societies in the developed world. So it should seem quite plausible that many Americans may be reluctant to publicly express agnostic or atheistic beliefs for fear of giving offense to someone who may well be a member of that (purported) "vast majority" in our society, thereby incurring his or her disapproval. And this fear of interpersonal or social isolation should have the effect of maintaining the conformity of public beliefs so frequently observed in our public opinion polls as well as in many other social situations in life.
Not surprisingly perhaps, the religious worldview appears to be nearly universal in the United States, so much so that it has become theoretically plausible to some that there is an evolutionary biological basis for religious belief. The data from other developed nations in the ISSP  (1991 and 1993) presented here, however, tell us that the degree of religious belief is nowhere near as universal as it seems to be from the perspective of the American culture. The similarity of the United States to nations such as Ireland, Northern Ireland, the Philippines, and Poland and its striking dissimilarity to countries like The Netherlands, West Germany, Great Britain, and Norway - not to mention the antithetical, former Soviet bloc nations - suggest that any biological propensity toward animistic or anthropomorphic interpretations, for example, must necessarily interact with the cultural availability of such explanations, religious and otherwise (e.g., "New Age" spiritualism). As E. O. Wilson might put it, it is this gene-culture co-evolution that may account for the widespread persistence of supernatural and other transcendental worldviews.  But such worldviews are not inevitable, and are probably best thought of as incidental, cultural by-products of more basic evolved psychological mechanisms that have some adaptive function. An imitation mechanism producing conformity of religious (and other) beliefs and behavior within one's societal group, for example, would bestow both survival and reproductive benefits to their carriers. 
Socioculturally speaking, the persistence of the religious worldview in America may be due in significant measure, as some scholars have argued, to the strength of the cultural fundamentalist movement in our society in recent years that has succeeded in getting its message and agenda into the public schools, the mass media, and other social institutions. One powerful indicator of the success of this movement is the low level of scientific literacy about human evolution in American society as compared to other developed nations. But this movement may represent, as Inglehart and others have argued, the throes of a religious minority whose traditional values and way of life are deeply threatened by the relentless secularization of our culture and the steady growth of the scientific worldview. The persistence of the religious worldview in America may be, moreover, the result of a "spiral of silence" that surrounds the expression of agnosticism, atheism, and alternative "spiritual" beliefs-particularly as we say, in polite company.
It was Sigmund Freud who probably best argued that the scientific worldview would eventually replace the "religious stage" of superstitious (and other unscientific modes of) thinking in the evolution of human civilization.  In Austria, where he spent much of his career, in the land of Darwin and Huxley, where he spent his final years, and in much of the modern world, the religious worldview seems to have declined significantly, as he would have predicted. But the decline has yet to materialize in America where the scientific worldview has still, as Stephen Jay Gould puts it, to "Complete Darwin's Revolution." 
Tables Table 1. Percentage Saying "I know God exists and I have no doubts about it" by Nation in the 1991 International Social Survey Table 2. Percentage Saying They Definitely Believe in "Life after Death," by Nation in the 1991 International Social Survey Table 3. Percentage Saying They Definitely Believe "The Bible is the actual word of God and it is to be taken literally, word for word," by Nation in the 1991 International Social Survey Table 4. Percentage Saying They Definitely Believe in "The Devil," by Nation in the 1991 International Social Survey Table 5. Percentage Saying They Definitely Believe in "Hell," by Nation in the 1991 International Social Survey Table 6. Percentage Saying They Definitely Believe in "Heaven," by Nation in the 1991 International Social Survey Table 7. Percentage Saying They Definitely Believe in "Religious Miracles," by Nation in the 1991 International Social Survey Table 8. Ranking of 21 Nations on Knowledge Question about Human Evolution, 1993 International Social Survey
George Bishop is Professor of Political Science at the University of Cincinnati.