Profiles of the Godless; Results from a survey of the nonreligious

Luke W. Galen

The nonreligious segment of the population is not only increasing but is also increasingly visible in the public square. Still, self-described religious believers constitute the vast majority of the American population, and so more attention has been paid by social scientists and survey researchers to distinctions such as religious denomination (say, evangelicals vs. mainline) or political leanings than to characteristics descriptive of a nonreligious orientation. This article describes a survey that represents a departure from previous studies that have tended to use broad categories, often lumping together the nonreligious into artificial groupings that are actually heterogeneous.

When surveys are conducted of the general population, the majority of which is religious, meaningful differences between distinct types of nonbelievers (say, secular humanist vs. atheist) have been neglected. Although commentators frequently speculate about distinctions among the godless, actual empirical studies of populations sufficiently large to permit reliable distinction between subtypes of nonreligious individuals are more difficult to come by. Given the stereotypical lens—if not manifestly negative outlook—with which the nonreligious are viewed, it is important to characterize the “varieties of nonreligious experience” to determine who precisely constitutes this growing demographic category.

In Hunsberger and Altemeyer’s Atheists: A Groundbreaking Study of America’s Nonbelievers (2006), an exception to this dearth of research, a survey of several atheist groups revealed that these active atheists tended to be highly educated, older males. Most had childhoods with little parental emphasis placed on religion, but a quarter of the sample had experienced at least a moderately religious childhood. One controversial interpretation of the study pertained to the distinction between atheists and agnostics in regards to personality traits such as dogmatism. The results indicated that atheists saw themselves as being less likely to give up their views in the future, and thus their outlook could be interpreted as being more rigid. However, in contrast to other samples of religious believers surveyed, the nonreligious, atheist, and agnostic were markedly less authoritarian and dogmatic.

In Amazing Conversions: Why Some Abandon Faith and Others Turn to Religion (1997), a study also conducted by Altemeier and Hunsberger, the authors examined the pathways by which nonbelievers reached their philosophical and religious conclusions. For example, some previously religious “apostates” experienced a mixture of costs and benefits as a result of their movement to irreligion. Although it was common for the previously religious to report positive feelings of intellectual autonomy, many reported negative interactions with families or the loss of emotional or familial support. Other surveys have similarly found that parental conflict is often associated with the jettisoning of familial religion. In How We Believe: The Search for God in an Age of Science (1999), author Michael Shermer describes a survey conducted using the readership of Skeptic magazine as well as a sample drawn from the general public. One characteristic predictive of lower religiosity (or at least lower fundamentalism) was the personality trait “Openness to experience,” which involves a high need for cognition, intellectual engagement, and xenophilia (interest in new experiences).

Although results from these studies offer valuable insights into characteristics associated with the varieties of nonbelief in God, the information gathered to this point has seldom included data on nonbelievers’ social relationships and mental well-being, much less any detailed breakdown of philosophical shadings. Regarding the latter, for example, the endorsement of philosophical beliefs (such as disbelief in God) can be compared with the terms that nonbelievers use to label themselves (such as atheist or humanist). Thus, there may be meaningful differences among those who choose such labels and those who may be de facto atheists and humanists but choose not to label themselves as such.


A survey instrument was prepared to probe some of the distinctions among the nonreligious so frequently neglected in previous studies. In order to gather preliminary data and to test the survey instrument, a request soliciting participation was sent to all subscribers of the Center For Inquiry/Michigan branch e-mail and group newsletter. For comparison purposes, we also approached two local churches that had sizable memberships. This was done in order to provide some range on survey instrument items such as belief in God as well as to support the testing of hypotheses regarding characteristics distinguishing between religious and nonreligious individuals residing in the same community. The CFI/Michigan and church samples were similar in size (n=333 and 325, respectively), age, and income. However several demographic distinctions stood out. The CFI/Michigan members differed from their churched counterparts by being predominantly male, more highly educated, more likely to be never married or cohabiting (although the “typical” CFI/Michigan member was married), and had fewer children living at home. In regard to metaphysical beliefs, not surprisingly, 95 percent of the church group reported being absolutely certain that God existed, with members distributed roughly equally among the self-labels of “religious,” “spiritual,” and “theistic.” Beliefs were more varied in the CFI/Michigan group. Although 48 percent were self-described atheists, the remainder was distributed among agnostics, humanists, spirituals, and “others” with the range of belief certainty much wider as well.

The measures of mental well-being showed few differences between the CFI/Michigan group and their churched counterparts. For example, reported life-satisfaction was well within the average range for both groups. One area of identifiable difference was that the churched participants perceived themselves as having a greater degree of social support from their social network relative to the CFI/Michigan members. Our survey included a standard measure of personality, the “Big Five” scales as described by McCrae and Costa. Personality theorists have identified five major dimensions of personality—ways in which individuals differ from one another that lend themselves to measurement on linear scales, or continua. One of the more familiar of these dimensions is extroversion versus introversion; however, this trait did not differ between the groups. The dimension that showed the greatest distinction between religious and nonreligious was the previously mentioned “Openness to experience.” As was the case in Shermer’s survey, nonreligious individuals reported being more intellectually oriented and unconventional. Even controlling for the large differences between religious and nonreligious individuals in regard to education, gender, marriage, and child-rearing, openness still was the strongest predictor of both lower religious belief and membership in CFI/Michigan as opposed to the churches. Another personality dimension that distinguished the religious from the nonreligious was “agreeableness” (a quality of being amiable or nonconfrontational as opposed to skeptical of others). The church sample was higher in agreeableness. Following the successful pilot study, it was determined that the survey instrume
nt, with minimal modification, could be used with a larger nonreligious sample.


In contrast to the goal of general population surveys (such as those conducted by Gallup or Harris), which is to obtain a representative estimate of the proportion of individuals within a population, the present study was designed to examine characteristics within the nonreligious population (such as the social and personal qualities that characterize different subgroups of nonreligious individuals). An e-mail request was sent to the membership of Center for Inquiry/Transnational. Some of the respondents referred the researchers to other nonreligious groups and individuals via a “snowball sampling” method. A total of 5,831 individuals completed the entire online form. Eighty-three percent of our sample resided in the United States An additional 8 percent were from Canada; 2 percent each were from the United Kingdom and Australia, and the remainder were from other countries.

It must be stated at the outset that this sample is not necessarily representative of all nonreligious individuals; rather, it is likely to be skewed toward those who are actively involved in secular-related issues (such as readers of CFI-affiliated publications or members of national and local nonreligious groups). However, the sample shows characteristics similar to most other surveys of the nonreligious in the literature. Among these characteristics, the demographic factor that most distinguished the nonreligious from the U.S. population as a whole was a high level of education. Forty-one percent of our respondents had a master’s, doctorate, or professional degree, and 31 percent reported earning more than $100,000 per year. Our sample was 74 percent male, 53 percent married, with an average age of forty-eight years. Therefore, the demographic characteristics indicate a stable and educated sample characterized by high socio-economic status.

Regarding religious background, the nonreligious have been raised in a wide range of childhood environments. At one end of this continuum, 15 percent reported having grown up in a household where religion was either mildly or not at all emphasized (only a small proportion reported that religion was actively discouraged); at the other end, 35 percent reported being raised with a strong or very strong religious emphasis. Also relevant is that those from the latter, high-religiosity background were more likely (relative to those from other religious backgrounds) to report that they have poorer relationships with family. An interesting curvilinear effect involved age and childhood religion. Reports of growing up with greater childhood religion were highest for the cohort currently in their sixties (i.e., born 1938–1948), and lowest for those currently in their twenties and those in their eighties. Thus, a greater proportion of those who lived their formative teen and young-adult years in the postwar 1950s reported having a higher level of religion in their household at that time, whereas the nonreligious who grew up either before the post–World War II period or who were later “Generation X’ers” reported less religious childhoods.


Although numerous articles and columns in Free Inquiry have explored the philosophical differences between belief labels—or the social, political, and pragmatic significance of, say, aligning with “humanism” as opposed to “atheism”—our survey allowed a purely empirical examination of those who choose various self-designations. Respondents were allowed to endorse multiple religious and philosophical views or labels (such as “spiritual,” “agnostic,” and “humanistic”), but they were also asked to choose the single term that best described themselves. This self-identification term served as a basis for categorization. Despite the option of selecting among a dozen labels such as “deist” or “polytheist,” the overwhelming majority of respondents were divided amongst four preferred labels: 57 percent atheist, 24 percent humanist, 10 percent agnostic, and 2 percent spiritual.

Interesting distinctions appear when examining the difference between an inclusive selection (which is to say, when respondents were allowed to select more than one label) versus when they were asked to set all others aside to choose the most descriptive single label. For example, although 9 percent of the sample chose “spiritual” among multiple labels, when asked to pick a single self-identification, only 2 percent chose “spiritual.” This large proportional reduction indicates that far fewer chose spiritual as their sole label than were willing to include it among other labels. The label “agnostic” was similarly “jettisoned” by a relatively high proportion of individuals. In fact, many respondents appear to use “agnostic” and “atheist” interchangeably; among those who selected “agnostic” as one of their multiple labels, they evenly split between “atheist” and “agnostic” when choosing a sole identification label. It therefore appears that “agnostic” is used alongside other labels but frequently discarded when push comes to shove. “Humanist” seems to be a popular secondary label and contrasts in that regard to “atheist.” For example, around two-thirds of self-described humanists also consider themselves atheists; half of both atheists and agnostics also consider themselves humanists. However, the “supplemental” nature of humanism is evident in that, whereas two-thirds of the sample included “humanist” among their multiple self-identifications, only a quarter chose that as their sole label.

The nonreligious often debate distinctions such as between positive atheists (who actively deny the existence of God) and negative atheists (who claim only that no evidence for God exists) or whether the latter should really be deemed a form of agnosticism. Our data suggest that, perhaps out of a wish to avoid appearing as “dogmatic atheists,” many individuals actually use the terms agnostic and atheist interchangeably while a large proportion are essentially de facto atheists. Contrast the attrition from “spiritual,” “agnostic,” and “humanist” when reverting to a single label with the three-quarters who included “atheist” as one of their multiple self-identifications; 57 percent of the latter retained “atheist” as their as sole label. In other words, those respondents who included “atheist” among other labels were most likely to end up retaining it when choosing one self-identification. Thus, atheist appears to be more of a “bridge-burning” term; those who define themselves as atheists are less likely to shed that term or to dilute it with other labels. This indicates that although humanist is one of many hats that nonreligious individuals wear, when push comes to shove most of these individuals are “really” atheists. This raises the question: are there individuals who are for all metaphysical intents and purposes either atheists or agnostics but do not label themselves as such? What characteristics distinguish those who otherwise metaphysically agree in unbelief but choose to describe themselves differently?


A comparison of respondents’ chosen belief labels to their responses on the philosophical-belief items allowed us to characterize respondents who were de facto atheists or agnostics but did not so self-label. Participants who responded that they ranged from at least “not sure” to “absolutely sure” that God did not exist and who also responded that they “did not believe in anything beyond the physical world” were labeled as de facto atheists/agnostics. This designation was cross-referenced with their self-identification. Most of those who met de facto atheist/agnostic criteria but did not self-label as such (and who could be termed atheist/agnostic deniers) tended to use other labels such as “spiritual” or “humanist.” When these individuals were compared to those who actually self-labeled as atheists/agnostics, several differences between these two groups emerged. The most salient distinction between the “admitters” and the “deniers” was age. Those nonbelievers who chose to self-label as “spiritual” and “humanist” were older (average of fifty-three and fifty-one, respectively) than those choosing “agnostic” and “atheist” (forty-nine and forty-seven, respectively). This would seem to indicate a cohort effect, such that the term atheist is becoming more common despite a shared de facto philosophical outlook with self-labeled humanists.


Perhaps not surprisingly, self-labeling appears to coincide with social identification; the philosophical views of self-labeled atheists and humanists take a more central place in their lives (as measured by selection of the descriptive phrase “I am emotionally invested in my philosophical views”) than agnostics and spirituals. One demographic distinguishing characteristic of spirituals is sex composition. Even though women represented only a fourth of the total sample, they represented one-half of spirituals. Spirituals also reported having more close social confidants. The number of both family and nonfamily confidants reported by spirituals was higher than the other three belief labels. (The greater number of confidants of the spirituals is not attributable to the higher proportion of women in that group; the analysis still displayed significant differences when limited to males-only.) However, spirituals reported lower satisfaction with their lives than those with other belief labels.

In regard to personality, the four main types of self-labeled nonbelievers differed not in openness to experience (which is more of a distinguishing characteristic between believers and the nonreligious) but on two other Big Five personality dimensions: the aforementioned agreeableness and neuroticism (that is, negative emotionality). In personality terms, spirituals are more agreeable than are humanists, who are in turn more agreeable than atheists and agnostics. However, spirituals also report more negative emotionality (i.e., are less emotionally stable) than atheists and humanists, who are relatively more stable. Agreeableness does not reflect the preference for social contact per se (as does extroversion), as there were no significant differences among the belief labels in sheer number of social contacts. However, the agreeableness differences may indicate a certain willingness among the spirituals to try to “get along with” or trust others (as mentioned above, they report having more close confidants) in contrast to the atheists, who display relatively greater willingness to go against the social grain. One can debate whether or not this confirms or disconfirms the “angry loner atheist” stereotype. It must be emphasized that the average number of close personal contacts reported by all groups in the sample was relatively high compared with other general population data. These personality findings appear to indicate that the atheists place less emphasis, relative to humanists and spirituals, on pleasing or trusting others.

One argument frequently made by social psychologist D.G. Myers, among others, using general population samples is that psychological and emotional well-being and life satisfaction are higher in proportion to greater religious belief. The problem with most research using such samples is that, due to smaller numbers on the low end of the belief continuum, the unsure or weakly religious (say, those reporting sporadic church attendance or weak belief in God) are often obscured by, or lumped together with, the completely nonreligious. Similarly, lack of church attendance is often equated with lack of religious belief, when in fact many believers do not attend religious services (a factor likely to be linked to lower mental health). This has often resulted in findings that appear to demonstrate a linear increase in psychological health with increasing religiosity. This problem can be addressed by using more substantial numbers of complete nonbelievers in addition to those merely at the low end of the religious belief range.

The relationship between certainty of beliefs and emotional well-being in our nonbeliever sample was a mirror image of general population studies. In overwhelmingly religious samples, certainty and confidence in one’s beliefs tends to be related to characteristics of emotional health and a sense of purpose, whereas religious uncertainty and doubt often correlate with anxiety and depression. (See in particular Smith, McCullough, and Poll; see also Hunsberger, Pancer, Pratt, and Alisat.) When we distinguished strong varieties of nonbelief, such as atheism, from weaker nonbelief, a curvilinear relationship emerged (see Shaver, Lenauer, and Sadd). Those nonbelievers most confident in their nonbelief tended to be the most emotionally healthy, relative to the “fence sitters” who reported more negative emotions. Similarly, life satisfaction was lower among the spirituals relative to the other three belief labels. Therefore, having uncertainty regarding one’s religious views appears to be associated with relatively greater emotional instability. Taken together with the personality findings, confident nonbelievers (and apparently confident religious believers) are better situated emotionally, although the lower agreeableness indicates that strong nonbelievers appear to be somewhat less likely to acquiesce to or to trust others. These findings bring up interesting questions regarding the relative value priorities individuals may assign to personal belief certainty versus a need for social acceptance.


To summarize, relative to the religious or churched segment of the population, the nonreligious are distinguished both demographically (more likely to be male, highly educated, never married or cohabiting) and by their personality (more open to new experience and intellectually oriented, less agreeable). Although overall life satisfaction and social contact in our nonreligious sample was equivalent to the religious comparison group, the latter perceived a higher level of social support, possibly provided by their religious organizations. Among our large survey of the nonreligious, there was a range of philosophical beliefs: respondents included self-labeled atheists, agnostics, humanists, and spirituals. The label “atheist” appears to be becoming more common among younger individuals, suggesting that fewer nonreligious young people are choosing more tentative labels relative to older cohorts. Finally, in contrast to many general population studies that lump together those who are confident in their nonbelief with those who may be weakly religious, the present study allows the ability to distinguish degrees of nonbelief, yielding interesting results. Confident nonbelievers such as atheists were more emotionally well-adjusted relative to tentative nonbelievers; the latter, though, appear to place a greater emphasis on being agreeable to, and trusting of, others. The present study indicates that the common assumption of greater religiosity relating to greater happiness and satisfaction is overly simplistic. Many of the nonreligious, particularly those involved with an increasingly visible movement or community characterized by stronger varieties of nonbelief, are actually as well-adjusted and satisfied as the highly religious, with those uncertain of their beliefs showing more distress.

More research remains to be done, for example regarding the factors that differentiate individuals who are raised in a religious context who remain religious versus those who become nonreligious. Those with high openness to experience and lower agreeableness may not be satisfied with “tradition” and may seek out experiences that further reinforce irreligious tendencies. A less agreeable, more individualistic style may lead one to assert confidently a disbelief in socially required spiritual platitudes, with a resulting trade-off between greater emphasis on personal integrity but lower social acceptance. Many nonreligious individuals with such personality traits likely select life experiences throughout their educational and social development that result in further skepticism and increased certainty of nonbelief. These various pathways to irreligion will become increasingly relevant as the nonreligious continue to grow as a proportion of the population.


Thanks to Jeff Seaver, executive director and chair of the Center for Inquiry/Michigan, for his help in this project. For more information on the study or specific findings, go to the Web site

Further Reading

  • Altemeyer, R.A., and B.E. Hunsberger. Amazing Conversions: Why Some Abandon Faith and Others Turn to Religion. Amherst, New York: Prometheus Books, 1997.
  • Hunsberger, B.E., M. Pratt, and S.M. Pancer. “Religious Versus Nonreligious Socialization: Does Religious Background have Implications for Adjustment?” International Journal for the Psychology of Religion 11 (2001): 105–28. Canadian students with no religious background were not different in personal adjustment from those with religious backgrounds.
  • Hunsberger, B.E., S.M. Pancer, M. Pratt, and S. Alisat. “The Transition to University: Is Religion Related to Adjustment?” Research in the Social Scientific Study of Religion 7(1996): 181–99. Religious doubts and questioning were associated with anxiety.
  • Hunsberger, B.E., and R.A. Altemeyer. Atheists: A Groundbreaking Study of America’s Nonbelievers. Amherst, N.Y.: Prometheus Books, 2006.
  • McCrae, R.R., and P.T. Costa. Personality in Adulthood. New York: The Guilford Press, 1990.
  • Myers, D.G. The Funds, Friends, and Faith of Happy People. American Psychologist (2000): 55, 56–67.
  • Shaver, P., M. Lenauer, and S. Sadd, “Religiousness, Conversion, and Subjective Well-being: The ‘Healthy-minded’ Religion of Modern American Women.” American Journal of Psychiatry 137(1980): 1563–1568. A curvilinear relationship was found between belief and mental health.
  • Shermer, M. How We Believe: The Search for God in an Age of Science. New York: Freeman, 1999. A survey of Skeptic magazine readership found that, among other predictors, the nonreligious were characterized by higher education, familial conflict, and openness to experience.
  • Smith, T.B., M.E. McCullough, and J. Poll, “Religiousness and Depression: Evidence for a Main Effect and the Moderating Influence of Stressful Life Events.” Psychological Bulletin 129 (2003): 614–36. A modest but significant relationship existed between religiosity and lower depression, particularly when religion buffered stressful life events.

Luke W. Galen

Luke W. Galen is associate professor of psychology at Grand Valley State University in Michigan. Jeremy Beahan teaches world religions, aesthetics, and introduction to philosophy at Kendall College of Art and Design in Grand Rapids. He is the producer and cohost of the popular skeptical podcast Reasonable Doubts.

The nonreligious segment of the population is not only increasing but is also increasingly visible in the public square. Still, self-described religious believers constitute the vast majority of the American population, and so more attention has been paid by social scientists and survey researchers to distinctions such as religious denomination (say, evangelicals vs. mainline) or …

This article is available to subscribers only.
Subscribe now or log in to read this article.