用户名: 密码:
成果Scholarsmore
您所在的位置:首页成果 》 论文
Religious Involvement and Depression: Evidence for Curvilinear and Stress-Moderating Effects
信息来源:Among YoungWomen in Rural China       发布日期:2013-10-23

Religious Involvement and Depression: Evidence for Curvilinear and Stress-Moderating Effects Among YoungWomen in Rural China

DEDONG WEI   (Department of Religious Studies Renmin University of China)
ERIC Y. LIU (Center for China Studies Buffalo State College State University of New York; and Duke University Medical Center)

 


Debates about whether the relationship between religiousness and depression is nonlinear have continued, but no definitive conclusions thus far have been drawn. Unlike most previous research in this area, which has been based on U.S. populations, this study focuses on a sample of 882 young rural Chinese women. Results from analyses reveal an inverse U-shaped relationship between intrinsic religiosity and depression. In addition, results show that intrinsic religiosity and religious activities exacerbate the deleterious effects of particular life events such as childbirth issues and marital conflicts. Implications of the findings for future studies on the subject are discussed.
Keywords: religiosity, depression, curvilinear relationship, stress moderating.


INTRODUCTION
Sociologists interested in the relationship between religion and mental health began by examining the effect of religious affiliation (Durkheim [1897] 1951). Although there have been meaningful findings about the role of religious affiliation (e.g., Breault 1986; Pescosolido and Georgianna 1989; Pope and Danigelis 1981), a growing trend in recent decades has gradually shifted the focus from religious affiliation onto religious involvement, a multifaceted concept that includes a broad array of religious orientations, attitudes, beliefs, devotional practices, and organizational activities (Ellison 1991; Koenig, McCullough, and Larson 2001; Smith,
McCullough, and Poll 2003). Thus far, scholars equipped with this more balanced approach to religion and mental health have reported a large body of empirical evidence consistently pointing to the salutary effects of religiosity, in spite of some notable exceptions (for a review, see Koenig, King, and Carson 2012).
That being said, researchers have yet to reach a general consensus regarding how religious involvement is related to psychological distress (Ellison et al. 2001). In particular, there have been significant, though scattered, debates and discussions involving the curvilinear as well as stress-buffering effect of religiosity on depression (Eliassen, Taylor, and Lloyd 2005; Ross 1990; Schnittker 2001; Shaver, Lenauer, and Sadd 1980). On one hand, most previous research shows an inverse U-shaped relationship between psychological distress and strength of religious Note: The following files were submitted by the author for peer review, but cannot be converted to PDF. You must view these files (e.g. movies) online.
Acknowledgments: The authors wish to thank Harold Koenig, Jie Zhang, Rong He, Christopher Ellison, Scott Schieman,
and Grace Ying Ning, as well as anonymous reviewers, for their helpful comments on early drafts of this article. This
study was supported by a grant from the National Natural Science Foundation of China (No. 70973132).
Correspondence should be addressed to Eric Y. Liu, Center for China Studies, State University of New York Buffalo State
College, 1300 Elmwood Ave., Buffalo, NY 14222, USA. E-mail: liuey@buffalostate.edu
Journal for the Scientific Study of Religion (2013) 52(2):349?67
C 
2013 The Society for the Scientific Study of Religion
350 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
belief (Ross 1990) and subjective religiosity (Eliassen, Taylor, and Lloyd 2005). On the other hand, some previous work discovered a U-shaped relationship between religious salience and depression (e.g., Schnittker 2001).
According to Eliassen, Taylor, and Lloyd (2005), discrepancies exist in the literature not only
because most of previous research employed single measures of religiousness and differential
control variables (e.g., stress exposure), but also because prior studies were based on data samples collected only from U.S. populations. To address weaknesses in the previous literature, this study examines the relationships between aspects of religious involvement梥uch as intrinsic religiosity, strength of religious belief, private practices, and religious attendance梐nd depression in a
Chinese context. The polytheistic Chinese religious system is different from the monotheistic Judeo-Christian-Islamic tradition in beliefs and practices, and it remains unclear whether the relationship between religiousness and depression is nonlinear in the Chinese setting. As will be discussed later, this study focuses on a sample of young, rural Chinese women. Our focus on
rural Chinese women is noteworthy for two substantive reasons. First, rates of mental disorders among young Chinese women are substantially higher than those of their male counterparts.
Second, there has been a vigorous religious growth in rural China. We use a newly developed
culture-specific Hoge Intrinsic Religiosity (IR) Scale (Liu and Koenig 2012) to test for three
specific types of effects: linear, curvilinear, and stressmoderating.
CURVILINEAR EFFECTS
Although multiple aspects of religiosity and their nonlinear relationships with psychological
distress have spurred increasing scholarly interest, there have been just a handful of theoretical and
empirical studies. Following Shaver, Lenauer, and Sadd (1980), Ross (1990) identified an inverse
U-shaped curvilinear relationship, indicating that individuals with high levels of commitment
to religious belief and those with no religious belief both reported less psychological distress
than those who had a moderate degree of commitment to their religious beliefs. Ross (1990)
also found that differences in belief systems across religious affiliations do not explain variation
in levels of psychological distress. Ross argued that the highest distress levels are associated
with persons who belong to a religion not out of voluntary choice, but out of indifference. These findings are broadly consistent with the view that religion provides cognitive, personal, and social resources that ultimately benefit psychological well-being (Antonovsky 1987; Ellison et al. 2001;
Hackney and Sanders 2003; Idler 1987; Petersen and Roy 1985; Pollner 1989; Veroff, Douvan, and Kulka 1981; Watson et al. 1988). More than a decade later, two empirical studies emerged with inconsistent findings. First, like Ross (1990), Eliassen, Taylor, and Lloyd (2005) observed an inverse U-shaped curvilinear relationship between religiosity and psychological distress. Their study, which used a measure of subjective religiosity, found that moderately religious respondents reported the highest levels of depressive symptoms, controlling for stress exposure. The study also indicated that the nonlinear association appears only for female study subjects.
In sharp contrast, however, Schnittker抯 (2001) examination of religious salience detected
the opposite U-shaped effect: very low or very high levels of religious salience were linked with
more depressive symptoms, adjusting for stress exposure. One possible explanation is that the Ushaped
association was due partly to the effects of religion on self-perception (including personal
mastery and self-esteem). In many ways, this argument is in line with the perspective emphasizing
the pathological characteristics and harmful consequences of religiosity (Ellis 1986, 1988; Ross
1990; Shaver, Lenauer, and Sadd 1980; Spilka 1987). For example, if religious groups become
overly restrictive, returns of religious attendance will diminish (Ellis 1986; Pescosolido and
Georgianna 1989; Spilka 1987). Moreover, extreme strength of religious belief may undermine
mastery and self-esteem (McDonald, Coval, and Luckett 1983). For instance, Ellis (1980, 1988)
argued that religion suppresses self-directedness and self-acceptance, leading devout persons to
RELIGIOUS INVOLVEMENT AND DEPRESSION 351
be inflexible, closed, intolerant, and unchanging. Certain beliefs, such as divine omnipotence,
divine punishment, divine control, and original sin, may result in especially adverse mental health
outcomes (Branden 1994; Ellis 1962; Schieman et al. 2006), since these beliefs can cultivate feelings
of shame, guilt, and self-doubt (Ellison 1993, 1994) and can be related to spiritual struggles,
fear and uncertainty, and their associated risks for mental illnesses (Flannelly et al. 2009).
MODERATING EFFECTS
Previous research suggests that the negative relationship between religious involvement
and mental disorders is even stronger among individuals who experience stressful life events
(Ellison et al. 2001; Schnittker 2001; Smith, McCullough, and Poll 2003; Tabak and Mickelson
2009; Williams et al. 1991). Several mechanisms may explain this stress-buffering effect. The
existential coherence and clarity afforded by strong religious belief may be especially helpful
for persons during times of trouble; religions offer a 搗ocabulary of suffering?and a 搑epertoire
of methods for their relief?(Wilson 1992:31). Persons who hold strong religious beliefs may
perceive stressful life events as a result of God抯 will or divine plan. They may regard life
difficulties as special opportunities offered by the supernatural to achieve spiritual growth or
learning. They may also think that since God is in control of everything, adversities are not
as severe as they appear, and they may be instilled with a sense of meaning and hope when
faced with challenging life circumstances (Ellison et al. 2001; McIntosh, Silver, and Wortman 1993; Musick 2000). Devotional activities such as prayer and religious counsel may also serve as important stress buffers (Ellison and Levin 1998; Pollner 1989; Veroff, Douvan, and Kulka 1981; Watson et al. 1988). Individuals engaged in devotional activities may gain psychological benefits such as greater confidence or an enhanced sense that with God抯 mercy and direction, one can better face and handle life stresses (Ellison 1994; Idler 1995; Pargament 1997; Pargament et al. 1990; Schieman, Nguyen, and Elliott 2003). Religious attendance may also moderate
the negative consequences of life stressors. Group sanctions may prohibit unhealthy behaviors
(McLaughlin et al. 1997). Social integration and support among fellow believers may help relieve
distress during times of adversities (Ellison and Levin 1998). Participation in ritual activities
may enhance self-esteem (Carothers et al. 2005) and promote a sense of meaning or life purpose
(Steger and Frazier 2005). Recent research has indicated that religious involvement is especially
powerful for buffering health-related stress (Liu, Koenig, and Wei 2012; Mattlin, Wethington,
and Kessler 1990; Stack 1983).
Contrary to the stress-buffering perspective, some scholars speculate about stressexacerbating
effects of religious involvement on depressive symptoms (Brown, Ndubuisi, and
Gary 1990; Ellison et al. 2001; Sorenson, Grindstaff, and Turner 1995; Strawbridge et al. 1998).
According to this line of research, aspects of religious involvement intensify the harmful influences of life stressors (Ellison et al. 2001; Schieman 2011; Strawbridge et al. 1998). For example, under certain stressful circumstances (e.g., disease and sickness, death grief, family and marital crises, domestic violence, child abuse, challenges with one抯 children, caregiver pressures,
financial problems, and legal disputes), some aspects of religious involvement may magnify the deleterious consequences of stressors on depression (Ellison 1994). Belief systems, religious values and worldviews, and social norms within religious communities may result in sources of negative inner feelings (e.g., guilt and shame), social stigmatization, disapproval, or denunciation, and withdrawal of support from fellow believers. Indeed, it has been noted that certain religious
values and practices play a potent role in exacerbating the negative consequences of familial problems (Ellison et al. 2001).
LIMITATIONS OF THE EXTANT LITERATURE
As discussed earlier, previous findings about the curvilinear and moderating effects of religiosity
on depressive symptoms are far from conclusive. Little, if any, previous research has
352 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
looked into the curvilinear relationship of depression to an important aspect of religiousness:
religious orientation. However, the effect of religious orientation on depression is worth studying
because, unlike other religious measures (e.g., religious beliefs, attitudes, practices, and affiliations),
religious orientation is a distinctly different dimension of religiousness (Allport and Ross
1967). The notion of religious orientations 揷aptures, in efficient form, some of the common pathways
people take and destinations they seek through religion?(Pargament 1997:59) The essence
of religious orientation derives from the involvement of the supernatural in the configuration of
means and ends (Pargament 1997). According to Allport (1966), for instance, individuals with an
intrinsic religious orientation see faith as a supreme value in its own right; by contrast, individuals
with an extrinsic religious orientation use religion for their own purposes (e.g., to obtain security
and solace, sociability, status, and self-justification). Intrinsically-oriented people center primarily
around God, whereas extrinsically-oriented people center around the self (Allport 1966).
To date, Allport抯 religious orientations are among the most heavily used frameworks in psychological
research on religion. Numerous previous studies have interpreted intrinsic religiosity
as deep religious devotion, distinguishing it from public rituals/behaviors or nominal religious
affiliation (Genia 1998; Lavric and Flere 2008; Tapanya, Nicki, and Jarusawad 1997; Watson
et al. 2002; Wink, Dillon, and Larson 2005). Indeed, empirical research confirms that compared
with a wide array of measures of religiousness, intrinsic religiosity assumes a unique role in
influencing depression. McCullough and Larson (1999), for instance, indicated that the negative
association between religiousness and depression became particularly strong when religiousness
was measured in terms of intrinsic religiosity; by comparison, the relationship was weaker when
religiousness was measured in terms of private religiousness, especially the strength with which
people hold particular religious beliefs. In their meta-analysis of 147 prior studies on religiousness
and depression, Smith, McCullough, and Poll (2003:628) identified only one statistically
significant difference between several categories of measures of religiousness: studies including
measures of intrinsic religiousness showed stronger negative correlations with symptoms of depression
than did studies using measures of religious attitudes and beliefs. More recent empirical
investigations and review research have, once again, verified this general pattern regarding religiousness
and depressive symptoms (Salsman and Carlson 2005;Wink, Dillon, and Larson 2005;
for an overview, also see Francis 2007).
Smith,McCullough, and Poll抯 meta-analysis (2003) also emphasizes the paucity of research
on the association between intrinsic religiousness and depression in Asian-American and Asian
populations.Nevertheless, there has been an emerging small literature examining the linear associations
between intrinsic religiousness and other mental health domains, such as worry/anxiety and
psychological well-being, in Asian populations (on Thailand, see Tapanya, Nicki, and Jarusawad
1997; on Japan, see Lavric and Flere 2008 and Roemer 2010; on Taiwan, see Liu, Schieman, and
Jang 2011). This literature has identified a key challenge in empirical research with a focus on
predominantly non-Judeo-Christian-Islamic societies: the original scales of intrinsic religiousness
揻unction well only within the U.S. sample?and thus are not directly applicable in other
cultures (Lavric and Flere 2008:172). To overcome this measurement issue, a culture-specific
intrinsic religiosity scale is essential (Tapanya, Nicki, and Jarusawad 1997). For instance, in their
study of Thai Buddhists, Tapanya, Nicki, and Jarusawad (1997) modified the Age-Universal I-E
Scale, replacing words such as church, God, and Bible with equivalent Buddhist terminology.
More recently, Liu and Koenig (2012) have adapted the Hoge IR Scale for mental health
research in China. Unlike the monotheistic Judeo-Christian-Islamic tradition, the traditional
Chinese religious system is strongly characterized by syncretism (Yang 1961). It combines
a wide array of supernatural beliefs (e.g., the Jade Emperor, Buddha, ancestor spirits, gods,
deities, and even ghosts), while melding practices drawn from a range of traditions including
Taoism, Confucianism, Buddhism, and popular ritual practices (e.g., praying, ancestor worship,
sitting meditation, cultivating Qi, chanting Buddha names, reciting mantras, making incense
or food offerings, and pilgrimages). Because of the absence of strict exclusivism, a Chinese
RELIGIOUS INVOLVEMENT AND DEPRESSION 353
person can select and mix religious beliefs and practices in accordance with day-to-day needs
or personal preferences. It has been suggested that the pursuit of health, high social status,
and good luck is a universal theme underlying the wide array of Chinese religious beliefs and
practices (Freedman 1974; Overmyer 1986). It is not unusual, for example, for a Chinese person
to make an incense offering to the God of Doctors (Baosheng Dadi) for better health in a
Taoist temple, worship the statue of Avaloketisvara Bodhisattva (Guanyin) as the giver of sons
at a Buddhist festival, and pray for family protection in front of the ancestor shrine at home.
Recent empirical research indicates that in the Chinese/Asian context, religiousness is often
positively associated with mental health outcomes (for a review, see Liu 2011; Liu, Schieman,
and Jang 2011). There is also evidence that belief in Chinese gods (e.g., Jade Emperor) buffers
health-related stress through positive emotions (Liu, Koenig, and Wei 2012). What, however,
are the nonlinear and stress-moderating effects of religiousness on depression in the Chinese
population? To fill this gap in the extant literature, we use the newly developed culture-specific
Hoge IR Scale (Liu and Koenig 2012), along with other religious measures, in an examination
of nonlinear and stress-moderating effects of intrinsic religiousness on depression in the Chinese
context.
HYPOTHESES
Recent research has investigated the possibility that the relationship between religiousness
and mental health may vary by gender (Eliassen, Taylor, and Lloyd 2005; Ellison et al. 2009; Idler
1987; Mirola 1999). As discussed earlier, the curvilinear relationship between religiousness and
psychological distress is more pronounced among women (Eliassen, Taylor, and Lloyd 2005).
Indeed, the fact that women are more religious than men is a common phenomenon across
societies and cultures, with only a few rare exceptions (Cornwall 2009; Freese 2004; Miller
and Stark 2002; Roth and Kroll 2007; Sullins 2006). This gender gap in religiousness also
characterizes Mainland China (Stark and Liu 2011). Religion has played an important role in
influencing the quality of Chinese women抯 lives. This is especially so in rural settings, where
religious involvement (especially festival attendance) never completely waned in 20th-century
China (Duara 1988). Especially during the reform era since the late 1970s, traditional cultural
norms have revived to the point where religious involvement is considered one of women抯 major
domestic responsibilities in rural China (Law 2005). Although the importance of religion for
women抯 life quality in China has raised significant research questions (Jaschok 2003), empirical
research on religiousness and depression in Chinese women has been astoundingly meager. A
few ethnographic studies have observed that during the reform era, Chinese women refused to
stop participating in temple fairs and festivals, showing that they felt helpless in the face of rapid
social changes and unfair treatment (Law 2005). Rural Chinese women regularly attend temple
services and hire shamans to perform rituals for matters such as giving birth to boys, curing
illness, and various marital and family issues (Law 2005). Thus, Law (2005) indicates that the
revival of religious practices serves the function of alleviating women抯 frustration with changes
in the domestic and economic spheres of rural village life. This perspective seems consistent
with previous research showing that supernatural beliefs serve the crucial function of mitigating
life difficulties confronted by women in traditional Chinese society (Yang 1970). We therefore
propose the following hypotheses:
H1: Religiousness will be associated with less depressive symptoms.
H2: The negative relationship between religiousness and depression will be inversely
U-shaped.
H3: Religiousness will buffer the harmful effects of stressful life events on depressive
symptoms.
354 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
DATA AND METHOD
Subjects
Participants in this study are women aged 18?4 living in the rural villages of Xinle,
Hebei Province, China. Participants were asked to complete a self-administered questionnaire in
February 2010. Before participation in the study, each participant read and signed a consent
form. Cluster sampling procedures with probability proportional to size (PPS) were applied to
recruit 1,200 participants from 10 randomly sampled villages of Xinle. The response rate was
86.6 percent (for more information about the data sample, see Liu and Koenig 2012).
The samplewas focused on rural young Chinesewomen, considering that ruralwomen梬ho
constitute the majority of Chinese female population梤epresent one of the most disadvantaged
social groups in China (Dai 1991). Rates of mental disorders among young women far exceed
those of their male counterparts in rural China (Pearson 1995; Zhang et al. 2010). In addition,
it has long been recognized that Hebei Province is an ideal place for studying the rural Chinese
population (Yan 2010), since for centuries Hebei has been a traditional agricultural province
with over 75 percent of its population classified as rural (National Bureau of Statistics of China
2001). More recently, China scholars have observed vigorous religious growth in rural villages
of Hebei Province (DuBois 2005; Fan 2003; Overmyer 2009) as religious revival has swept
throughout the country (Overmyer 2003; Stark and Liu 2011). Hebei provides a unique context
for examining how religious involvement may be associated with depression among rural young
Chinese women.
Dependent Variable
The Center for Epidemiologic Studies-Depression Scale (CES-D) is a self-report scale designed
to assess depression levels in the general population (Radloff 1977) that has been used in
previous U.S.-based literature to test for curvilinear and stress-moderating effects of religiousness
on depression (Eliassen, Taylor, and Lloyd 2005; Ellison et al. 2009; Schnittker 2001). The
scale contains 20 items measuring depressed mood, loneliness, psychomotor retardation, loss of
appetite, sleep problems, and personal feelings of guilt, restlessness, hopelessness, and failure
(Appendix S1). Respondents were asked to report how often during a typical week they experienced
these depressive symptoms. Responses were coded 1 = 搉ot at all,?2 = ? or 2 days a
week,?3 = ? or 4 days a week,?and 4 = 揳lmost every day.?The Chinese version of CES-D
has been shown to display good reliability and validity (Lin 1989; Zhang and Norvilitis 2002;
Zhang et al. 2003). The summed score of the unstandardized scale ranges from 0 to 74. Factor
analysis of the present data suggested a strong single factor, and the Cronbach抯 alpha for the scale
was .79.
Independent Variables
Intrinsic Religiosity
The Hoge IR Scale, originally derived from the Allport-Ross Scale (Allport and Ross 1967),
includes 10 statements that assess the intrinsic dimension of religiosity (Hoge 1972). Responses
are graded on a Likert-like scale of 1?; higher scores indicate greater agreement with the
statements. The Hoge IR Scale has been adapted for use in mental health studies in the Chinese
context (Liu andKoenig 2012).1 The Chinese version of the scale has two specific modifications in
1Previous research on intrinsic religiosity and psychological distress employed the scale developed by Gorsuch and
associates. Specifically, Lavric and Flere抯 (2008) research used the Revised Age Universal Religious Orientation Scale
RELIGIOUS INVOLVEMENT AND DEPRESSION 355
wording: (1) 揋od?was changed to 揼od(s), Buddha(s), Bodhisattvas, ancestral spirits, deities or
ghosts? and (2) 揃ible?was changed to 搑eligious texts?(see Appendices S2 and S3). Cronbach抯
alpha for the scale was .81.
Strength of Religious Belief
Following standard practice in prior research in the Chinese context (Liu, Schieman, and
Jang 2011), our scale of supernatural beliefs selects 10 questionnaire items that assess the extent
to which respondents believe in the following: 搆armic laws,?揾eavenly fate,?揧in-Yang and
five elements,?揃uddha and Bodhisattvas,?搒pirits and ghosts,?揳ncestral spirits,?搇uck,?搇ife
after death,?揳fter a person passes away, his/her spirit and soul still exist,?and 揾ell.?Response
choices are coded (1) 搉o, not at all,?(2) 搉o, not really,?(3) 搉ot sure,?(4) 搚es, somewhat,?and
(5) 搚es, very much.?The responses were summed over the items to create a supernatural beliefs
index. A factor analysis indicates that the items load quite highly on one underlying construct;
factor loadings range from .62 to .84. The alpha reliability coefficient was .92.
Religious/Spiritual Practices
Participants were asked: 揌ave you engaged in the following spiritual practices in the past
year??The individual items include: 搒itting in meditation,?揃uddhist chanting,?搑eciting
mantras,?損racticing Qi,?搑eading religious scriptures,?揼iving thanks, repenting, or praying
every day,?搒eeking shamans?help with health issues,?揹onation,?損racticing fortune telling
and fengshui,?搘orshiping ancestor spirits,?and 搈aking incense offering to Buddha.?Response
choices for these items are coded (0) 搉o?and (1) 搚es.?Following previous literature (Liu and
Mencken 2010), we calculated a count variable for religious/spiritual practices.
Religious Attendance
Respondents were asked: 揌ow often do you attend religious services or other activities at
temples and/or churches??Response choices are coded (1) 搉ever,?(2) 搒eldom,?(3) 搒everal
times a year, but less than once a month,?(4) 揳t least once every month,?(5) 搊nce every week,?br /> (6) 揳t least twice every week.?Because most respondents seldom or never attended religious
services, we follow prior research (Liu and Mencken 2010) in creating a dummy variable for
religious attendance (1 = yes, 0 = no). Appendix S4 provides a correlation matrix for the religion
variables.
Control Variables
We control for demographic and socioeconomic variables. Among these are age (in years),
Youth League or Chinese Communist Partymembership (1=yes, 0=no), and marital status (1=
married, 0 = separated, widowed, divorced, and single/never married). Education was originally
coded 1 = none/elementary school, 2 = middle school, 3 = high school, 4 = college, and 5 =
graduate school, but we include dummy variables for each education category to allow for the
emergence of potential nonlinear relationships between education level and depressive symptoms.
College and graduate school were merged into one category due to the extremely small number
of respondents who claimed to have undertaken graduate-level education. Elementary school
education or less was used as the contrasting category.
developed by Gorsuch and McPherson (1989), but only included three of the original items on intrinsic religiousness.
Tapanya, Nicki, and Jarusawad抯 (1997) study was built upon the Age Universal Religious Orientation Scale developed by
Gorsuch and Venable (1983). Terms such as church, God, and Bible were adapted to the Buddhist tradition in Thailand.
356 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Sense of Mastery
Sense of mastery is measured using the Pearlin Mastery Scale (Pearlin and Schooler 1978)
adapted for the Chinese context (Liu 2009). This scale includes seven items measuring one抯 sense
of mastery, two of which are positively coded. Response categories range from 1=strongly agree
to 4=strongly disagree. The score for the scalewas summed over the recoded ratings. Cronbach抯
alpha for the scale was .60.
Self-Esteem
Rosenberg抯 Self-Esteem Scale is a widely used 10-item instrument that gauges one抯 perceptions
and presentations of oneself (Rosenberg 1965). Scores on each item range from 1=strongly
disagree to 4 = strongly agree, indicating the extent to which respondents agree or disagree with
statements such as 揑 take a positive attitude toward myself,?揂ll in all, I am inclined to feel that
I am a failure,?and 揙n the whole, I am satisfied with myself.?Higher scores indicate greater
self-esteem. The score for the scale was summed over all 10 items. The alpha reliability of the
scale was .45, which is marginally acceptable (see in Schnittker 2001).
Stressful Life Events
A revised 64-item version of Paykel抯 Interview for Recent Life Events (IRLE) was implemented
to gauge the number of life events that had occurred in the past year. This adapted scale
also included 19 culture-specific items for China (Zhang and Liu 2012). The respondents were
asked whether they considered each of the life events to be positive or negative. Only the negative
life events were included in our analyses because of our focus on stress. Responses were coded
in two ways. First, the responses were coded into two count measures, with 1 = yes and 0 =
no for a single negative life event and multiple negative life events. We also created a series of
dummy variables by event type, with 1 = yes and 0 = no for family conflicts, marital conflicts,
work issues, childbirth problems, legal disputes, severe illness, and financial problems.
Social Support
The Multidimensional Scale of Perceived Social Support (MSPSS) contains three subscales,
each of which is comprised of four items assessing levels of perceived support from family,
friends, and significant others (Zimet et al. 1988). Respondents were asked to rate their level of
agreement with various statements on a scale of 1 = strongly disagree to 4 = strongly agree.
Higher scores indicate greater perceived social support. The MSPSS scale has been validated
in a wide range of samples, including women, adolescents, older adults, doctor trainees, and
patients (Kazarian and McCabe 1991; Stanley, Beck, and Zebb 1998; Zimet et al. 1990), and in
the Chinese context (Cheng and Chan 2004). It also has been shown to be relatively free of social
desirability bias (Dahlem, Zimet, and Walker 1991; Kazarian and McCabe 1991).
Analytic Strategy
We begin by presenting descriptive statistics computed to examine the distribution of the
data. We subsequently estimate ordinary least squares (OLS) regression models with depression
as the endogenous variable. The first OLS model estimates the main effects of religious variables
on depression, controlling for age, marital status, education, and Youth League or Chinese
Communist Party membership. A second model includes perceived social support, adjusting for
other variables in the previous model. Next, nonlinear effects of the religious variables are tested
in the model by including squared terms for intrinsic religiosity and strength of religious beliefs.2
2Quadratic terms may provide the most powerful tool for testing for potential nonlinearities (Schnittker 2001). An
alternative way of examining nonlinearities is to code responses on each of the scale items into a series of dummy
variables; supplementary analyses yielded results substantively similar to those based on the quadratic terms.
RELIGIOUS INVOLVEMENT AND DEPRESSION 357
Table 1: Descriptive statistics
Range Mean s.d. n (%)
Age 18?4 25.80 4.61
Marital status (married) 1 = yes 68
Party/league membership 1 = yes 48
Elementary school 1 = yes 1
Middle school 1 = yes 52
High school 1 = yes 29
College or graduate school 1 = yes 17
Intrinsic religiosity 1?6 36.43 7.34
Strength of religious belief 8?0 20.36 7.02
Religious practices 0? .74 1.01
Religious attendance 1 = yes 18
Self-esteem 0?7 27.28 3.71
Mastery 0?4 15.72 2.66
Social support 0?7 45.47 6.97
Stressful life events 0? 1.23 1.86
N = 882 (after deleting missing cases).
Keeping with standard practice (Schnittker 2001), subsequent models control for self-esteem
and personal mastery and continue to test for interactions between the religious measures and
measures of stressful life events.
RESULTS
Descriptive Statistics
Table 1 provides the descriptive statistics for the variables used in this study. The average
age of 26 indicates that most of the respondents have just entered adulthood. Moreover, most
of the rural young Chinese women are currently married, which is consistent with the previous
literature. Nearly half of the sample claimed to be members of the Youth League or the Chinese
Communist Party. Although the vast majority of the young women respondents have received
education above the elementary-school level, more than half of them did not have an opportunity
for high school education. Twenty-nine percent of the female respondents completed high school
but did not continue their education after graduation. Only 17 percent of the women went to
college, and even fewer entered graduate school. The sample also reported moderate levels
of self-esteem and mastery, as well as high levels of perceived social support. Most of the
young women said they had experienced no more than two stressful life events in the past
year.
Table 1 makes it clear that the young female respondents engaged in only a limited number of
private religious practices: about one in five participated in religious services or other activities at
temples and/or churches. Strength of religious belief is also modest among the women. However,
the average level of intrinsic religiosity is surprisingly high, suggesting that intrinsic religiosity
is particularly crucial for assessing the religiousness of rural young Chinese women.
Main Effects
OLS regression examines linear and nonlinear relationships between the religious variables
and depression, controlling for age, marital status, education, and Communist Party membership.
358 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Table 2: Depression regressed on main effects of selected independent variables
Model 1 Model 2 Model 3 Model 4
Age .043 .030 .040 .052
(.077) (.077) (.076) (.072)
Married ?.566* ?.439 ?.311 ?.120
(.787) (.785) (.775) (.738)
Youth League/party member ?.023 ?972 ?790 ?339
(.660) (.654) (.647) (.617)
Middle school ?.246 ?.668 ?.760 ?178
(2.388) (2.370) (2.340) (2.231)
High school ?868 ?.094 ?.253 .463
(2.438) (2.417) (2.386) (2.276)
College or graduate school ?.254 ?.151 ?.171 ?440
(2.504) (2.481) (2.449) (2.336)
Intrinsic religiosity ?210*** ?182*** .694*** .793***
(.041) (.041) (.194) (.184)
Intrinsic religiosity2 ?014*** ?014***
(.003) (.003)
Religious belief .054 .045 .345 .262
(.042) (.042) (.195) (.186)
Religious belief2 ?008 ?006
(.005) (.004)
Religious practice ?157 ?039 ?073 ?123
(.297) (.296) (.292) (.279)
Religious attendance .305 .166 .144 ?144
(.787) (.780) (.770) (.734)
Perceived social support ?175*** ?169*** ?019
(.041) (.041) (.043)
Self-esteem ?498***
(.088)
Mastery ?585***
(.108)
Intercept 35.451*** 43.014*** 27.115*** 39.033***
R2 .051 .071 .096 .185
*p < 0.10; ***p < .001.
Model 1 in Table 2 presents linear specifications of the religious variables and theoretically
relevant control variables. Results show that intrinsic religiosity (p < .001) and marriage (p <
.05) both are associated with fewer depressive symptoms, even though no association appears
between other measures of religiousness and depression. Thus, H1 is partially supported.
Model 2 includes the measure of perceived social support. The estimated effect of intrinsic
religiosity remains significant (p < .001), with negligible diminishment in the coefficient size.
Marriage loses statistical significance, implying that support received from one抯 spouse and
family members may be the underlying mechanism for the negative association between marriage
and depression.
Model 3 adds quadratic terms of intrinsic religiosity and strength of religious belief, with
results confirming the only inverse U-shaped association between intrinsic religiosity and depression.
Women with low levels of intrinsic religiosity, as well as those with high levels of intrinsic
religiosity, report fewer depressive symptoms than women with moderate levels of intrinsic
RELIGIOUS INVOLVEMENT AND DEPRESSION 359
Figure 1
CES-D by intrinsic religiosity
Y-axis=Depression X-axis=Intrinsic religiosity
religiosity.3 Figure 1 illustrates the inverse U-shaped effect of intrinsic religiosity for a respondent
with mean scores on the remaining religious measures and control variables. Since the
U-shaped effect is observed for intrinsic religiosity, H2 is partially supported.
Model 4 is an expanded regression that further controls for personal mastery and self-esteem.
Not surprising, both measures of self-concept are related with lower levels of depression. The
inverse U-shaped relationship between intrinsic religiosity and depression remains consistent with
the results presented in the previous model.4 Moreover, the inclusion of self-esteem and mastery
reveals a pattern of suppression. Although the coefficient size of the quadratic term of intrinsic
religiosity remains stable, the coefficient size of intrinsic religiosity increases by more than
14 percent when the self-concept measures are held constant. This may suggest that the estimated
negative effect of intrinsic religiousness on depression comes in part through the mechanism of
self-conception.5
Interaction Effects
Does religious involvement buffer or exacerbate the negative effects of stressful life events?
Since previous research indicates that the interaction effect may vary with how stressful life
events are measured (e.g., Schnittker 2001), we distinguish between the number of stressful
life events and event type. Panel A in Table 3 tests for interactions between religious measures
and event counts, whereas Panel B tests for interactions between religious measures and specific
types of stressful life events.
3The mean for the scale of intrinsic religiosity was relatively high, and the inflection point was somewhere in the middle,
meaning that for about 80 percent of the sample the association was simply negative.
4The inclusion of additional controls did not alter the inverse U-shaped pattern between models 3 and 4, as the coefficient
of the quadratic term of intrinsic religiosity remained stable across models. In supplementary analyses, no nonlinear
relationships were detected between depression and such controls as self-esteem and personal mastery.
5In auxiliary analyses, the relationships between the self-esteem items, mastery items, and the depression scale were
checked individually. Results were substantively similar to those based on the indices. The relatively low alpha scores
for the mastery and self-esteem scales may reflect cultural differences. Previous literature suggests that reported levels
of self-esteem and mastery tend to be lower among Chinese and Japanese people than they are among North Americans
because the Chinese and Japanese cultures encourage individuals to maintain interpersonal relationships by presenting
the self as inferior to others (Heine et al. 1999; Sastry and Ross 1998; Yik et al. 1998). Although high levels of mastery
and self-esteem may represent a norm violation for East Asians, prior empirical studies show that self-esteem and mastery
remain a universal need; therefore, they are important predictors of depressive symptoms in the Asian context (Cai, Wu,
and Brown 2009; Sastry and Ross 1998).
360 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Table 3: Summary of the effects of the interaction between negative life events and religious
variables on depression
Intrinsic Religious Private Religious
Religiosity Belief Practice Attendance
Panel A: Multiple Event Models
Interaction with:
One event ?118 .213* .443 .286
(.097) (.094) (.668) (1.880)
Two or more events ?012 ?115 ?034 ?204
(.079) (.086) (.543) (1.424)
Panel B: Discrete Event Models
Interaction with:
Legal problems ?191 .069 .867 1.606
(.101) (.105) (.629) (1.789)
Severe illness .049 ?021 .460 ?433
(.086) (.096) (.596) (1.556)
Work issues ?062 ?009 .476 ?.331
(.103) (.123) (.744) (1.796)
Financial problems .047 ?109 ?268 .103
(.087) (.101) (.636) (1.591)
Childbirth problems .315* ?009 ?227 1.077
(.145) (.170) (.933) (3.162)
Family conflicts ?158 .274 .395 1.257
(.108) (.146) (.834) (2.119)
Marital conflicts ?250 .000 2.035* 8.427***
(.141) (.145) (.838) (2.073)
Note: Models also include controls for age, marital status, education, perceived social support, and the main effects of
the religious measures and negative life events.
N = 882; *p < .05; **p < .01; ***p < .001 (two-tailed tests). Robust standard errors in parentheses.
Several significant interaction terms are detected (Table 3), all suggesting that some religious
measures may exacerbate the effects of stressful life events. First, the strength of religious belief
increases the estimated negative effect of the one-event dummy variable (p < .05). No significant
interactions between the religious measures and the multiple-events dummy emerge from the test
results. Additional analyses are performed, using different random subsamples; results suggest
that the interaction between strength of religious belief and the one-event dummy variable is no
more than we would expect by chance at the p < .05 level.
Moreover, Table 3 shows that intrinsic religiosity amplifies the estimated effect of traumatic
childbirth events, while private religious practice exacerbates the negative influence of marital
conflicts. Similarly, there is a robust interaction between religious attendance and marital conflicts
(p < .001). Auxiliary tests suggest that the interactions between the religious variables梬hich
were examined individually and simultaneously梐nd these two specific life events are not likely
due to chance. Thus, H3 is not supported.
DISCUSSION
The primary goal of this study has been to investigate nonlinear and stress-moderating
effects of religious involvement on depression. Overall, our findings indicate that (1) among
young women in rural China there is an inverse U-shaped association between intrinsic religiosity
RELIGIOUS INVOLVEMENT AND DEPRESSION 361
and depressive symptoms; and (2) intrinsic religiosity, private religious practices, and religious
attendance tend to exacerbate the estimated deleterious effects of certain types of stressful life
events such as marital conflicts and childbirth problems. These findings merit further discussion.
Perhaps the most important result of our analyses is the nonlinear relationship between religiousness
and depression. Two previous studies have identified an inverse U-shaped relationship
between religious involvement and depression (Eliassen, Taylor, and Lloyd 2005; Ross 1990)
while one has identified a U-shaped distribution in the U.S. population (Schnittker 2001). This
study confirms the inverse U-shaped relationship among rural young Chinese women. Further,
consistent with prior U.S.-based research (Smith, McCullough, and Poll 2003), we show that
intrinsic religiosity is a more powerful measure than religious belief and other religious measures
in a Chinese sample. On the other hand, our findings are inconsistent with those of Schnittker
(2001), who found a U-shaped relationship between religious salience and depression. From
Schnittker抯 perspective, religious salience might more accurately reflect the degree to which
religion regulates an individual抯 behavior on a daily basis, at least in comparison with measures
of religious belief. Because there is a growing gulf between private and public faith and related
practices, public and private religious practice may have different effects on depression, with only
some forms of involvement having optimal or significant effects (Schnittker 2001). However, this
perspective may not be an adequate explanation of why the relationship between intrinsic religiosity
and depression is inversely U-shaped, as discovered in this study, especially considering that
intrinsic religiosity is also a widely adopted measure of the degree to which religion regulates an
individual抯 day-to-day behavior. One explanation could be that the divergence results from how
religiousness is measured (religious salience measured with a single item vs. intrinsic religiosity
measured with a scale). An alternative explanation may be that the nonlinear relationship is not
universal but culture specific. In any event, this study by no means indicates that the prior finding
of the U-shaped relationship is necessarily incorrect; instead, it suggests that continuing research
is needed.
In addition, this study identifies conditional relationships between religious involvement
and depression: negative life events are positively related to depression, but the slope of this
association is steeper for women who self-identify as high in religiosity. Specifically, although
the behavioral measures of religious involvement bore no direct relationship to depression,
this study finds significant stress-exacerbating effects of intrinsic religiosity, private religious
practices, and religious attendance among young rural Chinese women. The religious measures
had an exacerbating effect on depression through interaction with certain types of stressful life
events such as marital conflict and childbirth issues.Women who experienced marital conflict and
childbirth problems, as well as those who had greater intrinsic religiosity or engaged in private
or public religious practices, also reported more depressive symptoms. In the patrilineal Chinese
society, familial and marital relationships are highly male-dominated (Yang 1970). As a result,
Chinese family and marriage are characterized by male domination and even chauvinism, which
hardly functions as a protective factor against mental disorders for women (Zhang et al. 2010).
Indeed, family issues (including domestic violence and spousal conflicts) lead to increased levels
of depression among young Chinese women (Cong et al. 2012; Lau 2011; Lee et al. 1997; Wang
et al. 2012). Childbirth-related problems, such as miscarriage and giving birth to daughters,
also result in more depressive symptoms among young Chinese women (Qiu and Xiao 2010;
Ramashwar 2007; Tao et al. 2011). Moreover, traditional Chinese religions, especially Chinese
folk religions and Confucianism, typically serve as tools for perpetuating gender discrimination
in the patrilineal society (Law 2005). For instance, Confucianism teaches that women are inferior
to men, that women should not receive education or participate in social activities outside of the
home, and that the wife should take good care of the husband and give birth to a son to continue
her husband抯 family bloodline. Such religious norms put Chinese women in a highly vulnerable
position, leading to increased levels of female depression and higher suicide rates (Zhang and
Liu 2012). Thus, it is conceivable that traditional Chinese religiousness reinforces and amplifies
362 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
the harmful effects of certain stressful life events such as marital conflicts and childbirth issues  for women.
There are several limitations to this study. First, we employ a sample of rural Chinese women
aged 18?4. As discussed earlier, there are sound justifications for the study抯 focus on young
women who reside in rural China. For one thing, prior U.S.-based research suggested that the
nonlinear relationship between religiousness and depression exists only amongwomen, not among
men (Eliassen, Taylor, and Lloyd 2005). For another, young rural Chinese women experience
substantially higher rates of mental disorders than their male counterparts, and tend to draw on
religious resources to deal with daily affairs and events (e.g., Law 2005; Pearson 1995). Although
the sample is innovative, the findings of this study may not be representative of the general
population in China or of people in other cultures and societies. Caution should be exercised
against overgeneralizing the results beyond the sample population. Further empirical investigation
should explore, from a comparative perspective, whether there is a nonlinear relationship between
religiousness and depression for men and older individuals in the Chinese and broader non-U.S.
contexts.
Second, it is impossible with cross-sectional data to determine the causality of the linear
relationship between religious involvement and depressive symptoms. For instance, levels of
depression are greater in women who attended religious services and experienced childbirth
difficulties than they are in those who did not.We might put forth a plausible argument thatwomen
who experience postpartum depression might be more likely to participate in religious rituals
promoting health. It is equally legitimate to argue that attendance at ritual services might increase
depression levels in women faced with negative life events related to childbirth. Nevertheless, it
is less so for the nonlinear relationship between religiosity and depression. In fact, the U-shaped
relationship is evidence of causal order (Ross 1990). As Ross asserts, 搕he fact that distress was
lowest among the very religious and among the nonreligious suggests that religion probably
affects distress rather than vice versa. It is unlikely that low levels of distress would lead a person
to become either very religious or not at all religious?(1990:243).
In addition to limitations related to the sample, it is also worthwhile to discuss the modified
measure of intrinsic religiosity employed in this study. Because of our focus on China, we
adopted the recently modified Hoge IR Scale that includes several questions about beliefs in
Buddha, Bodhisattva, ancestral spirits, deities, and ghosts (Liu and Koenig 2012). This modified
scale replaces the monotheistic Judeo-Christian-Islamic God with a range of entities. Since each
of these is a distinct entity, we might expect different connections, interpretations, and emotional
reactions to each (ranging from ambivalence to fear to comfort). For instance, Roemer (2010)
examines the different interactions Japanese have between (1) Shinto gods, deities, and spirits
and (2) ancestors; he finds opposite effects on psychological distress in some cases. Especially
because the survey design of this study includes the different Chinese entities combined in a
single question, potential distinctions are worth further investigation in China. With this caveat
in mind, it is true that radically different images of God coexist even within monotheism (Froese
and Bader 2010).We might continue to suspect that neither the original nor the modified versions
of the Hoge IR Scale can fully capture the rich variations in supernatural beliefs. However, it is
important to understand that at a basic level, individuals with intrinsic religious orientations are
搈ore concerned with the quality of their faith and connectedness with God (in whatever form)?br /> (Thoresen, Harris, and Oman 2001:19, emphasis added). Just as Pargament (1998) notes, religious
orientations describe general tendencies or inclinations that do not speak to the particulars of any
situation; instead, they are cross-situational phenomena.
In summary, this study explores the curvilinear and stress-moderating effects of religious
involvement on depressive symptoms. Using a sample of Chinese women, this study aims to
enhance understanding of the relationship between religion and mental health in a less-commonly
studied region of the world. Results from our analyses confirm the inverse U-shaped distribution
identified by two previous studies conducted in the United States (Eliassen, Taylor, and Lloyd
RELIGIOUS INVOLVEMENT AND DEPRESSION 363
2005; Ross 1990). The analyses also show that religious involvement does not universally benefit
mental health among young rural Chinese women. In fact, religiosity exacerbates the effects of
certain negative life events on psychological distress. Because this study employs data from a
non-U.S. sample and employs a measure of religious involvement (intrinsic religiosity) that has
previously not been used to assess these relationships, the generalizability of these and previous
findings is extended. This generalizability is especially important for a study centered on China,
where梒ompared to predominantly Christian societies梩emple visits are not very common
(especially in the form of regular group worship).
REFERENCES
Allport, Gordon W. 1966. Religious context of prejudice. Journal for the Scientific Study of Religion 5(4):447?7.
Allport, Gordon W. and J. Michael Ross. 1967. Personal religious orientation and prejudice. Journal of Personality and
Social Psychology 5(4):432?3.
Antonovsky, Aaron. 1987. Unraveling the mystery of health. San Francisco: Jossey-Bass.
Branden, Nathaniel. 1994. The six pillars of self-esteem. New York: Bantam Books.
Breault, Kevin D. 1986. Suicide in America: A test of Durkheim抯 theory of religious and family integration, 1933?980.
American Journal of Sociology 92(3):628?6.
Brown, Diane R., Samuel C. Ndubuisi, and Lawrence E. Gary. 1990. Religiosity and psychological distress among blacks.
Journal of Religion and Health 29(1):55?8.
Cai, Huajian, Qiuping Wu, and Jonathon D. Brown. 2009. Is self-esteem a universal need? Evidence from the People抯
Republic of China. Asian Journal of Social Psychology 12(2):104?0.
Carothers, Shannon S., John G. Borkowski, Jennifer Burke Lefever, and Thomas L. Whitman. 2005. Religiosity and the
socioemotional adjustment of adolescent mothers and their children. Journal of Family Psychiatry 19(2):263?5.
Cheng, Sheung-Tak and Alfred C.M. Chan. 2004. Themultidimensional scale of perceived social support: Dimensionality
and age and gender differences in adolescents. Personality and Individual Differences 37(7):1359?9.
Cong, E., Y. Li, C. Shao, J. Chen, W. Wu, X. Shang, Z. Wang, Y. Liu, L. Liu, C. Gao, Y. Li, J. Wu, H. Deng, J. Liu, W.
Sang, G. Liu, H. Rong, Z. Gan, L. Li, K. Li, J. Pan, Y. Li, Y. Cui, L. Sun, L. Liu, H. Liu, X. Zhao, Y. Zhang, R.
Zhang, Y. Chen, X. Wang, H. Li, Y. Chen, Y. Lin, K. S. Kendler, J. Flint, and S. Shi. 2012. Childhood sexual abuse
and the risk for recurrent major depression in Chinese women. Psychological Medicine 42(2):409?7.
Cornwall, Marie. 2009. Reifying sex difference isn抰 the answer: Gendering processes, risk, and religiosity. Journal for
the Scientific Study of Religion 48(2):252?5.
Dahlem, Nancy W., Gregory D. Zimet, and Robin R. Walker. 1991. The multidimensional scale of perceived social
support: A confirmation study. Journal of Clinical Psychology 47(6):756?1.
Dai, Kejing. 1991. The life experience and status of Chinese rural women from observation of three age groups.
International Sociology 6(1):5?3.
Duara, Prasenjit. 1988. Culture, power and the state: Rural north China, 1900?942. Palo Alto, CA: Stanford University
Press.
DuBois, Thomas David. 2005. The sacred village: Social change and religious life in rural north China. Honolulu:
University of Hawaii Press.
Durkheim, ?Emile. [1897] 1951. Suicide: A study in sociology. Glencoe, IL: Free Press.
Eliassen, Henry A., John Taylor, and Donald A. Lloyd. 2005. Subjective religiosity and depression in the transition to
adulthood. Journal for the Scientific Study of Religion 44(2):187?9.
Ellis, Albert. 1962. Reason and emotion in psychotherapy. New York: Lyle Stuart.
棗? 1980. Psychotherapy and atheistic values: A response to A. E. Bergin抯 揚sychotherapy and Religious Values.?br /> Journal of Consulting and Clinical Psychology 48(5):635?9.
棗?1986. Do some religious beliefs help create emotional disturbances? Psychotherapy in Private Practice 4(4):101?br /> 06.
棗? 1988. Is religiosity pathological? Free Inquiry 8(1):27?2.
Ellison, Christopher G. 1991. Religious involvement and subjective well-being. Journal of Health and Social Behavior
32(1):80?9.
棗? 1993. Religious involvement and self-perception among black Americans. Social Forces 71(4):1027?5.
棗?1994. Religion, the life stress paradigm, and the study of depression. In Religion in aging and health: Theoretical
foundations and methodological frontiers, edited by Jeffrey S. Levin, pp. 78?21. Thousand Oaks, CA: Sage.
Ellison, Christopher G., Jason D. Boardman, David R. Williams, and James S. Jackson. 2001. Religious involvement,
stress, and mental health: Findings from the 1995 Detroit Area Study. Social Forces 80(1):215?9.
Ellison, Christopher G., Brian K. Finch, Daniell Nicole Ryan, and Jennifer J. Salinas. 2009. Religious involvement and
depressive symptoms among Mexican-origin adults in California. Journal of Community Psychology 37(2):171?3.
364 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Ellison, ChristopherG. and Jeffrey S. Levin. 1998. The religion-health connection: Evidence, theory, and future directions.
Health Education and Behavior 25(6):700?0.
Fan, Lizhu. 2003. Popular religion in contemporary China. Social Compass 50(4):449?7.
Flannelly, Kevin, Kathleen Galek, Christopher G. Ellison, and Harold G. Koenig. 2009. Beliefs about God, psychiatric
symptoms, and evolutionary psychiatry. Journal of Religion and Health 49(2):246?1.
Francis, Leslie J. 2007. Introducing the New Indices of Religious Orientation (NIRO): Conceptualization and measurement.
Mental Health, Religion and Culture 10(6):585?02.
Freedman, Maurice. 1974. On the sociological study of Chinese religion. In Religion and ritual in Chinese society, edited
by Arthur P. Wolf, pp. 19?1. Palo Alto, CA: Stanford University Press.
Freese, Jeremy. 2004. Risk preferences and gender differences in religiousness: Evidence from theWorld Values Survey.
Review of Religious Research 46(1):88?1.
Froese, Paul and Christopher Bader. 2010. America抯 four gods: What we say about God桝nd what that says about us.
New York: Oxford University Press.
Genia, Vicky. 1998. Religiousness and psychological adjustment in college students. Journal of College Student Psychotherapy
12(3):67?7.
Gorsuch, Richard L. and Susan E. McPherson. 1989. Intrinsic/extrinsic measurement: I/E-revised and single-item scales.
Journal for the Scientific Study of Religion 28(3):348?4.
Gorsuch, Richard L. and G. Daniel Venable. 1983. Development of an 揳ge universal?I-E scale. Journal for the Scientific
Study of Religion 22(2):181?7.
Hackney, Charles H. and Glenn S. Sanders. 2003. Religiosity and mental health: A meta-analysis of recent studies.
Journal for the Scientific Study of Religion 42(1):43?5.
Heine, Steven J., Darrin R. Lehman, Hazel Rose Markus, and Shinobu Kitayama. 1999. Is there a universal need for
positive self-regard? Psychological Review 106(4):766?4.
Hoge, Dean R. 1972. A validated intrinsic religious motivation scale. Journal for the Scientific Study of Religion 11(4):369?6.
Idler, Ellen L. 1987. Religious involvement and the health of the elderly: Some hypotheses and an initial test. Social Forces 66(1):226?8.
棗? 1995. Religion, health, and nonphysical senses of self. Social Forces 74(2):683?04.
Jaschok, Maria. 2003. Violation and resistance: Women, religion, and Chinese statehood. Violence against Women
9(6):655?5.
Kazarian, Shahe S. and Scott B. McCabe 1991. Dimensions of social support in the MSPSS: Factorial structure, reliability,
and theoretical implications. Journal of Community Psychology 19(2):150?0.
Koenig, Harold G., Dana E. King, and Verna B. Carson. 2012. Handbook of religion and health, 2nd ed. New York: Oxford University Press.
Koenig, Harold G., Michael E. McCullough, and David B. Larson. 2001. Handbook of religion and health. New York: Oxford University Press.
Lau, Ying. 2011. A longitudinal study of family conflicts, social support and antenatal depressive symptoms among Chinese women. Archives of Psychiatric Nursing 25(3):206?9.
Lavric, Miran and Sergej Flere. 2008. The role of culture in the relationship between religiosity and psychological
well-being. Journal of Religion and Health 47(2):164?5.
Law, Pui-Lam. 2005. The revival of folk religion and gender relationships in rural China: A preliminary observation.
Asian Folklore Studies 64(1):89?09.
Lee, D. T. S., T. K. H. Chung, C. M. Leung, and L. Cheung. 1997. Depression after miscarriage in Chinese women.
Psychosomatic Medicine 59(1):94?5.
Lin, Nan. 1989. Measuring depressive symptomatology in China. Journal of Nervous and Mental Disease 117(3):121?1.
Liu, Eric Y. 2009. Beyond the West: Religiosity and the sense of mastery in modern Taiwan. Journal for the Scientific
Study of Religion 48(4):774?8.
棗? 2011. Religion and mental health in China. In Religion and health: Toward a sociological theory of religion and
health, edited by Antony Blasi, pp. 141?4. Boston: Brill.
Liu, Eric Y. and Harold G. Koenig. 2012. Measuring intrinsic religiosity: Scales for use in mental health studies in
China桝 research report. Mental Health, Religion, and Culture 16(2):215?4.
Liu, Eric Y., Harold G. Koenig, and DedongWei. 2012. Discovering a blissful island: Religious involvement and personal
happiness in Taiwan. Sociology of Religion 73(1):46?8.
Liu, Eric Y. and Fredrick Carson Mencken. 2010. Fatalistic voluntarism and life happiness in post-socialist China.
Sociological Spectrum 30(3):270?8.
Liu, Eric Y., Scott Schieman, and Sung Joon Jang. 2011. Religiousness, spirituality, and psychological distress in Taiwan.
Review of Religious Research 53(2):137?9.
Mattlin, Jay A., Elaine Wethington, and Ronald C. Kessler. 1990. Situational determinants of coping and coping effectiveness.
Journal of Health and Social Behavior 31(1):103?2.
McCullough, Michael E. and David B. Larson. 1999. Religion and depression: A review of the literature. Twin Research 2(2):126?6.
RELIGIOUS INVOLVEMENT AND DEPRESSION 365
McDonald, Coval B. and Jeffrey B. Luckett. 1983. Religious affiliation and psychiatric diagnoses. Journal for the Scientific Study of Religion 22(1):15?7.
McIntosh, Daniel N., Roxane Cohen Silver, and Camille B. Wortman. 1993. Religion抯 role in adjustment to a negative
life event: Coping with the loss of a child. Journal of Personality and Social Psychology 65(4):812?1.
McLaughlin, Caitlin S., Chuansheng Chen, Ellen Greenberger, and Cornelia Biermeier. 1997. Family, peer and individual
correlates of sexual experience among Caucasian and Asian American late adolescents. Journal of Research on
Adolescence 7(1):33?3.
Miller, Alan S. and Rodney Stark. 2002. Gender and religiousness: Can socialization explanation be saved? American
Journal of Sociology 6(4):1399?423.
Mirola, William A. 1999. A refuge for some: Gender differences in the relationship between religious involvement and
depression. Sociology of Religion 60(4):419?7.
Musick,MarcA. 2000. Theodicy and life satisfaction among black andwhite Americans. Sociology of Religion 61(3):267?br /> 87.
National Bureau of Statistics of China. 2001. Chinese population census 2000. Available at http://www.stats.gov.
cn/english/statisticaldata/yearlydata/YB2000e/index1.htm, accessed January 21, 2013.
Overmyer, Daniel L. 1986. Religions of China: The world as a living system. Prospect Heights, IL: Waveland Press.
棗? 2003. Religion in China today: Introduction. China Quarterly 2003(174):307?6.
棗?2009. Local religion in north China in the twentieth century: The structure and organization of community rituals
and beliefs. Boston: Brill.
Pargament, Kenneth I. 1997. The psychology of religion and coping: Theory, research, practice. New York: Guilford
Press.
Pargament, Kenneth I., David S. Ensing, Kathryn Falgout, Hannah Olsen, Barbara Reilly, Kimberly Van Hatisma, and
RebeccaWarren.1990. God help me (I): Religious coping efforts as predictors of the outcomes of significant negative
life events. American Journal of Community Psychology 18(6):793?24.
Pearlin, Leonard I. and Carmi Schooler. 1978. The structure of coping. Journal of Health and Social Behavior 19(1):2?1.
Pearson, Veronica. 1995. Goods on which one loses: Women and mental health in China. Social Science Medicine
41(8):1159?3.
Pescosolido, Bernice A. and Sharon Georgianna. 1989. Durkheim, suicide, and religion: Toward a network theory of
suicide. American Sociological Review 54(1):33?8.
Petersen, Larry R. and Anita Roy. 1985. Religiosity, anxiety, and meaning and purpose: Religion抯 consequences for
psychological well-being. Review of Religious Research 27(1):49?2.
Pollner,Melvin. 1989. Divine relations, social relations, and well-being. Journal of Health and Social Behavior 30(1):92?br /> 104.
Pope, Whitney and Nick Danigelis. 1981. Sociology抯 搊ne law.?Social Forces 60(2):495?16.
Qiu, Jianyin and Xiao Zeping. 2010. Reproductive-related depression in women: A review of two decades of research in
China. Asia-Pacific Psychiatry 2(1):19?5.
Radloff, Lenore Sawyer. 1977. The CES-D scale: A self-report depression scale for research in the general population.
Applied Psychological Measurement 1(3):385?01.
Ramashwar, S. 2007. In China,womenwho give birth to girls face an increased risk of postpartum depression. International
Family Planning 33(4):191?2.
Roemer, Michael K. 2010. Religion and psychological distress in Japan. Social Forces 89(2):559?3.
Rosenberg, Morris. 1965. Society and the adolescent self-image. Princeton, NJ: Princeton University Press.
Ross, Catherine E. 1990. Religion and psychological distress. Journal for the Scientific Study of Religion 29(2):236?5.
Roth, Louise Marie and Jeffery C. Kroll. 2005. Risky business: Assessing risk-preference explanations for gender
differences in religiosity. American Sociological Review 72(2):205?0.
Salsman, John M. and Charles R. Carlson. 2005. Religious orientation, mature faith, and psychological distress: Elements
of positive and negative associations. Journal for the Scientific Study of Religion 44(2):201?9.
Sastry, Jaya and Catherine E. Ross. 1998. Asian ethnicity and the sense of personal control. Social Psychology Quarterly
61(2):101?0.
Schieman, Scott. 2011. Religious beliefs and mental health: Applications and extensions of the stress process model. In
Handbook of mental health and mental disorder: Perspectives from social science, edited by David Pilgrim, Anne
Rogers, and Bernice Pescosolido, pp. 179?01. Thousand Oaks, CA: Sage.
Schieman, Scott, Kim Nguyen, and Diana Elliott. 2003. Religiosity, socioeconomic status, and the sense of mastery.
Social Psychology Quarterly 66(3):202?1.
Schieman, Scott, Tetyana Pudrovska, Leonard I. Pearlin, and Christopher G. Ellison. 2006. The sense of divine control
and psychological distress: Variations across race and socioeconomic status. Journal for the Scientific Study of
Religion 45(4):529?9.
Schnittker, Jason. 2001. When is faith enough? The effects of religious involvement on depression. Journal for the
Scientific Study of Religion 40(3):393?11.
Shaver, Phillip, Michael Lenauer, and Susan Sadd. 1980. Religiousness, conversion, and subjective well-being: The
揾ealthy-minded?religion of modern American women. American Journal of Psychiatry 137(2):1563?8.
366 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Smith, Timothy B., Michael E. McCullough, and Justin Poll. 2003. Religiousness and depression: Evidence for a main
effect and the moderating influence of stressful life events. Psychological Bulletin 129(4):614?6.
Sorenson, Ann Marie, Carl F. Grindstaff, and R. Jay Turner. 1995. Religious involvement among unmarried adolescent
mothers: A source of emotional support? Sociology of Religion 56(1):71?1.
Spilka, Bernard. 1987. Functional and dysfunctional roles of religion: An attributional approach. Journal of Psychology
and Christianity 8(3):5?5.
Stack, Steven. 1983. The effect of religious commitment on suicide: A cross-national analysis. Journal of Health and
Social Behavior 24(4):362?4.
Stanley, Melinda A., J. Gayle Beck, and Barbara J. Zebb. 1998. Psychometric properties of the MSPSS in older adults.
Aging and Mental Health 2(3):186?3.
Stark, Rodney and Eric Y. Liu. 2011. The religious awakening in China. Review of Religious Research 52(3):282?9.
Steger, Michael F. and Patricia Frazier. 2005. Meaning in life: One link in the chain from religion to well-being. Journal
of Counseling Psychology 52(4):574?2.
Strawbridge, William J., Sarah J. Shema, Richard Cohen, Robert E. Roberts, and George A. Kaplan. 1998. Religiosity
buffers effects of some stressors on depression but exacerbates others. Journal of Gerontology: Social Sciences
53B(3):S118?6.
Sullins, D. Paul. 2006. Gender and religion: Deconstructing universality, constructing complexity. American Journal of
Sociology 112(3):838?0.
Tabak, Melanie A. and Kristin D. Mickelson. 2009. Religious service attendance and distress: The moderating role of
stressful life events and race/ethnicity. Sociology of Religion 70(1):49?4.
Tao, Ming, Yihan, Li, Dong Xie, Zhiyang Wang, Jianying Qiu, Wenyuan Wu, Jing Sun, Zhoubing Wang, Danhong Tao,
Hongsu Zhao, Tian Tian, Jingxuan Zhang, Chengge Gao, Qihui Niu, Qiang Li, Shanming Liu, Jia Liu, Yunshu
Zhang, Qiang He, Han Rong, Zhaoyu Gan, Jianying Li, Xiansheng Chen, Jiyang Pan, Yi Li, Yanping Cui,Wei Han,
Huan Ma, Shoufu Xie, Guixing Jin, Ling Li, Ruiling Zhang, Qingrong Tan, Jun Zhang, Jing Guan, Shenxun Shi,
Yiping Chen, Kenneth S, Kendler, Jonathan Flint, and Jingfang Gao. 2011. Examining the relationship between
lifetime stressful life events and the onset of major depression in Chinese women. Journal of Affective Disorders
135(1?):95?9.
Tapanya, Sombat, Richard Nicki, and Ousa Jarusawad. 1997. Worry and intrinsic/extrinsic religious orientation among
Buddhist (Thai) and Christian (Canadian) elderly persons. International Journal of Aging and Human Development
44(1):73?3.
Thoresen, Carl E., Alexis H. S. Harris, and Doug Oman. 2001. Spirituality, religion, and health: Evidence, issues, and
concerns. In Faith and health: Psychological perspectives, edited by Thomas G. Plante and Allen C. Sherman,
pp. 15?2. New York: Guilford Press.
Veroff, Joseph, Elizabeth Douvan, and Richard A. Kulka. 1981. The inner American: A self-portrait from 1957 to 1976.
New York: Basic Books.
Wang, Lina, Dongdong Qiao, Yihan Li, Liwei Wang, Jianer Ren, Kangmei He, Jing Sun, Zhoubing Wang, Tian Tian, Ce
Chen, Lei Yang, Jian Hu, Hong Deng, Qian Wang, Keqing Li, Jiyang Han, Han Rong, Zhaoyu Gan, Hong Yang,
Pingliang Zhou, Jiyang Pan, Cong Zhou, Yanping Cui, Libo Song, Yuzhang Zhu, Ying Li, XueyiWang, Lanxian Ye,
Wei Liang, Yunchun Chen, Qingjun Tang, Jing Guan, Shenxun Shi, Kenneth S. Kendler, Jonathan Flint, and Lanfen
Liu. 2012. Clinical predictors of familial depression in Han Chinese women. Depression and Anxiety 29(1):10?5.
Watson, P. J., Nima Ghorbani, H. Kristl Davison, Mark N. Bing, Ralph W. Hood Jr., and Ahad Framarz Ghramaleki.
2002. Negatively reinforcing personal extrinsic motivations: Religious orientation, inner awareness, and mental
health in Iran and the United States. International Journal for the Psychology of Religion 12(4):255?6.
Watson, P. J., Ralph W. Hood Jr., Shelly G. Foster, and Ronald J. Morris. 1988. Sin, depression, and narcissism. Review
of Religious Research 29(3):295?05.
Williams, David R., David B. Larson, Robert E. Buckler, Richard C. Heckmann, and Carolyn Pyle. 1991. Religion and
psychological distress in a community sample. Social Science and Medicine 32(11):1257?2.
Wilson, Bryan R. 1992. Religion in sociological perspective. New York: Oxford University Press.
Wink, Paul,Michele Dillon, and Britta Larsen. 2005. Religion asmoderator of the depression-health connection: Findings
from a longitudinal study. Research on Aging 27(2):197?20.
Yan, Ming. 2010.. [The history of Chinese sociology]. Beijing: Tsinghua University Press.
Yang, C. K. 1961. Religion in Chinese Society: A Study of Contemporary Social Functions of Religion and Some of Their
Historical Factors. Berkeley: University of California Press.
棗? 1970. Religion in Chinese society: A study of contemporary social functions of religion and some of their
historical facts. Berkeley: University of California Press.
Yik, M. S. M., M. H. Bond, and D. L. Paulhus. 1998. Do Chinese self-enhance or self-efface?: It抯 a matter of domain.
Personality and Social Psychology Bulletin 24:399?06.
Zhang, Jie, Yeates Conwell,William F.Wieczorek, Chao Jiang, Shuhua Jia, and Li Zhou. 2003. Studying Chinese suicide with proxy-based data. Journal Nervous and Mental Disease 191(7):450?7.
Zhang, Jie, Chao Jiang, Shuhua Jia, and William F. Wieczorek. 2010. An overview of suicide research in China. Archive of Suicide Research 6(2):167?4.
RELIGIOUS INVOLVEMENT AND DEPRESSION 367
Zhang, Jie and Eric Y. Liu. 2012. Confucianism and youth suicide in rural China. Review of Religious Research 54(1):93?br /> 111.
Zhang, Jie and Jill M. Norvilitis. 2002. Measuring Chinese psychological well-being withWestern developed instruments.
Journal of Personality Assessment 79(3):492?11.
Zimet, Gregory D., Nancy W. Dahlem, Sara G. Zimet, and Gordon K. Farley. 1988. The multidimensional scale of
perceived social support. Journal of Personality Assessment 52(1):30?1.
Zimet, Gregory D., Suzanne S. Powell, Gordon K. Farley, Sidney Werkman, and Karen A. Berkoff. 1990. Psychometric
properties of the multidimensional scale of perceived social support. Journal of Personality Assessment 55(3?br /> 4):610?7.
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online version of this article at the
publisher抯 website:
Appendix S1. Chinese translation of the CES-D scale
Appendix S2. 10-Item Hoge Intrinsic Religiosity scale, English and Chinese versions
Appendix S3. Hoge IR item scores and correlations with total scores
Appendix S4. Correlations between religious measures

中国人民大学国际佛学研究中心    技术支持:北京维创科技