Religion and well-being in a church without a creed

Mental Health, Religion & Culture March 2007; 10(2): 109 126 Religion and well-being in a church without a creed MARTA ELLIOTT & R. DAVID HAYWARD University of Nevada, Reno, US Abstract This study of religion and well-being questions whether adherents of a religion without a specific creed may still derive well-being from their religious experience as do members of more orthodox religious institutions. Survey data from a Unitarian Universalist congregation were analyzed to test whether the relationship between religion and individual wellbeing in a non-christian, non-creedal religious institution is similar to that observed in mainstream Christian denominations. Hierarchical linear regression analysis indicates that spirituality, social support, and the centrality of Unitarian Universalism (UU) to personal identity predict the amount of strength and comfort the respondents derived from religion, after controlling for demographic characteristics. Interaction effects between education and theological coherence and between gender and spirituality were also found. These results support the hypothesis that members of non-doctrinal religious groups benefit from religion by means of similar processes as members of religious groups that espouse more orthodox belief with one exception: theological incoherence is actually more beneficial to well-being of the highly educated in this context than is theological coherence. Introduction: Religion and well-being in a church without a creed Membership in a religious group is, in a wide variety of circumstances, associated with increased levels of individual well-being (Pargament, 2002). But because religion is such a complex phenomenon, and both religiosity in general and religious group membership in particular can serve an array of interacting functions, it is not clear by what mechanism(s) its benefits are conferred. Most of the research in the area of religious group membership and well-being has been conducted in the context of mainstream Christian churches in the English-speaking world with fairly clear-cut belief systems (see Suhail & Haroon, 2004 for a rare exception), rendering it important to assess the relationship between church Correspondence: Marta Elliott, PhD, Department of Sociology/300, University of Nevada, Reno; Reno, NV, 89557. Tel: (775) 784 4884. Fax: (775) 784 1358. E-mail: melliott@unr.edu ISSN 1367-4676 print/issn 1469-9737 online ß 2007 Taylor & Francis DOI: 10.1080/13694670500386069

110 Marta Elliott & R. David Hayward membership and well-being in a non-christian religious context. The current study assesses the association of church experience on individual well-being among members of a non-denominational religious movement with no explicit creed known as Unitarian Universalism (UU). Most studies of religion and well-being focus on church affiliation in contrast to a lack thereof to demonstrate the benefits of religion to well-being (e.g. Strawbridge, Cohen, Shema, & Kaplan, 1997). However, a growing number of studies take into account other interacting aspects of religious belonging that affect well-being, such as the opportunity for spiritual experience (Ellison, 1991; Schwenka, 2000), the provision of social support, and the provision of a cognitive framework for responding to existential questions (Ellison, 1991). In this study, we explore variation among individual church members in terms of their experiences of spirituality, social support, and theological coherence to better understand variation in their well-being. The UU are a particularly interesting group to examine because they have explicitly rejected any form of theological orthodoxy or creed, but maintain a congregational social structure similar to that of most Christian denominations in America. Comparing the pattern of personal well-being within this group with those observed in existing studies focusing on mainstream churches will help to determine whether the positive effects of a religion are contingent upon its ability to foster a standard set of beliefs among group members, or whether a religion without a creed has similar functions for its adherents. Spirituality The spiritual experience associated with religious affiliation is one potential mechanism by which the religion enhances subjective well-being. Religious groups tend to foster spiritual experiences by encouraging activities such as prayer, meditation, and contemplation. These activities are in themselves beneficial (Greene & Hiebert, 1988; Kumar & Hussain, 2003; Poloma & Pendleton, 1989) and so contribute to the overall well-being of the religious affiliates. There is evidence that certain subsets of spirituality have important effects on well-being. Schwenka (2000), using a definition of mysticism as the spiritual experience of perceiving a unity among all things, found that mystical experience was associated with a greater life satisfaction. To the extent that a religious group encourages or focuses attention on such experience, its members would be expected to benefit. The act of independent prayer is another aspect of spirituality that is often suggested as a mechanism by which religious belonging confers benefits upon individuals. Correlations have been observed between the frequency of personal prayer and various measures of well-being, including positive affect, balanced affect, and the absence of negative affect (Ferris, 2002). Particularly pertinent to the conception of prayer as a spiritual experience, Poloma and Pendleton (1989) produced evidence that the benefits associated with prayer vary depending on the style of prayer that an individual practices. Abstract meditational prayer was positively related with existential elements of well-being

Religion and well-being without a creed 111 and colloquial or conversational prayer was associated with overall happiness, but neither ritualistic nor petitionary prayer was associated with such positive outcomes in terms of quality of life. This suggests that the effectiveness of prayer in improving the well-being may be partly mediated by the extent to which that prayer reflects a spiritual experience of a personal connection to the divine. Social support Social support is perhaps the most thoroughly researched mediator between religious affiliation and well-being. It has long been suggested that religious groups provide their members the opportunity to become part of a cohesive congregation-based social network that confers social benefits including facilitating the formation of friendships and providing help accessing community resources (Ellison, 1991). There is a significant body of evidence that integration into a religious community is associated with higher personal well-being. For example, Ferriss (2002) found that the frequency of attending religious services was the most important predictor of individuals happiness, when compared to denominational affiliation and espousal of particular religious beliefs. Greater integration into a religious group also appears to improve one s ability to use the group to enhance well-being through religious coping (Pargament, Tarakeshwar, Ellison, & Wulff, 2001). Even membership in religious clubs, which presumably offer less in the way of formal spiritual structure than congregations, seems to confer some of the same benefits in terms of well-being (Frankel & Hewitt, 1994), which suggests that social integration itself plays an important role. Additional research suggests that social relations in the setting of a religious institution are sometimes characterized by conflicts that can act to the detriment of individual well-being (Krause, Chatters, Meltzer, & Morgan, 2000). Feeling overly criticized and being overburdened by church-related demands are both negatively associated with well-being, especially for those in positions of leadership in the church (Krause, Ellison, & Wulff, 1999). Thus, the relationships one acquires through religious involvement are not always socially supportive, but may sometimes be unsupportive, resulting in the opposite effect on psychological well-being. Identity Although it is often left out of general models of religion and well-being, the degree to which one s personal identity is defined by religion may profoundly influence the benefits of religion. Identity theory predicts that the more valued or salient a particular identity is, the greater is the potential that value has to exert a positive or negative influence on one s well-being (Thoits, 1991). Thus, one would expect that those for whom religion is a central aspect of identity would experience the effects of religious group membership most strongly. Furrow, King and White (2004), for example, found that the degree to which

112 Marta Elliott & R. David Hayward adolescents personal identities were defined by religion was positively associated with their sense of fulfillment in life. Self-stereotyping may account for some of this effect (Burris & Jackson, 2000); the more one identifies as a prototypical member of a religious group, the more one attributes the positive aspects of membership to the self. Mael and Ashforth (2001) discussed the literature supporting a link between religious identity and self-esteem, but cautioned that individual well-being may come at the expense of collective well-being, as religious group identity has historically often been built on derogation of other groups. Theological coherence Belonging to a religious group can also serve the function of providing and reinforcing a coherent framework for understanding the world. There is an influential line of reasoning that suggests that religion enhances individual well-being by offering a consensual system of meaning that reduces the anxiety associated with existential uncertainty (Greenberg et al., 1990). Religious groups, according to this argument, promote their members well-being in part by enforcing a uniform theological structure. Thus, the extent to which one derives benefits in terms of reduced anxiety from one s religion may be mediated by one s personal sense of theological coherence. This basic position was supported by Ellison (1991), whose analysis of General Social Survey data suggested that existential certainty based on religious beliefs was the most important predictor of perceived quality of life among the numerous religious variables measured. Similarly, Francis and Kaldor (2002) found that belief in God was associated with positive affect, independent of the effects of church attendance, and Suhail and Haroon (2004) found that holding religious beliefs was a significant and independent predictor of life satisfaction among Pakistanis. There is, however, some suggestion that this propensity for belief to mediate well-being may not be universal to all religions, but rather may be conditioned by the theology of the group. For example, Cohen (2002) found that holding specific beliefs consistent with one s religious tradition was a better predictor of happiness among Christians than among Jews. This may suggest that endorsing a consensual system of existential meaning is an important mediator of well-being in groups with theological traditions emphasizing faith and orthodoxy, such as Christianity and Islam, but that other mediating functions are more important among religious groups such as Judaism that emphasize identity and practice. Religion, well-being and demographic characteristics Several demographic characteristics are known to be associated with overall well-being in predictable ways. Although overall well-being is not equivalent to well-being derived from one s religion, it is expected that the associations with overall well-being will tend to hold when measuring church-specific well-being, as we do in this study.

Religion and well-being without a creed 113 In terms of gender, women tend to have lower levels of well-being than men (Piccinelli & Wilkinson, 2000). Age tends to have a curvilinear relationship with well-being such that well-being increases until about age 45 and then begins its decline into old age (Mirowsky & Ross, 1992). People who are married tend to have greater well-being than people who are not (Mookherjee, 1997), whereas parents tend to have lower well-being than non-parents (MccLanahan & Adams, 1987). Finally, education bears a strong positive association with well-being (Ross & Van Willigen, 1997). Furthermore, the effects of these five demographic characteristics on well-being may vary as a function of spirituality, social relations, identity, and theological coherence. For example, there is a wealth of empirical evidence that ties gender to religion and spirituality. Researchers have consistently found, using a variety of measures, that women are more religious, more spiritual, attend religious services more frequently, and are more orthodox in their beliefs than men (Miller & Hoffman, 1995). Gender also has been observed to interact with various aspects of religion when it comes to predicting well-being. Religious attendance reduces mortality more dramatically for women than for men (House, Robbins, & Metzner, 1982; Strawbridge et al., 1997). Other research suggests that while women receive more church-based social support, men benefit more in terms of overall health from the support that they do get (Krause, Ellison, & Marcum, 2002). Age is associated with more religiosity (Barna, 1996) and more frequent church attendance (Gallup & Jones, 1989), although the latter effect is tempered by decreased attendance due to poor health and lack of access to transportation among some older adults (Ainley, Singleton, & Swigert, 1992). The negative effects of religious doubt become less strong as age increases (Krause, 1999). Married people may enjoy greater well-being associated with their religious involvement, especially when the couple attends together. Parents, however, may not derive the same degree of benefit if they are distracted from religious participation by the exigencies of bringing children to church. Religion, education, and well-being are related to one another in ways that are complex and not yet fully understood. Ellison (1991) found that, while education had an overall positive effect on life satisfaction, higher levels of education also reduced the positive effects of theological coherence. In terms of religious participation, there is evidence that education is positively associated with attendance within groups, but negatively associated with attendance between groups (Glaeser & Sacredote, 2002). That is, it is the most educated members of each religious group who tend to participate the most, but as these groups improve in average level of education, overall participation drops. Unitarian Universalism Unitarian Universalism (UU) is considered a religion of the free mind. In many ways, it exemplifies what might be called intellectual religion, in contrast to popular religion. The distinction between these two styles of belief is not

114 Marta Elliott & R. David Hayward unique to contemporary America; it reflects a remarkably consistent pattern across times and cultures. Whether in ancient Greece, imperial China, or renaissance Europe, the class of highly educated intellectuals has tended to hold religious convictions sharply at odds with the majority of the population (Stark, 2001). Popular religion is almost always centered around a God or gods conceived of as conscious, concerned, and active in human affairs, while intellectual religion tends to focus on abstractions and essences that preclude or minimize interaction between the human and the divine. From these very distinct God images, other important differences follow. Popular religion tends to be particularistic, claiming that access to the ultimate truth comes only through adherence to a single orthodoxy, while intellectual religion tends towards universalism, claiming that valid religious wisdom can be found in many traditions. Popular religion usually emphasizes divine agency concepts like grace and revelation in religious truth, while intellectual religion more often claims that human agency concepts like reason and meditation is sufficient to provide the ultimate answers. When broadly characterized, popular religion tends to encourage vigorous participation in social institutions and to inspire missionary zeal, whereas intellectual religion, since it makes no claims to have exclusive access to the truth, tends to discourage active proselytizing as a kind of illegitimate imperialism. These two styles of religion can coexist within a single religious tradition, and even within the same organization. But intellectual religion tends to be concentrated among members of wealthy and well-established organizations with stagnant or shrinking membership rolls, while popular religion flourishes in socially marginal but fast-growing groups (Stark & Finke, 1992). Originally two separate religions, Unitarianism and Universalism emerged in response to specific controversies within the Christian church, offering a home to skeptics who doubted the mainstream ideologies of their time. Unitarianism was born from the belief that God was not a trinity, i.e., God the Father, God the Son, God the Holy Ghost, but that there was only one God. Considered heretical in its time, Unitarians count Michael Sevetus, who was executed for his 1553 On The Errors of the Trinity as one of its many martyrs. Universalism is based on the belief that there is no hell, and that God will grant salvation to all people regardless of their deeds or sins. Both the Unitarian and Universalist movements have been champions of social justice since at least the 19th century, backing the abolition of slavery and women s rights among other issues. In 1961, the two organizations merged into a single association supportive of both religious tolerance and religious questioning. Although both movements originated out of Christianity, and many presentday members of UU churches identify as Christian, just as many do not, embracing a variety of personal theologies including humanism, theism, wiccan, Buddhism, and even atheism. The by-laws of the Unitarian Universalist Association (UUA) specify an explicit guarantee of freedom of belief. In fact, diversity, both in beliefs and among members, is actively encouraged. Nonetheless, there are seven unifying principles (see Appendix) that UU are called upon to affirm and promote. What differentiates the seven principles

Religion and well-being without a creed 115 from an official creed more typical of Christian churches, for example, is that the principles are not binding upon UU members. In sum, UU is a religion that supports freedom of religious belief and expression constrained only by a set of guiding principles to be pondered and applied as individual members find fit (Provost, 1992). Hypotheses Spirituality hypothesis: People who have more frequent experiences of spirituality tend to find greater strength or comfort in Unitarian Universalism. Social support hypothesis: People who receive more social support from the church will tend to find greater strength or comfort in Unitarian Universalism. Social conflict hypothesis: People who experience more social conflict in the church will tend to find less strength or comfort in Unitarian Universalism. Identity hypothesis: The more salient Unitarian Universalism is to one s identity, the more one will tend to find strength and comfort in Unitarian Universalism. Theological coherence hypothesis: The more coherent one s personal theology, the more one will tend to find strength and comfort in Unitarian Universalism. Data and methods Survey The design of this survey was based on a prototype supplied by the UUA, but was modified to meet the specific interests of a Unitarian Universalist Fellowship as they were searching for a minister, and of the first author who oversaw the development and administration of the survey. The survey was produced according to the total design method (Dillman, 2000) as an attractive booklet with a color cover, and was distributed along with an invitation to participate that was individually addressed and hand-signed by each member of the Search Committee. Careful attention to detail coupled with intensive follow-up efforts yielded an unusually high response rate for a survey of this nature. One-hundred and ninety-eight surveys were distributed by mail in late November of 2002 to 140 members and 58 friends (non-members). As of January 12, 2003, 72% (142) surveys were returned. Of these, 77% (105) were from members and 23% (32) were from friends. Thus, the overall response rate was 72%, the response rate for members was 75%, and the response rate for friends was 55%. Measurement Measurement of all variables is described in detail in Table I. Overall psychological well-being was not measured in the survey, but there was a direct measure of

116 Marta Elliott & R. David Hayward Table I. Variable descriptions. Variable Description Dependent variable Strength or comfort from religion Independent variables Female Age Married/partnered How much is Unitarian Universalism a source of strength or comfort to the respondent? 1 ¼ Not at all 2 ¼ A little bit 3 ¼ A fair amount 4 ¼ A great deal Respondent gender 0 ¼ Male 1 ¼ Female In what year was the respondent born? age ¼ 2002 year of birth What is the respondent s marital status? 0 ¼ Never married, separated, divorced, or widowed 1 ¼ Currently married/partnered Parent Does the respondent have any children under the age of 18 who are living at home? 0 ¼ No 1 ¼ Yes Education Member Length with congregation What is the respondent s education level? 1 ¼ Less than college degree 2 ¼ College degree 3 ¼ Graduate degree Is the respondent a friend or a member of the congregation? 0 ¼ Friend 1 ¼ Member How long has the respondent been a member/friend of the congregation? 1 ¼ Less than six months 2 ¼ Six months to 1 year 3 ¼ Over 1 year but less than 3 years 4 ¼ Over 3 years but less than 5 years 5 ¼ Over 5 years but less than 10 years 6 ¼ Over 10 years but less than 20 years 7 ¼ Over 20 years but less than 30 years 8 ¼ Over 30 years Positive social support Sum of the Emotional Support Received from Others subscale of the Religious Support Scale Short Form with modified response categories: 1. How often do people in our Fellowship make you feel loved and cared for? 2. How often do the people in our Fellowship listen to you talk about your personal problems and concerns? 1 ¼ Never 2 ¼ Rarely 3 ¼ Sometimes 4 ¼ Often (Continued )

Religion and well-being without a creed 117 Table I. Continued. Variable Negative social support Spirituality Identity Coherence of theologies Description Sum of the Negative Interaction subscale of the Religious Support Scale Short Form with modified response categories: 1. How often do the people in our Fellowship make too many demands on you? 2. How often are the people in our Fellowship critical of you and the things you do? 1 ¼ Never 2 ¼ Rarely 3 ¼ Sometimes 4 ¼ Often Sum of responses to 5 selected items from the Daily Spiritual Experiences Scale with modified response categories: 1. I experience a connection to all life. 2. During our Sunday service, I feel joy that lifts me out of my daily concerns. 3. I feel deep inner peace or harmony. 4. I feel a selfless caring for others. 5. I accept others even when they do things I think are wrong. 1 ¼ Never 2 ¼ Rarely 3 ¼ Sometimes 4 ¼ Often How central versus peripheral is Unitarian Universalism to the respondent s identity as an individual? 1 ¼ Highly peripheral 2 ¼ Somewhat peripheral 3 ¼ Neither central nor peripheral 4 ¼ Somewhat central 5 ¼ Highly central 1 ¼ coherent, 0 ¼ incoherent (complete description in text) the extent to which respondents felt they derived strength or comfort from UU. This single-item measure is treated as a proxy for well-being derived from the church and is the dependent variable for all reported analyses. Background variables include gender, age in years, marital and parenthood status, education, membership status in the congregation, and (among members only) length of membership. Social support was measured with a Short Form of the Religion Support Scale (National Institute on Aging/Fetzer Institute, 1999) consisting of four items, two indicative of positive social support and two of negative social interactions. We measured positive and negative social interactions as two two-item indexes because their alpha reliabilities were considerably higher when measured this way (0.55 and 0.57 versus 0.42 for the 4-item index). Furthermore, positive social support and negative social interactions are distinct phenomena expected to have countervailing effects on well-being.

118 Marta Elliott & R. David Hayward Spirituality was measured as an index with five items taken from the Daily Spiritual Experience Scale (Underwood, 2002) that capture spiritual experiences both within and outside of the church. Religious identity was measured with a single-item measure that asked respondents how central versus peripheral they considered UU to be to their identity. And finally, theological coherence was measured indirectly by classifying respondents self-identified theology(ies) into coherent versus incoherent. The theological preference item consisted of a list of 15 categories with instructions for respondents to check all that characterized their personal theologies. The response categories, with the percentage of respondents indicating each in parentheses, were Agnosticism (32.9%), Atheism (10.3%), Buddhism (18.5%), Christianity (24.0%), Deist (9.6%), Hinduism (6.8%), Humanism (43.2%), Islam (2.7%), Judaism (5.5%), Mysticism (14.4%), Naturalistic Theism (14.4%), Paganism (14.4%), Wicca (2.7%), None (9.6%), and Other (12.3%). Individual respondents selected between zero (6.2%) and 13 (1.4%) of these categories to describe their personal theologies. A plurality (45.9%) indicated only one of these categories, and most of the rest selected two (17.8%) or three (17.8%) theological preferences. Respondents who indicated only one personal theology, e.g., humanism, were classified as coherent, while those who indicated two or more of them were classified based on an a priori set of criteria identifying inconsistent pairs of theologies. Two theologies were deemed inconsistent if any of their generally accepted key points of belief were mutually exclusive. For example, Christianity entails a belief in the divinity of Christ, and since this claim is specifically rejected by both Judaism and Islam, Christianity was defined as inconsistent with each of these theologies. Similarly, since the Buddhist tradition is based on an explicit rejection of Hinduism s gods, as well as of its conception of the soul, those two theologies were classified as an inconsistent pair. Agnosticism, Humanism, and Mysticism were construed broadly as theological modifiers that could potentially be consistent with any religious tradition, but all of the remaining theological categories were construed to entail particular beliefs that could be at variance with one another. Any complex theological combination containing at least one inconsistent pair was also coded as incoherent. Thus, for example, a respondent identifying himself or herself as Christian, Humanist, and Mystic would be coded as theologically coherent, but one identifying as Muslim, Hindu, and Mystic would be considered theologically incoherent under this set of criteria. Analytic strategy Initially, we present descriptive statistics for all variables in the analysis. Then we proceed to multivariate analysis in which we regress our dependent variable, strength and comfort derived from UU, on all our independent variables in a series of equations. In the first equation, we regress strength and comfort on our background variables to test and control for variation according to gender, age, marital and parenthood status, education, and length of membership

Religion and well-being without a creed 119 in the Fellowship. In the second equation, we add indicators of religious experience and beliefs, including spirituality, positive and negative social support, identity as a Unitarian Universalist, and coherence of theology to test their associations with strength and comfort after holding constant demographic factors and length of membership. Finally, we test whether or not religious factors moderate the influence of background characteristics by entering statistical interaction terms, or multiplicative terms between two variables. A unique interaction term is constructed for each pair of background variables (7) and religious variables (5), resulting in 35 multiplicative or interaction terms. To test the effects of these 35 terms, we add them one at a time to Equation 2. Once we determine which interaction terms are significant when entered individually, we enter all the significant interaction terms together. Those that remain significant when entered as a set are presented in the final model, Equation 3. Results Descriptive statistics Table II presents minimum and maximum values, means and standard deviations for all variables included in the analyses. The mean value of 2.88 for the dependent variable indicates that on average, respondents reported receiving a fair amount of strength and comfort from Unitarian Universalism. Sixty percent of the sample was female and the average age was 58 years. Sixty-five percent were currently married or living with a partner and 22% had one or more children under the age of 18 living at home. The congregation, on average, was well educated with a mean level of education falling between a college and a graduate degree. Seventy-three percent of respondents were members and the average length of membership was somewhere between 3 and 10 years. The average Table II. Descriptive statistics (N ¼ 142). Variable Minimum value Maximum value Mean SD Dependent variable Strength or Comfort from Religion 1 4 2.88 0.86 Independent variables Female 0 1 0.60 0.49 Age 28 91 57.81 13.88 Married/Partnered 0 1 0.65 0.48 Parent 0 1 0.22 0.41 Education 1 3 2.49 0.71 Member 0 1 0.73 0.44 Length with congregation 1 8 4.77 2.01 Positive social support 2 8 6.25 1.22 Negative social support 2 7 3.97 1.24 Spirituality 5 20 15.56 2.23 Identity 1 5 3.23 1.17 Coherence of theologies 0 1 0.83 0.38

120 Marta Elliott & R. David Hayward value for positive social support, 6.25, indicates a typical response of sometimes to frequency of positive social support whereas the average value for negative social interaction of 3.97 suggests such experiences are more rare. The spirituality index averaged 15.56 indicating that on average, respondents sometimes experience spirituality as described by the index items. Religious identity averaged just above the neutral value for the scale slightly in favor of UU being central to one s identity. Finally, 83% of the sample was classified as having a coherent personal theology according to the classification rules described above. Multivariate analyses The results of the multivariate analyses are present in Table III. Equation 1 presents the regression of strength and comfort on background variables, indicating that none of the demographic indicators is associated with the dependent variable. While membership is positively associated in the bivariate case, only length of membership is positively associated in the multivariate case. Thus, regardless of gender, age, marital or parenthood status, or education, those who had been members of the church for long were more likely to report it as a source of strength or comfort to them than those who either were not members or had only been members for a short time. In other words, membership appears to confer psychological benefits over time regardless of one s demographic characteristics. Table III. Regression of finding Unitarian Universalism a source of strength and comfort on demographic and individual factors and relevant interaction effects (N ¼ 142). Equation 1 Equation 2 Equation 3 b b b Female 0.195 0.111 0.051 0.029 0.035 0.020 Age 0.004 0.070 0.008 0.120 0.006 0.102 Married/Partnered 0.093 0.051 0.046 0.026 0.027 0.015 Parent 0.008 0.004 0.135 0.065 0.129 0.062 Education 0.029 0.024 0.091 0.075 0.506* 0.417 Member 0.304 0.156 0.199 0.103 0.218 0.112 Length with congregation 0.078* 0.181 0.027 0.063 0.031 0.072 Positive social support 0.170*** 0.242 0.163** 0.232 Negative social support 0.026 0.038 0.021 0.030 Spirituality 0.090** 0.232 0.029 0.075 Identity 0.290*** 0.392 0.303*** 0.055 Coherence of theologies 0.289 0.126 0.308 0.134 Coherence education 0.478* 0.359 Spirituality female 0.116* 0.213 Constant 1.851 0.057 0.052 Adjusted R 2 0.057 0.331 0.365 S.E.E. 0.84 0.71 0.69 *p 0.05, **p 0.01, ***p 0.001.

Religion and well-being without a creed 121 Equation 2 tests the five hypotheses by adding indicators of positive social support, negative social interaction, spirituality, identity, and theological coherence to the variables entered in Equation 1. Length of membership became non-significant with the addition of these five new predictors, three of which were significantly associated with the dependent variable. More specifically, positive social support, spirituality, and religious identity were all three positively associated with deriving strength or comfort from UU in the multivariate case even when controlling for background characteristics and length of membership. Thus, the results support the spirituality hypothesis, the social support hypothesis, and the identity hypothesis. However, no support is found for the social conflict or the theological coherence hypotheses. The final stage of the analysis explored whether any of the hypothesized effects varied according to members background characteristics. For example, was the positive effect of social support on strength and comfort stronger for men than women, as has been shown in other studies (Krause et al., 2002), or was the influence of spirituality contingent on educational background? To explore these possibilities, we regressed interaction terms between each of our five key variables (positive social support, negative social interaction, spirituality, identity, and coherence of theology(ies) and each of our seven background characteristics (gender, age, marital and parenthood status, education, membership, and length of membership) on strength and comfort derived from Unitarian Universalism. Of the 35 possible statistical interactions, two were significant: education has a stronger positive effect on strength and comfort for those with an incoherent theology, and spirituality has a stronger positive effect on women s strength and comfort than on men s. Given the small number of interactions that were significant, these results could be due to chance alone. Nonetheless, given their consistency with previous research and with our theoretical model, we will explore their implications with caution. An examination of the complete effects of all components of the interaction terms aids in their interpretation. When coherent theology ¼ 1 and education ¼ 1 (less than college degree), the main effect of coherent theology, 0.308, may be added to the main effect of education, 0.506, to obtain 0.814, from which the interaction effect, 0.478, is subtracted, yielding a total positive effect of 0.336 for participants without a college degree who have a coherent theology. Among those with a college degree (education ¼ 2), the total effect becomes 0.364, and with a graduate degree, it is 0.392. In other words, the effect of a coherent theology is increasingly positive as education increases. When coherent theology ¼ 0, the importance of education to strength and comfort increases substantially, equaling 0.506 for those with less than a college degree, 1.012 for those with a college degree, and 1.518 for those with a graduate degree. Thus, it appears that while education has a direct positive effect on strength and comfort in general, its effect is magnified among those who do not subscribe to a coherent theology. Because spirituality is an interval rather than an ordinal measure (like education), we calculated total effects of spirituality on strength and comfort for men

122 Marta Elliott & R. David Hayward and women separately at fixed values at, above and below the mean level of spirituality in the sample. The average value for spirituality in the sample was 15.56 on a scale from 5 to 20. Thus, for women with an average value on spirituality, the total effect of spirituality on strength and comfort was (1)(0.035) þ (15.56)(0.029) þ (1)(15.56)(0.116) or 2.29 whereas for men, the total effect was (0)(0.035) þ (15.56)(0.029) þ (0)(15.56)(0.116) or 0.451. In other words, the effect of spirituality on strength and comfort was more than four times greater for women than for men who had average levels of spirituality. The same contrast was calculated for women and men who scored one standard deviation below the mean on spirituality resulting in contrasting total effects of 1.968 and 0.387 respectively. Similarly, at one standard deviation above the mean value of spirituality, the total effect for women was 2.615 while for men it was only 0.516. In sum, spirituality is positively associated with strength and comfort but its association is stronger for women than for men. Furthermore, the difference between women and men diminishes slightly at higher levels of spirituality. Discussion The purpose of this study was to test a model of religion and well-being in the context of a non-christian church that lacks an explicit creed. We wished to know if the common predictors of well-being derived from religion in more mainstream Christian churches would also apply in a different setting. Furthermore, given the demonstrated importance of theological coherence to well-being in Christian settings, we wanted to explore the complicated matter of just how well-being is to be found when theological coherence is not offered by the church itself. The results indicate that several predictors of well-being in Christian churches hold among members of a Unitarian Universalist church. First, self-reported experiences of spirituality are associated with greater strength and comfort found in the church, lending support to our spirituality hypothesis. Second, reports of positive social support from fellow church members is associated with greater strength and comfort, indicating that social support in a religious setting improves well-being even in the absence of a clear-cut belief system. Finally, consistent with identity theory, we found that those who view UU as more central to their identity tend to derive greater strength and comfort from it. Perhaps not surprisingly, theological coherence did not predict well-being derived from church as it has in other religious settings. In the Unitarian Universalist church, members are encouraged to develop their own personal theologies, and to continue to question and refine their beliefs as a life-long process. Those whose beliefs were not considered coherent by our classificatory system may have been lacking a well-conceived personal theology, or they may have developed a personal theology that went beyond the limits imposed by boundaries of the world s religions. The finding that those with an incoherent theology still tend to derive a fair amount of strength and comfort from their

Religion and well-being without a creed 123 religion provided they are well-educated supports this idea. That is, those who are well-educated may be better able to conceive of a complex personal theology that appears superficially incoherent but is in fact coherent at a higher level of reasoning. The interaction effect between theological coherence and level of education has some important theoretical implications. The results indicate that not only is an incoherent theology more beneficial to well-being than a coherent one, but that this effect increases with education. These findings contradict Ellison (1991) and may suggest that a different pattern generally prevails among members of the highest educational strata. Alternatively, there may be benefits to theological incoherence associated with the non-creedal context of the Unitarian Universalist group that are only fully exploited by those with the highest levels of education. The findings of the present study with respect to theological incoherence are far from conclusive; more precise measures that are better able to capture and compare the contents of individual systems of belief must be developed in order to properly study this relationship. The data used in this study were collected from a single Unitarian Universalist Fellowship and therefore do not represent UU let alone adherents of intellectual religion. Furthermore, the data were cross-sectional, forcing us to infer causal directionality from theory rather than observe it more directly. Nonetheless, the data have several strengths, including the high response rate, and the numerous questions that allowed us to measure our key concepts: well-being, spirituality, social support, religious identity, and theological coherence. Further research on religion and well-being could improve on our study in several ways. First, the population under study could be expanded to include other intellectual religions, such as the United Church of Christ. Second, the population could include individuals who subscribe to intellectual religion but do not attend any particular religious organization. Third, the measure of theological coherence could be developed with more in-depth questions regarding theological beliefs. Finally, adherents of intellectual religion could be compared to their counterparts who subscribe to more popular religions. Among the questions that could be addressed from such a comparison include the relative benefits of intellectual versus popular religion for well-being, and a comparison of the mechanisms that link religion to well-being between these two broad approaches to theology. What is clear from the present study is that intellectual religion does improve psychological well-being in some of the same ways as popular religion, via spirituality, social support, and an identity grounded in religious faith. References Ainley, S. C., Singleton, R., & Swigert, V. L. (1992). Aging and religious participation: Reconsidering the effects of health. Journal for the Scientific Study of Religion, 31, 175 188. Barna, G. (1996). The index of leading spiritual indicator. Dallas, TX: Word Publishing. Burris, C. T., & Jackson, L. M. (2000). Social identity and the true believer: Responses to threatened self-stereotypes among the intrinsically religious. British Journal of Social Psychology, 39, 257 278.

124 Marta Elliott & R. David Hayward Cohen, A. B. (2002). The importance of spirituality in well-being for Jews and Christians. Journal of Happiness Studies, 3, 287 310. Dillman, Don A. 2000. Mail and internet surveys: The tailored design method (2nd ed.). New York: Wiley & Sons, Inc. Ellison, C. G. (1991). Religious involvement and subjective well-being. Journal of Health and Social Behavior, 32, 80 99. Ferriss, A. L. (2002). Religion and the quality of life. Journal of Happiness Studies, 3, 199 215. Francis, L. J., & Kaldor, P. (2002). The relationship between psychological well-being and Christian faith and practice in an Australian population sample. Journal for the Scientific Study of Religion, 41, 179 185. Frankel, B. G., & Hewitt, W. E. (1994). Religion and well-being among Canadian university students: The role of faith groups on campus. Journal for the Scientific Study of Religion, 33, 62 74. Furrow, J. L., King, P. E., & White, K. (2004). Religion and positive youth development: Identity, meaning, and prosocial concerns. Applied Developmental Science, 8, 17 26. Gallup, G., & Jones, S. (1989). One hundred questions and answers: Religion in America. Princeton, NJ: Princeton Religious Research Center. Glaeser, E., & Sacredote, B. (2002). Education and religion. (Harvard Institute of Economics Research Paper No. 1913). Cambridge, MA: Harvard Institute of Economics. Greenberg, J., Pyszczynski, T., Solomon, S., Rosenblatt, A., Veeder, M., Kirkland, S., et al. (1990). Evidence for terror management theory II: The effects of mortality salience on reactions to those who threaten or bolster the cultural worldview. Journal of Personality and Social Psychology, 58, 303 318. Greene, Y. N., & Hiebert, B. (1988). A comparison of mindfulness meditation and cognitive self-observation. Canadian Journal of Counseling, 22, 25 34. House, J. S., Robbins, C., & Metzner, H. L. (1982). The association of social relationships and activities with mortality: Prospective evidence from the Tecumseh Community Health Study. American Journal of Epidemiology, 116, 123 140. Krause, N. (1999). Aging, religious doubt, and psychological well-being. Gerontologist, 39, 525 533. Krause, N., Chatters, L.M., Meltzer, T., & Morgan, D.L. (2000). Negative interaction in the church: Insights from focus groups with older adults. Review of Religious Research, 41, 510 533. Krause, N., Ellison, C. G., & Marcum, J. P. (2002). The effects of church-based emotional support on health: Do they vary by gender? Sociology of Religion, 63, 21 47. Krause, N., Ellison, C.G., & Wulff, K.M. (1999). Church-based emotional support, negative interaction, and psychological well-being: Findings from a national sample of Presbyterians. Journal for the Scientific Study of Religion, 37, 725 741. Kumar, K. G., & Hussain, A. M. (2003). Meditation: A harbinger of subjective well-being. Journal of Personality and Clinical Studies, 19, 93 102. Mael, F. A., & Ashforth, B. E. (2001). Identification in work, war, sports, and religion: Contrasting the benefits and risks. Journal for the Theory of Social Behaviour, 31, 197 222. McLanahan, S., & Adams, J. (1987). Parenthood and psychological well-being. Annual Review of Sociology, 13, 237 257. Miller, A. S., & Hoffman, J. P. (1995). Risk and religion: An explanation of gender differences in religiosity. Journal for the Scientific Study of Religion, 60, 419 437. Mirowsky, J., & Ross, C.E. (1992). Age and depression. Journal of Health and Social Behavior, 33, 187 205. Mookherjee, H.N. (1997). Marital status, gender, and perception of well-being. Journal of Social Psychology, 137, 95 105. National Institute on Aging/Fetzer Institute. (1999). Multidimensional measurement of religiousness/ spirituality for use in health research. Kalamazoo, MI: John E. Fetzer Institute. Pargament, K. I. (2002). The bitter and the sweet: An evaluation of the costs and benefits of religiousness. Psychological Inquiry, 13, 168 182. Pargament, K. I., Tarakeshwar, N., Ellison, C. G., & Wulff, K. M. (2001). Religious coping among the religious: The relationship between religious coping and well-being in a national sample of Presbyterian clergy, elders, and members. Journal for the Scientific Study of Religion, 40, 497 514.

Religion and well-being without a creed 125 Piccinelli, M., & Wilkinson, B. (2000). Gender differences in depression: Critical review. British Journal of Psychiatry, 177, 486 492. Poloma, M. M., & Pendleton, B. F. (1989). Exploring types of prayer and quality of life: A research note. Review of Religious Research, 31, 46 53. Provost, G. (1992). Brief history of Unitarian Universalism. (UUA Pamphlet Commission Publication #3005). Boston: UUA Publications. Ross, C.E., & Van Willigen, M. (1997). Education and the subjective quality of life. Journal of Health and Social Behavior, 38, 275 297. Schwenka, S. (2000). Mysticism as a predictor of subjective well-being. International Journal for the Psychology of Religion, 10, 259 270. Servetus, Michael. (1969). The Two Treatises of Servetus on the Trinity: On the Errors of the Trinity, seven books, A. D. MDXXXI: now tranlated into English by Earl Morse Wilbur. New York: Kraus Reprint. Stark, R. (2001). One true God: Historical consequences of monotheism. Princeton: Princeton University Press. Stark, R., & Finke, R. (1992). The churching of America, 1776 1990: Winners and losers in our religious economy. New Brunswick, NJ: Rutgers University Press. Strawbridge, W. J., Cohen, R. D., Shema, S. J., & Kaplan, G. A. (1997). Frequent attendance at religious services and mortality over 28 years. American Journal of Public Health, 87, 957 961. Suhail, K., & Haroon, R. C. (2004). Predictors of subjective well-being in an eastern Muslim culture. Journal of Social and Clinical Psychology, 23, 359 376. Thoits, P. A. (1991). On merging identity theory and stress research. Social Psychology Quarterly, 54, 101 112. Underwood, L. G. (2002). The Daily Spiritual Experience Scale: Development, theoretical description, reliability, exploratory factor analysis, and preliminary construct validity using health-related data. Annals of Behavioral Medicine, 24, 22 33. Appendix: Principles of the Unitarian Universalist Association Unitarian Universalist Association principles and purposes We, the member congregations of the Unitarian Universalist Association, covenant to affirm and promote. The inherent worth and dignity of every person;. Justice, equity and compassion in human relations;. Acceptance of one another and encouragement to spiritual growth in our congregations;. A free and responsible search for truth and meaning;. The right of conscience and the use of the democratic process within our congregations and in society at large;. The goal of world community with peace, liberty, and justice for all;. Respect for the interdependent web of all existence of which we are a part. The living tradition which we share draws from many sources:. Direct experience of that transcending mystery and wonder, affirmed in all cultures, which moves us to a renewal of the spirit and an openness to the forces which create and uphold life;. Words and deeds of prophetic women and men which challenge us to confront powers and structures of evil with justice, compassion, and the transforming power of love;. Wisdom from the world s religions which inspires us in our ethical and spiritual life;