JOURNAL for the
SCIENTIFIC STUDY of RELIGION
Sleep Quality and the Stress-Buffering Role of
Religious Involvement: A Mediated
CHRISTOPHER G. ELLISON
Department of Sociology
University of Texas at San Antonio
REED T. DEANGELIS
Department of Sociology
University of North Carolina at Chapel Hill
TERRENCE D. HILL
School of Sociology
University of Arizona
Department of Sociology
Although several studies have documented an inverse association between stressful events and sleep quality, much
less is known about the factors that might moderate or buffer against the adverse effects of psychosocial stress
on sleep. Building on previous research, we employ national cross-sectional survey data from the 2017 Baylor
Religion Survey (n=1,410) to test whether the association between recent stressful events and sleep quality varies
according to several dimensions of religious involvement. We also formally assess whether any attenuation of the
association between stressful events and sleep quality is at least partially mediated or explained by lower levels
of depressive symptoms (mediated moderation). Our moderation analyses indicate that the inverse association
between stressful events and sleep quality is in fact attenuated by religious cognitions (secure attachment to God
and assurance of salvation), but not religious attendance or private religiousness. We also observe direct evidence
of mediated moderation through depressive symptoms for both religious cognitions. Taken together, our results
demonstrate that religious cognitions may buffer against stress-related sleep disturbance by helping people avoid
symptoms of depression.
Keywords: stress, sleep, depression, religion, religious beliefs, attachment to God, afterlife.
A growing body of work in multiple disciplines has demonstrated the importance of sleep
quality for a range of health outcomes. Brieﬂy, sleep quality encompasses both quantitative (e.g.,
duration and latency) and qualitative (e.g., depth and soundness) aspects of sleep (Pilcher, Ginter,
and Sadowsky 1997). Unfortunately, optimal sleep remains elusive for many people, typically
due to insomnia, sleep apnea, restless legs syndrome, and other forms of sleep disruption. Such
difﬁculties with sleep have been linked with an arrayof complications, including but not limited to
cardiovascular and circulatory dysfunction, obesity and diabetes, affective disorder, and reduced
life expectancy (Davies et al. 2014; Grandner et al. 2010; Riemann, Berger, and Voderholzer
2001). Sleep problems have also been shown to impair cognitive functioning, thereby limiting
work and academic performance and increasing the risk of accidents, as well as to undermine
interpersonal and familial relationships (Engle-Friedman et al. 2003; Williamson and Feyer 2000).
Given the far-reaching effects of suboptimal sleep, numerous studies have explored the epi-
demiology of sleep-related outcomes according to various social and behavioral factors, including
age, gender, race-ethnicity, socioeconomic status, substance use, and exercise (Grandner et al.
Correspondence should be addressed to Christopher Ellison, Department of Sociology, University of Texas at San Antonio,
One UTSA Circle, San Antonio, TX 78249. E-mail: email@example.com
Journal for the Scientiﬁc Study of Religion (2019) 0(0):1–18
2019 The Society for the Scientiﬁc Study of Religion
2JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
2016; Hoch et al. 1997; Mezick et al. 2008; Middlekoop et al. 1996). One particularly robust
predictor of impaired sleep is psychosocial stress (Kim and Dimsdale 2007). Indeed, major life
events and chronic conditions can take a signiﬁcant toll on sleep (Hale, Hill, and Burdette 2010;
Mai et al. 2018; Williamson et al. 1995). Consequently, researchers have worked to identify
factors that may mitigate the effects of psychosocial stress on sleep outcomes, such as social
support and coping skills, among others (Morin, Rodrigue, and Ivers 2003).
One avenue of research that has received surprisingly little attention is whether religious
involvement is linked with sleep. In fact, a recent review identiﬁed only seven population-based
studies that explore associations between religious factors and sleep outcomes (Hill, DeAngelis,
and Ellison 2018).1This general pattern of neglect contrasts sharply with the burgeoning theoret-
ical and empirical literature on religion and other mental and physical health outcomes, including
mortality risk (Hill, Burdette, and Bradshaw 2016; Koenig, King, and Carson 2012; Schieman,
Bierman, and Ellison 2013). Although the modest existing literature points to salutary associa-
tions between religion and sleep, several important limitations need to be addressed: (1) heavy
reliance on rudimentary religion measures, typically emphasizing organizational and nonorga-
nizational practices such as attendance and prayer; (2) neglect of the potential stress-buffering
effects of religious factors vis-`
a-vis sleep outcomes (for a rare exception, see White et al. 2018);
and (3) inattention to the processes that mediate or explain any stress-buffering effects of religion
on sleep (mediated moderation). Our study contributes to the literature by addressing each of
In the pages that follow, we brieﬂy review the connections between one particularly important
type of psychosocial stress—namely, stressful life events—and sleep. We then develop a series
of theoretical arguments explaining why religious involvement should buffer the noxious effects
of life events on sleep quality. Although we consider the roles of organizational and nonorgani-
zational religious practices, we also highlight two distinct religious cognitions: (1) a perceived
secure attachment to God; and (2) a subjective certainty in personal salvation, or the conviction
that one is destined for a blissful afterlife. Next, we develop a theoretical argument outlining a
process of mediated moderation in which religious involvement buffers the association between
stressful life events and sleep by reducing symptoms of depression. Relevant hypotheses are then
tested with national, cross-sectional survey data from the 2017 Baylor Religion Survey (n=
1,410). After presenting results from multivariate regression analyses, we discuss a number of
implications, limitations, and promising directions for future research.
THEORETICAL AND EMPIRICAL BACKGROUND
Stressful Life Events and Sleep
Psychosocial stress has been established as a robust determinant of sleep impairment (Kim
and Dimsdale 2007). Although research in this area has focused on various sources of stress
(e.g., Hale, Hill, and Burdette 2010; Mai et al. 2018), the study of major life events has a long
history in the literature on sleep outcomes (Guastella and Moulds 2007; Healey et al. 1981;
Williamson et al. 1995). Stressful events may increase the risk of insomnia and other forms of
sleep disruption partly because distressed individuals experience high levels of cognitive arousal
and rumination (Guastella and Moulds 2007; Morin, Rodrigue, and Ivers 2003). For example, they
may obsessively review the details and circumstances surrounding a given event, the problem-
solving or emotion-regulation efforts required for adaptive response, as well as any accompanying
1We should note that although our study centers on population-based research, a number of clinical studies have also
made important contributions to the study of religion and sleep (e.g., Khoramirad et al. 2015; Kopp et al. 2017; Vidal and
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 3
challenges to their sense of self or major life goals. All of this may make it difﬁcult to relax, fall
asleep, and sustain normal sleep architecture (Guastella and Moulds 2007; Morin, Rodrigue, and
Ivers 2003). Following a stressful event, people may also develop anxiety about the very prospect
of sleep loss and its cumulative consequences for health and role performance, further amplifying
the noxious effects of stressful life events on sleep (Harvey 2002). The above considerations lead
to our ﬁrst hypothesis:
H1: The number of recent life events will be inversely associated with sleep quality.
The Stress-Buffering Role of Religious Practices
Religion is a complex, multidimensional phenomenon (e.g., Idler et al. 2003) and various
facets of religious involvement may be differentially associated with sleep (Hill, DeAngelis, and
Ellison 2018). At a minimum, researchers typically distinguish between organizational (e.g.,
service attendance) and nonorganizational (e.g., prayer, scripture reading) forms of religious
involvement. In addition, scholars have become increasingly interested in the health implications
of religious cognitions, particularly beliefs about God and the afterlife (Flannelly 2017; Park
2017; Schieman, Bierman, and Ellison 2013). In this study, we explore the possibility that each
of these religious dimensions may moderate the association between stressful events and sleep.
Although a small body of literature has linked religious factors with sleep outcomes in diverse
samples (for a recent review, see Hill, DeAngelis, and Ellison 2018), the possible stress-buffering
effects of religion on sleep appear to have been addressed only once, and only among a sample of
active-duty U.S. military personnel (White et al. 2018). In what follows, we explain why certain
religious practices and cognitions should buffer the effects of major life events on sleep outcomes
in the general U.S. population.
Organizational Religious Involvement
Why might organizational religious involvement—speciﬁcally attendance at worship
services—mitigate the effects of stressful life events on sleep quality? Although this issue has
rarely been explored in empirical research on sleep, the broader literature on religion and well-
being suggests several plausible mechanisms. First, religious congregations bring together indi-
viduals who share common beliefs, values, and interests on a regular basis to engage in collective
worship activities and rituals to which they ascribe sacred signiﬁcance (Ellison and George 1994;
Idler et al. 2009). These joint activities tend to build solidarity among co-religionists and con-
tribute to a shared sense of meaning and purpose. Interactions with co-religionists may therefore
promote positive cognitions and emotions, thus diverting attention from personal problems and
challenges (e.g., Krause 2009). Sermons and other formal religious communications may also
reinforce the plausibility of religious belief systems, in turn promoting subjective certainty in
the validity of one’s religious worldview (Ellison 1991; Sethi and Seligman 1993). This may
ultimately provide insights regarding (1) how to effectively manage negative emotions like grief
or anger in the wake of stressful events and (2) how to remedy stressful circumstances.
Second, church members often provide much needed assistance to their fellows, including
tangible (i.e., goods, services, or money) and socioemotional (i.e., love and compassion) support
(Krause 2008). Religious organizations also provide spiritual support through one-on-one inter-
actions and small group experiences. In the wake of stressful events, co-religionists may assist
each other with realizing the precepts of their faith and applying spiritual insights to the inter-
pretation of stressful events (Krause 2008). Church-based support can be especially comforting
for recipients to the extent they share common deﬁnitions of speciﬁc stressors and invoke shared
religious discourses with co-religionists (Ellison and George 1994; Krause 2006a). Moreover,
church-based support may promote the use of positive religious coping practices (Krause et al.
4JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
2001), which have been linked with better psychosocial adjustment to stressful events and con-
ditions (Ano and Vasconcelles 2005; Pargament 1997). For all of these reasons, it is plausible
that regular churchgoers may experience less agitation in the wake of negative life events, and,
ultimately, better quality sleep. This leads to our second hypothesis:
H2: Religious attendance will attenuate the association between recent life events and sleep
Nonorganizational Religious Practices
There are also several reasons to expect nonorganizational religious practices to buffer the
impact of stressful events on sleep quality. For one, private prayer may be especially important
in assisting people who are dealing with major life events. Brieﬂy, individuals build relationships
with a (perceived) divine other in much the same way they construct concrete social bonds,
namely, through interpersonal interaction and conversation (Pollner 1989). Although prayer is
sometimes petitionary or contemplative (Poloma and Gallup 1991), it is more often colloquial,
as believers tend to engage in an ongoing conversation with a divine other (e.g., God, Jesus) who
they view as an intimate friend (Cerulo and Barra 2008; Sharp 2010). Believers can gain comfort
and solace through regular conversations with God, as well as an underlying sense of security
from having a beneﬁcent, all-powerful conﬁdant who is available at any time for consultation
(Pollner 1989; Sharp 2010).
Interactions with God may also be augmented by scriptural study, which can facilitate a
process of religious role-taking and deepen conversational prayer (Pollner 1989). By reading the
Bible or other religious texts, believers gain a richer understanding of how God has interacted with
people throughout history, what God wants and expects from them, and what God promises to
the faithful (Wikstr¨
om 1987). All of this may be particularly important for dealing with stressful
life events. For example, many persons can cite speciﬁc scriptural passages from which they
derive great comfort in times of distress. Akin to regular prayer, distressed persons may ﬁnd
hope and relief through frequent Bible reading, easing cognitive arousal in the face of major life
events and ultimately facilitating sound sleep. Although few studies have focused speciﬁcally on
the mental or physical health implications of Bible study (or other scripture reading), one recent
study showed that frequent Bible study moderated the association between stressful life events
and psychological well-being among U.S. adults (Krause and Pargament 2018). This leads to our
H3: Private or nonorganizational religious practices (i.e., prayer and scripture reading) will
attenuate the association between stressful life events and sleep quality.
The Stress-Buffering Role of Religious Cognitions
Until recently, the possible mental and physical health implications of religious cognitions
and beliefs have received short shrift from empirical researchers (Park 2017; Schieman, Bierman,
and Ellison 2013). Over the past two decades, however, religion-health scholars have shown
new interest in religious beliefs (Flannelly 2017; Schieman, Bierman, and Ellison 2013). How
and why might religious beliefs moderate the association between stressful life events and sleep
outcomes? One important line of recent work offers a valuable insight: religiosity in general, and
religious cognitions in particular, are associated with positive reappraisal coping or the tendency
to reinterpret potentially troubling events and conditions in less threatening terms (DeAngelis
and Ellison 2017; Park 2017; Vishkin et al. 2016).
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 5
Beliefs About God
One important direction in this area has centered on beliefs about God. For example, in-
vestigators have examined the apparent psychosocial advantages associated with distinct types
of God imagery (Bradshaw, Ellison, and Flannelly 2008; Flannelly 2017; Schieman et al. 2017;
Stroope, Draper, and Whitehead 2013), as well as perceptions of divine control (DeAngelis 2018;
DeAngelis and Ellison 2017; Krause 2005; Schieman, Pudrovska, and Milkie 2005; Schieman
et al. 2006; Schieman, Bierman, and Ellison 2010). More recently, scholarly attention has shifted
increasingly toward believers’ attachments to God, namely, the nature and quality of their re-
lationships with the divine (Granqvist, Mikulciner, and Shaver 2010; Kirkpatrick 2005). This
work draws inspiration from a broader literature on attachment theory, which posits that early life
relationships with caregivers can shape subsequent relations with romantic partners, friends, and
even God, and that different attachment styles can have implications for psychosocial well-being
Studies have shown that believers who perceive God as a reliable source of assistance and
comfort tend to enjoy a wide array of salutary mental and physical health outcomes, while those
who perceive God as detached and unreliable tend to experience adverse outcomes (Bradshaw
and Kent 2018; Bradshaw, Ellison, and Marcum 2010; Ellison et al. 2014). According to the
limited body of work thus far, attachment to God also moderates the associations between
stressful events and psychological well-being (Ellison et al. 2012). It is likely that individuals
with a secure attachment to God feel conﬁdent that He will listen and respond to their needs and
concerns at any time (Sharp 2010). Having such a reliable and powerful ally can help believers
maintain positive outlooks in the face of major life events (DeAngelis and Ellison 2017), which
may ultimately surface as improved sleep quality. This suggests the following study hypothesis:
H4: Secure attachment to God will attenuate the association between stressful life events and
Beliefs About the Afterlife
Another important strand of research on religious cognitions involves belief in (or about)
the afterlife, which has only recently reached the agenda of health scholars (Schieman, Bierman,
and Ellison 2013). Those persons who hold favorable views of the afterlife—and their place in
it—tend to enjoy better psychosocial functioning and lower levels of psychopathology (Flannelly
et al. 2008). In particular, individuals who feel assured of spiritual salvation and eternal life
(e.g., Heaven) may have a different perspective on potentially stressful events and conditions in
this world (Ellison et al. 2001). If, as many of the faithful believe, material existence is really a
prelude to a blissful union with God, and perhaps reunion with loved ones, then the anxieties and
sorrows of this world should be less pronounced. At most, stressful events are temporary hurdles
or speed bumps along the road to eternal salvation. Put another way, one who believes that he
or she is spiritually saved and bound for Heaven might not “sweat the small stuff”—and viewed
from an eternal perspective, most earthly problems may be perceived as small. It is reasonable
to expect that this mindset may result in feelings of peace and reassurance despite negative life
events, which could reduce cognitive arousal and rumination and promote higher quality sleep.
Although this proposition has never been assessed empirically, studies have linked afterlife beliefs
with an array of salutary psychosocial outcomes (Flannelly et al. 2006; Schieman, Bierman, and
Ellison 2013). Furthermore, persons who believe in an afterlife have been shown to experience
fewer undesirable consequences from negative life events and other stressors (Bradshaw and
Ellison 2010; Ellison et al., 2001, 2009). Taken together, the foregoing discussion leads us to the
following study hypothesis:
H5: Assurance of spiritual salvation will attenuate the association between stressful life
events and sleep quality.
6JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Depressive Symptoms as Mediators
In their recent review of the religion-sleep connection, Hill, DeAngelis, and Ellison (2018)
lament the fact that few studies have ever explored the pathways (mediators) via which religious
factors may shape sleep outcomes. Indeed, with only a few exceptions (Ellison et al. 2011;
Krause et al. 2017), studies in this nascent area of research have focused primarily on direct
associations between religion and sleep. Hill, DeAngelis, and Ellison (2018:329) also argue that
“once we begin to consider subgroup variations in the effects of religious involvement, empirical
explanations for these patterns should also be formally tested (i.e., mediated moderation).”
Here, we focus on depressive symptoms as a potential explanatory mechanism. To be sure, the
association between sleep and psychiatric disorders is complex and almost certainly bidirectional
(Krystal 2012; Riemann, Berger, and Voderholzer 2001). Nevertheless, depression is an especially
plausible mediator of the religion–sleep relationships for two reasons (Cutler 2018; Nutt, Wilson,
and Paterson 2008). First, depression has shown to degrade sleep quality. Indeed, insomnia
and related sleep problems are widely considered to be symptoms of depression (e.g., Lin and
Stevens 2014; Radloff 1977). According to recent research, depression can adversely affect sleep
architecture by reducing sleep efﬁciency, increasing the time required to fall asleep, and leading
to more frequent and lengthier periods of wakefulness throughout the night (Nutt, Wilson, and
Paterson 2008). Moreover, depression causes less slow-wave sleep and also alters REM sleep
through (1) shorter REM latency, (2) increased duration of the ﬁrst REM period, and (3) increased
number of eye movements (REM density) (Cutler 2018; Nutt, Wilson, and Paterson 2008). In
sum, depression can undermine both the quantity and quality of sleep, which is why impaired
sleep is typically recognized as a common symptom of depression.
Second, a large and growing body of research reveals that multiple dimensions of religiosity—
including those discussed above—are positively associated with psychological well-being and
inversely related to major depression (Maselko, Gilman, and Buka 2009) and depressive symp-
toms (Schieman, Bierman, and Ellison 2013), especially for persons dealing with stress (Smith,
McCullough, and Poll 2003). For example, studies have found that religious attendance mod-
erates the association between stressors (e.g., ﬁnancial problems, poor health) and symptoms
of depression and distress (Bradshaw and Ellison 2010; Strawbridge et al. 1998; Williams et al.
1991). Private religious practices such as prayer and scripture reading can also assist with emotion
management and can moderate the links between stressors and depressive symptoms and other
psychological problems (Krause and Pargament 2018; Sharp 2010). A modest but growing liter-
ature has shown that religious cognitions—secure attachment to God, and belief in a favorable
afterlife—also moderate the association between stressors and depressive symptoms (Bradshaw,
Ellison, and Marcum 2010; Ellison et al., 2009, 2012). Taken together, these arguments suggest
a ﬁnal hypothesis:
H6: Any attenuation of the association between stressful life events and sleep quality will
at least be partially mediated or explained by lower levels of depressive symptoms
We tested our hypotheses with cross-sectional data from the 2017 Baylor Religion Survey.
This study was ﬁelded using a self-administered pen and paper questionnaire with mail-based
collection only. The sample was selected using ABS (address-based sample) methodology based
on a simple stratiﬁed sample design. In particular, the target population (U.S. adults) was stratiﬁed
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 7
by race-ethnicity and age group to ensure minimum coverage of minority subpopulations. Gallup
distributed questionnaires with an invitation letter, return envelope, and $1 USD cash incentive. A
total of 1,501 surveys were completed. The ﬁnal sample is comparable to the 2016 General Social
Survey on a number of key demographics, including religious and political characteristics, age,
gender, education levels, and marital status (see http://www.baylor.edu/baylorreligionsurvey).
Past-month sleep quality. Respondents answered the following three questions regarding
their sleep patterns during the past month: (1) their average duration of sleep each night; (2)
how often they had trouble falling asleep (reverse-coded); and (3) how often they felt rested
in the morning. Sleep duration was coded in number of hours, while trouble falling asleep and
feeling rested in the morning were coded as ordinal measures (1 =never, 4 =most of the time).
We standardized the three items before creating a mean index of past-month sleep quality, with
higher scores reﬂecting improved quality of sleep (alpha =.55). Principal component analysis
conﬁrmed all three standardized items load onto a single component with an eigenvalue of 1.59
and eigenvectors ranging from .54 to .60.
Recent life events. Respondents were asked if any of the following 10 major events happened
to them in the past year: (1) had a child, (2) got married, (3) got divorced/separated, (4) experienced
a death of a loved one, (5) got a new job/promotion, (6) lost a job, (7) got a long-term illness
or injury, (8) moved, (9) had house foreclosed, and (10) failed at something important to them.
Response choices were coded 0 (no) and 1 (yes). Responses were summed to create a checklist
inventory of recent life events. Because very few respondents (4 percent) reported more than
three recent life events, these outlying scores were collapsed into a single category of three or
Religious involvement. Organizational religious involvement was gauged with a single-item
measure of religious attendance that asked respondents: “How often do you attend religious
services at a place of worship?” Response categories ranged from 0 (never) to 7 (several times a
week). We measured nonorganizational or private religiousness with a two-item additive index of
frequency of prayer and scripture reading (alpha =.76). Respondents answered how often they
spent time alone (outside of religious services) either praying or reading “the Bible, Koran, Torah,
or other sacred book.” Responses for frequency of prayer ranged from 0 (never) to 5 (several
times a day). Frequency of reading scripture had response categories ranging from 0 (never) to 8
(several times a week or more).
Religious cognitions. We assessed religious cognitions with two separate measures for secure
attachment to God and assurance of salvation. To measure secure attachment to God, respondents
were ﬁrst asked: “Which one statement comes closest to your personal beliefs about God?”
Those who answered “I do not believe in God” were not asked any follow-up questions about
their relations with God (n=126). To account for these respondents in our analyses, all models
include a dummy variable coded 1 =“does not believe in God” and 0 =“believes in God.”
Secure attachment to God consisted of the following four items (1 =strongly disagree, 4 =
strongly agree): (1) God seems impersonal to me (reverse-coded); (2) God seems to have little or
no interest in my personal problems (reverse-coded); (3) God knows when I need support; and
(4) I feel that God is generally responsive to me. Responses were averaged, with higher scores
reﬂecting a more secure attachment to God (alpha =.88). Finally, assurance of salvation was
measured with a single item that asked respondents: “How certain are you that you will get into
Heaven?” Original response categories were ordinal and ranged from 1 (very certain) to 5 (not
at all certain), with additional responses for “I don’t know” and “I don’t believe in Heaven.” To
gauge assurance in personal salvation, we dummy-coded this measure such that 1 =very certain
and 0 =not very certain.
8JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Depressive symptoms. Depressive symptoms were measured with a 10-item version of the
Center for Epidemiological Studies Depression (CESD) index (Radloff 1977). Respondents were
asked how often in the past week they (1) were bothered by things that usually didn’t bother
them; (2) could not shake off the blues; (3) felt just as good as other people (reverse-coded); (4)
had trouble keeping their mind on what they were doing; (5) felt depressed; (6) felt too tired to do
things; (7) felt happy (reverse-coded); (8) enjoyed life (reverse-coded); (9) felt sad; and (10) felt
that people disliked them. Response choices ranged from 1 (never) to 4 (most or all of the time).
Responses to the 10 items were averaged to create a mean index, with higher scores indicating
greater depressive symptoms (alpha =.85).
Control variables. Models also controlled for religious afﬁliation (conservative Protestant,
mainline Protestant, black Protestant, Catholic, other, none), gender (male =0, female =1),
age (in years), race/ethnicity (white, black, other), marital status (single =0, married =1),
education (ordinal, 8th grade or less =1, postgraduate =9), employment status (unemployed
=0, employed =1), household income (ordinal, $10,000 or less =1, greater than $150,000 =
7), and residential status (large city, suburb, small city or town, rural area). Weighted descriptive
statistics of study variables are reported in Table 1.
To assess hypotheses 1 through 5, Table 2 reports a series of ﬁve ordinary least squares
(OLS) regression models that regress past-month sleep quality on recent life events, measures of
religious involvement, and their interactions. Model 1 reports estimated net effects of life events
and focal religion variables on sleep quality. Models 2 through 5 introduce separate interaction
terms between life events and religious attendance, private religious practices, secure attachment
to God, and assurance of salvation, respectively. Continuous variables are mean-centered be-
fore estimating interaction terms. All analyses adjust for control variables and poststratiﬁcation
weighting to better reﬂect the population of U.S. adults. Coefﬁcients for control variables are
excluded to conserve space (full models available upon request).
There are two general approaches to mediation models and we use both of these to test our
mediated moderation hypothesis. The ﬁrst is the “coefﬁcient change” approach, which assesses
the change in the coefﬁcient for the focal predictor before and after the mediator variable is added
to the regression equation. This approach is tested with the Clogg statistic (Clogg, Petkova, and
Haritou 1995). In this case, we assess the change in the coefﬁcients for our focal interaction terms
before and after controlling for depressive symptoms. The second is the “indirect effect” approach,
which assesses the statistical signiﬁcance of the product of two coefﬁcients: the coefﬁcient for the
effect of the focal predictor on the mediator and the coefﬁcient for the effect of the mediator on
the outcome. This approach is tested with the Sobel statistic (Sobel 1982). This latter approach
requires information for each link in the proposed causal process, namely, the product of the
coefﬁcient for the interaction term in the prediction of depressive symptoms and the coefﬁcient
for the effect of depressive symptoms on sleep. The Sobel test is a useful comparison because it
is considered to be a conservative mediation test (Hayes 2013).
To facilitate the presentation of complex mediated moderation analyses, we depict main
ﬁndings as linear prediction graphs (Figures 1 and 2). Both ﬁgures are constituted by two
separate graphs depicting sleep quality scores (y-axis) as a function of recent life events (x-axis)
and the focal religious measure. The graph on the left (Model A) shows the interaction term
before controlling for depressive symptoms, while the graph on the right (Model B) shows the
same interaction after controlling for depressive symptoms. Accompanying these ﬁgures is a
table of estimated marginal effects on linear predictions of sleep quality (Table 4). This table
reports linear associations between recent life events and sleep quality (1) while holding the focal
religious measure at certain speciﬁed values, and (2) before and after controlling for depressive
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 9
Table 1: Weighted descriptive statistics: 2017 Baylor Religion Survey (n=1,410)
Range Mean/Proportion SD
Sleep quality −2.83–2.17 −.46 .75
Recent life events 0–3 1.13 1.04
Religious attendance 0–7 3.03 2.56
Private religiousness 0–12 5.25 4.26
Secure attachment to God 1–4 3.10 .80
Assurance of salvation 0–1 .28
Depressive symptoms 1–4 1.89 .52
Age 18–98 49.16 17.65
Male (reference) 0–1 .47
Female 0–1 .53
White (reference) 0–1 .76
Black 0–1 .11
Other race 0–1 .13
Single (reference) 0–1 .49
Married 0–1 .51
Education 1–9 5.18 2.28
Unemployed (reference) 0–1 .33
Employed 0–1 .67
Household income 1–7 4.25 1.74
Large city (reference) 0–1 .25
Suburb 0–1 .29
Small city/town 0–1 .33
Rural 0–1 .13
Conservative Protestant (reference) 0–1 .28
Mainline Protestant 0–1 .12
Black Protestant 0–1 .07
Catholic 0–1 .25
Other religion 0–1 .09
No religious afﬁliation 0–1 .19
Believes in God (reference) 0–1 .90
Does not believe in God 0–1 .10
symptoms. Table 4 is basically a numerical representation of Figures 1 and 2, but with additional
tests of statistical signiﬁcance accompanying each interaction term slope.
Finally, the following variables had missing data: sleep quality (n=78), recent life events
(n=106), depressive symptoms (n=98), religious attendance (n=56), prayer (n=65), scripture
reading (n=24), religious afﬁliation (n=46), secure attachment to God (n=47), assurance
of salvation (n=30), gender (n=34), age (n=99), race (n=75), marital status (n=29),
education (n=32), employment status (n=85), household income (n=84), and residential
status (n=44). With the exception of our dependent variable (sleep quality), all analyses replaced
these missing values with 25 iterations of multiple imputation by chained equation (Johnson and
Young 2011). Main ﬁndings were comparable before and after imputation.
10 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Table 2: Ordinary least squares (OLS) regression models predicting past-month sleep quality
Model 1 Model 2 Model 3 Model 4 Model 5
Recent life events −.082** −.074** −.078** −.076** −.113***
(.027) (.027) (.028) (.027) (.032)
Religious attendance .009 .009 .010 .011 .010
(.016) (.016) (.016) (.016) (.016)
Private religiousness −.005 −.006 −.006 −.006 −.005
(.011) (.011) (.011) (.011) (.011)
Secure attachment to God .062 .058 .060 .052 .062
(.054) (.054) (.055) (.053) (.054)
Assurance of salvation .104 .111 .111 .111 .098
(.063) (.063) (.062) (.062) (.062)
Interactions [Recent life events ✕...]
Religious attendance – .017 – – –
– (.010) – – –
Private religiousness – – .009 – –
– – (.006) – –
Secure attachment to God – – – .049*–
– – – (.022) –
Assurance of salvation – – – – .122*
– – – – (.053)
Constant −.598*−.637** −.709** −.509*−.680**
Adjusted R2.074 .076 .075 .080 .078
Notes: Unstandardized coefﬁcients reported with robust standard errors in parentheses. All models adjust for probability
weighting and control variables.
*p<.05; **p<.01; *** p<.001 (two-tailed).
Life events by attachment to God on sleep quality: Before (Model A) and after (Model B)
controlling for depressive symptoms
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 11
Life events by assurance of salvation on sleep quality: Before (Model A) and after (Model B)
controlling for depressive symptoms
Table 2 reports results from OLS regression models predicting past-month sleep quality. In
Model 1, the number of recent life events was inversely associated with sleep quality and this
association was statistically signiﬁcant net of control variables (b=–.082, p<.01). Moreover,
the interaction terms in Models 2 and 3 were statistically insigniﬁcant, indicating the association
between recent life events and sleep quality did not vary according to levels of religious attendance
or private religiousness. Nonetheless, both interaction terms in Models 4 and 5 were positive and
statistically signiﬁcant (p<.05), indicating that secure attachment to God and assurance of
salvation attenuated the negative association between recent life events and sleep quality.
Table 3 reports results from OLS regression models testing for mediated moderation. Models
1a and 2a are identical to Models 4 and 5 in Table 2, serving only as contrasts to Models 1b
and 2b, which additionally control for depressive symptoms. We found that the interaction terms
between recent life events and religious cognitions were no longer statistically signiﬁcant after
controlling for depressive symptoms. Moreover, the Clogg statistics were statistically signiﬁcant
at conventional levels for the interactions with secure attachment to God (t=2.28, df =1,407,
p<.05) and assurance of salvation (t=2.28, df =1,407, p<.05), indicating a signiﬁcant
reduction in the magnitude of the interaction terms after accounting for symptoms of depression.
The Sobel statistic was marginally signiﬁcant for secure attachment to God (z=1.73, se =.01,
p<.10) and statistically signiﬁcant at conventional levels for assurance of salvation (z=
2.11, se =.02, p<.05), indicating a signiﬁcant indirect effect of the interaction term on
sleep quality through depressive symptoms. These results support our mediated moderation
hypothesis. In other words, both interactions were fully mediated or explained by depressive
Figures 1 and 2 provide a clearer picture of these results. In both ﬁgures, the graphs on the
left (Model A) show that respondents who scored high on secure attachment to God, or who were
very certain they were going to Heaven when they die, tended to report little or no difference in
sleep quality as their number of recent life events increased. On the other hand, respondents who
were less secure in their relationship with God, or who expressed doubts about their prospects of
a blissful afterlife, reported signiﬁcantly steeper declines in sleep quality as a function of recent
12 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Table 3: Ordinary least squares (OLS) regression models testing for mediated moderation (n=
Model 1a Model 1b aࣔb Model 2a Model 2b aࣔb
Recent life events −.076** −.031 −.113*** −.052
(.027) (.025) (.032) (.029)
Secure attachment to God .052 −.014 .062 −.008
(.053) (.045) (.054) (.045)
Assurance of salvation .111 .055 .098 .048
(.062) (.055) (.062) (.055)
Depressive symptoms – −.606*** –−.607***
– (.050) – (.050)
Interactions [Recent life events ✕...]
Secure attachment to God .049*.030 †––
(.022) (.020) – –
Assurance of salvation – – .122*.069 †
– – (.053) (.045)
Constant −.509*.715*** −.680** .746**
Adjusted R2.080 .236 .078 .235
RMSE .717 .654 .718 .654
Notes: Unstandardized coefﬁcients reported with robust standard errors in parentheses. All models adjust for probability
weighting, religious attendance, private religiousness, and control variables.
*p<.05; **p<.01; *** p<.001 (two-tailed).
“aࣔb” indicates Clogg tests for mediation.
†Indicates a statistically signiﬁcant (p<.05) difference between interaction term coefﬁcients after adjusting for depressive
Table 4: Marginal effects on linear predictions of sleep quality (n=1,410)
Variables Model 1a Model 1b Model 2a Model 2b
Recent life events:
@ Attachment to God =−1SD −.133*** −.065*––
@ Attachment to God =Mean −.077** −.031 – –
@ Attachment to God =+1SD −.020 .004 – –
Recent life events:
@ Assurance of salvation =0– –−.113*** −.052
@ Assurance of salvation =1 – – .009 .017
Notes: Based on models from Table 3. Model A =excluding depressive symptoms. Model B =controlling for depressive
*p<.05; **p<.01; *** p<.001 (two-tailed).
life events (see Table 4). Finally, the graphs on the right (Model B) show that the differences in
slopes become modest after controlling for depressive symptoms.
DISCUSSION AND CONCLUSION
Inadequate sleep has been linked to an array of negative outcomes, including impaired social
and occupational functioning and risk of premature death (Davies et al. 2014; Engle-Friedman
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 13
et al. 2003; Grandner et al. 2010). Although many studies have explored social inﬂuences on sleep,
the role of religion has received only limited attention to date. Our study augmented this literature
by examining (1) whether multiple dimensions of religious involvement buffered the link between
stressful life events and poor sleep quality, and (2) whether depressive symptoms mediated any
observed stress-buffering patterns. Relevant hypotheses were evaluated using data collected from
a recent national sample of U.S. adults. Several particular ﬁndings deserve discussion.
First, consistent with a number of previous studies, we found that the number of stressful
life events experienced during the year prior to the interview was inversely associated with sleep
quality (Healey et al. 1981; White et al. 2018; Williamson et al. 1995). For many individuals,
stressful events provoke a repetitive analysis of the causes and implications of events, resulting in
high levels of cognitive arousal and rumination (Guastella and Moulds 2007). In the aftermath of
stressful life events, individuals may also anticipate difﬁculty falling or staying asleep, and may
develop anxiety about sleep problems and their effects on daytime functioning (Harvey 2002).
This anxiety may further deepen the effects of stressful events on sleep.
Did religion moderate the deleterious effects of stressful life events on sleep? According
to our results, the answer depends on the dimension of religious involvement being considered.
Although previous studies have documented salutary associations between sleep outcomes and
organizational and nonorganizational religious practices, we found no evidence that either of these
dimensions of religious involvement mitigated the harmful consequences of stressful events on
sleep. Our ﬁndings were therefore largely consistent with White et al.’s (2018) aforementioned
study of U.S. military personnel, which also failed to uncover any stress-buffering effects of
religious attendance on sleep quality. On the other hand, White et al. did ﬁnd that the extent to
which soldiers relied on their religious/spiritual beliefs in their daily decision making signiﬁcantly
buffered the effects of traumatic combat experiences on sleep disturbance. Thus, an emergent
theme appears to be that religious/spiritual cognitions are most effective for buffering the impact
of stressors on sleep.
There is a possibility that this verdict could change with access to more ﬁne-grained measures
of religious practices, particularly measures that account for the speciﬁc contents and motiva-
tions behind each practice (see DeAngelis, Bartkowski, and Xu 2018). For example, the role of
organizational religiosity might become clearer with speciﬁc measures of church-based support.
In particular, it would be useful to distinguish between interactions that (1) make church mem-
bers feel loved and cared for (i.e., socioemotional support), and (2) assist people in applying
religious/spiritual insights to their daily lives and problems (i.e., spiritual support;) (Krause 2008,
2009). These and other distinct facets of congregational life could help soothe people who are
dealing with stressful events, perhaps culminating in more positive sleep outcomes (Idler et al.
2009). The same is true for frequency of prayer and scripture reading. We might learn more with
detailed measures of prayer types and expectancies—namely, whether prayer is intended to be
colloquial, meditative, petitionary, or ritualistic—as well as with measures of domain-speciﬁc
scripture readings and practices DeAngelis et al. 2018; Ellison et al. 2014; Poloma and Gallup
1991). All of these are promising avenues for further research. For now, though, the mere fre-
quency of praying and reading scripture appears to be inconsequential for moderating the effects
of stressful events on sleep.
In contrast to the negligible role of religious practices, two distinct religious cognitions—
secure attachment to God and assurance of salvation—did buffer the association between stressful
life events and sleep quality. These ﬁndings resonate with an emerging body of research on the
role of religious beliefs for health and well-being (Flannelly 2017; Schieman, Bierman, and
Ellison 2013). As Park (2017) has pointed out, religious cognitions can shape interpretations
of potential stressors, thereby inﬂuencing whether or to what extent challenging circumstances
provoke distress. Certain religious beliefs may therefore aid the faithful in evaluating potentially
stressful events in terms that are less threatening to their core perceptions of self and reality
(DeAngelis 2018; DeAngelis and Ellison 2017; Vishkin et al. 2016).
14 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Why should perceptions of a secure attachment to God or the conviction that one is spiritually
saved promote better sleep in the face of stressful events? For one, belief in a loving and
supportive deity can act as a “force multiplier” that facilitates productive coping skills (Bradshaw
and Kent 2018; Bradshaw, Ellison, and Marcum 2010). That is, perceiving God as all-powerful
and unconditionally loving can provide a secure base from which to engage with the world, and
ultimately curb existential uncertainties stemming from stressful or traumatic events (Granqvist,
Mikulciner, and Shaver 2010; Kirkpatrick 2005). Believers may be unable to comprehend why
misfortune has befallen them, but they may nevertheless sleep better at night knowing that the
universe is under the watchful eye of a deity who, at the end of the day, remains deeply concerned
with the well-being of the world and its inhabitants (DeAngelis and Ellison 2017). For similar
reasons, people who are convinced they are destined for a blissful afterlife consisting of union
with God, and perhaps reunion with loved ones, should also feel less troubled by worldly problems
(Flannelly et al. 2006, 2008). These persons may be less inclined to derive their sense of self
and purpose from material achievements or other earthly activities and statuses. Major life events
involving loss or other fundamental changes are thus likely to be perceived as less threatening
(Bradshaw and Ellison 2010; Ellison et al. 2009), resulting in diminished cognitive arousal and
better sleep in the wake of such events.
Another important contribution of this study is our examination of mediators via which
religion might buffer the effects of stressful events on sleep quality. A recent review of the religion-
sleep connection (Hill et al. 2018) identiﬁed mental health as an important potential mediator.
Depression has been shown to increase the risk of subsequent insomnia and other forms of sleep
disruption (Cutler 2018; Nutt, Wilson, and Paterson 2008), such that sleep problems are often
regarded as symptoms of depression (Lin and Stevens 2014). Both stressful events and religion
have also been associated with depression independently and interactively. Speciﬁcally, multiple
dimensions of religious involvement have shown to moderate associations between stressors and
depressive symptoms (e.g., Ellison et al. 2009, 2012; Smith, McCullough, and Poll 2003). To our
knowledge, however, this is the ﬁrst empirical study to specify and test a mediated moderation
model that provides an explanatory mechanism of the stress-buffering role of religious cognitions.
Although the present study has advanced the nascent religion-sleep literature, several issues
remain unaddressed. First, additional religious domains should be explored in the future, includ-
ing: (1) speciﬁc aspects of congregational support practices, including spiritual support (Krause
2008); (2) diverse styles and practices of prayer and scriptural study (DeAngelis et al. 2018;
Ellison et al. 2014; Poloma and Gallup 1991); (3) religious coping styles and practices (Ano and
Vasconcelles 2005; Pargament 1997); and (4) religiously inspired character strengths or virtues,
such as hope, gratitude, and forgiveness (Krause 2006b; Krause and Ellison 2003; Krause et al.
Second, religion-sleep researchers should explore other complex and contingent associations.
Numerous studies have shown that religious factors tend to confer particular health beneﬁts
among members of marginalized social groups, such as racial-ethnic minorities and persons of
low socioeconomic status (Bradshaw and Ellison 2010; Schieman et al. 2006, 2010). Religion
and health scholars have also begun to develop and test moderated moderation hypotheses (i.e.,
three-way interactions), uncovering contexts in which the stress-buffering effects of religion
are ampliﬁed among certain disadvantaged segments of the population, such as the elderly
and undereducated (DeAngelis and Ellison 2018; Jung 2018). Future studies may beneﬁt from
exploring whether similar complex patterns apply to sleep outcomes.
Third, a growing literature has linked “spiritual struggles” with negative health outcomes,
although few of these studies have focused on sleep (for an exception, see Ellison et al. 2011).
Such struggles are typically of three types: (1) divine or troubled relationships with God; (2)
intrapsychic or chronic and unresolved religious doubting; and (3) interpersonal or negative
interactions in religious settings (Exline et al. 2011, 2014). In addition to whatever direct
associations may exist between spiritual struggles and health, such struggles have shown to
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 15
exacerbate the effects of stressors on well-being (McConnell et al. 2006). The possible links
between spiritual struggles and sleep are clearly ripe for further exploration.
This study is also characterized by several shortcomings. First, our data are cross-sectional
and thus entail the usual caveats. Although we can document associations, we cannot make
strong causal claims because we lack data on the temporal ordering among our key variables.
Future studies using longitudinal data are needed. A second limitation involves the measurement
of our focal variables. Speciﬁcally, sleep quality is a self-report measure based on three items
comprising an index of subpar reliability. Research on religion and sleep can be advanced further
with the use of more extensive and well-validated measures such as the Pittsburgh Sleep Quality
Index (Buysse et al. 1989), as well as with biological markers of sleep architecture. Moreover,
although our religious items are adequate for gauging key constructs in our study, single-item
religion measures can still suffer from low reliability. The development of multi-item measures
of scripture reading or afterlife beliefs, for instance, would be a welcome addition to work in this
ﬁeld (e.g., DeAngelis, Bartkowski, and Xu 2018).
Such limitations notwithstanding, the current study has made signiﬁcant contributions to
the religion and sleep literature. First, our study explored the role of multiple dimensions of
religion in mitigating the noxious effects of stressful life events on sleep. Although public and
private religious practices have been central to the religion-sleep literature to date, we found
that only religious cognitions—namely, secure attachment to God and assurance of salvation—
attenuated the inverse association between stressful events and sleep quality. Second, our analyses
showed that the stress-buffering role of religious cognitions was explained by lower levels of
depressive symptoms. Given the importance of sleep for many facets of health and well-being,
further research along the lines sketched above will advance our understanding of the complex
relationships among religion, stress, and other social determinants of sleep.
Ano, Gene and Erin B. Vasconcelles. 2005. Religious coping and psychological adjustment to stress: A meta-analysis.
Journal of Clinical Psychology 61:461–80.
Bradshaw, Matt and Christopher G. Ellison. 2010. Financial hardship and psychological distress: Exploring the buffering
effects of religion. Social Science and Medicine 71:196–204.
Bradshaw, Matt, Christopher G. Ellison, and Kevin J. Flannelly. 2008. Prayer, God imagery, and symptoms of psy-
chopathology. Journal for the Scientiﬁc Study of Religion 47:644–59.
Bradshaw, Matt, Christopher G. Ellison, and John P. Marcum. 2010. Attachment to God, images of god, and psychological
distress in a nationwide sample of presbyterians. International Journal for the Psychology of Religion 20:130–47.
Bradshaw, Matt and Blake Victor Kent. 2018. Prayer, attachment to God, and changes in psychological well-being in
later life. Journal of Aging and Health 30:667–91.
Buysse, Daniel J., Charles F.Reynolds III, Timothy H. Monk, Susan R. Berman, and David J. Kupfer. 1989. The Pittsburgh
Sleep Quality Index: A new instrument for psychiatric practice and research. Psychiatry Research 28:193–213.
Cerulo, Karen A. and Andrea Barra. 2008. In the name of . . .: Legitimate interactants in the dialogue of prayer. Poetics
Clogg, Clifford C., Eva Petkova, and Adamantios Haritou. 1995. Statistical methods for comparing regression coefﬁcients
between models. American Journal of Sociology 100(5):1261–93.
Cutler, Andrew J. 2018. The role of insomnia in depression and anxiety: Its impact on functioning,
treatment, and outcomes. http://www.cmeinstitute.com/Psychlopedia/Pages/depression/27arrd/sec3/section.aspx.
Davies, Sarah K., Joo Ern Ang, Victoria L. Revell, Ben Holmes, Anuska Mann, Francesca P. Robertson, et al. 2014.
Effect of sleep deprivation on the human metabolome. Proceedings of the National Academy of Sciences of the
DeAngelis, Reed T. 2018. Goal striving stress and self-concept: The moderating role of perceived divine control. Society
and Mental Health 8(2):141–56.
DeAngelis, Reed T., John P. Bartkowski, and Xiaohe Xu. 2018. Scriptural coping: An empirical test of hermeneutic
theory. Journal for the Scientiﬁc Study of Religion. Online ﬁrst. https://doi.org/10.1111/jssr.12576
DeAngelis, Reed T. and Christopher G. Ellison. 2017. Kept in his care: The role of perceived divine control in positive
reappraisal coping. Religions 8:133–48.
16 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
———. 2018. Aspiration strain and mental health: The education-contingent role of religion. Journal for the Scientiﬁc
Study of Religion 57(2):341–64.
Ellison, Christopher G. 1991. Religious involvement and subjective well-being. Journal of Health and Social Behavior
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 80:215–49.
Ellison, Christopher G., Matt Bradshaw, Kevin J. Flannelly, and Kathleen C. Galek. 2014. Prayer, attachment to God,
and symptoms of anxiety-related disorders. Sociology of Religion 75:208–33.
Ellison, Christopher G., Matt Bradshaw, Nilay Kuyel, and John P. Marcum. 2012. Attachment to God, stressful life
events, and changes in psychological distress. Review of Religious Research 53:493–511.
Ellison, Christopher G., Matt Bradshaw, Jennifer E. Storch, John P. Marcum, and Terrence D. Hill. 2011. Religious doubts
and sleep quality: Findings from a nationwide study of presbyterians. Review of Religious Research 53:119–36.
Ellison, Christopher G., Amy M. Burdette, and TerrenceD. Hill. 2009. Blessed assurance: Religion, anxiety, and tranquility
among U.S. adults. Social Science Research 38:656–67.
Ellison, Christopher G. and Linda K. George. 1994. Religious involvement, social ties, and social support in a southeastern
community. Journal for the Scientiﬁc Study of Religion 33:46–61.
Engle-Friedman, Mindy, Suzanne Reila, Ram Golan, Ana M. Ventuneac, Christine M. Davis, Angela D. Jefferson, and
Donna Major. 2003. The effect of sleep loss on next day effort. Journal of Sleep Research 12:113–24.
Exline, Julie J., Crystal L. Park, Joshua M. Smyth, and Michael P. Carey. 2011. Anger toward God: Social-cognitive
predictors, prevalence, and links with adjustment to bereavement and cancer. Journal of Personality and Social
Exline, Julie J., Kenneth I. Pargament, Joshua B. Grubbs, and Ann Marie Yali. 2014. The religious and spiritual struggles
scale: Development and initial validation. Psychology of Religion and Spirituality 6:208–22.
Flannelly, Kevin J. 2017. Religious beliefs, evolutionary psychiatry, and mental health in America: Evolutionary threat
assessment systems. New York: Springer.
Flannelly, Kevin J., Christopher G. Ellison, Harold G. Koenig, and Kathleen C. Galek. 2008. Belief in life-after-death,
psychiatric symptompatology, and cognitive theories of psychopathology. Journal of Psychology and Theology
Flannelly, Kevin J., Harold G. Koenig, Christopher G. Ellison, Neal Krause, and Kathleen C. Galek. 2006. Belief
in life-after-death and mental health: Findings from a national survey. Journal of Nervous and Mental Disease
Grandner, Michael A., Lauren Hale, Melisa Moore, and Nirav Patel. 2010. Mortality associated with short sleep duration:
The evidence, the possible mechanisms, and the future. Sleep Medicine Reviews 14:191–203.
Grandner, Michael A., Natasha J. Williams, Kristen L. Knutson, Dorothy Roberts, and Girardin Jean-Louis. 2016. Sleep
disparity, race/ ethnicity, and socioeconomic position. Sleep Medicine 18:7–18.
Granqvist, Pehr, Mario Mikulciner, and Phillip R. Shaver. 2010. Religion as attachment: Normative processes and
individual differences. Personality and Social Psychology Review 14:49–59.
Guastella, Adam J. and Michelle L. Moulds. 2007. The impact of rumination on sleep quality following a stressful life
event. Personality and Individual Differences 42:1151–62.
Hale, Lauren, Terrence D. Hill, and Amy M. Burdette. 2010. Does sleep quality mediate the association between
neighborhood disorder and self-rated physical health? Preventive Medicine 51:275–78.
Harvey, Allison G. 2002. A cognitive model of insomnia. Behaviour Research and Therapy 40:869–93.
Hayes, Andrew. 2013. Introduction to mediation, moderation, and conditional processes analysis: A regression-based
approach. New York: Guilford Press
Healey, E. Shevy, Anthony Kales, Lawrence J. Monroe, Edward O. Bixler, Katherine Chamberlin, and Constantin
Soldatos. 1981. Onset of insomnia: Role of life-stress events. Psychosomatic Medicine 43:439–51.
Hill, Terrence D., Amy M. Burdette, and Matt Bradshaw. 2016. Health and biological functioning. In Handbook on
religion and society, edited by David Yamane, pp. 11–28. New York: Springer.
Hill, Terrence D., Reed T. DeAngelis, and Christopher G. Ellison. 2018. Religious involvement as a social determinant
of sleep: An initial review and conceptual model. Sleep Health 4:325–30.
Hoch, Carolyn C., Mary Amanda Dew, Charles F. Reynolds III, Daniel J. Buysse, Peter D. Nowell, Timothy J.
Monk, Sati Mazumdar, Michael D. Borland, Jean Miewald, and David J. Kupfer. 1997. Longitudinal changes in
diary- and laboratory-based sleep measures in healthy “old old” and “young old” subjects: A three-year follow-up.
Idler, Ellen L., David A. Boulifard, Erich Labouvie, Yung Y. Chen, Tyrone J. Krause, and Richard J. Contrada. 2009.
Looking inside the black box of “attendance at services”: New measures for exploring an old dimension in religion
and health research. International Journal for the Psychology of Religion 19:1–20.
Idler, Ellen L., Marc A. Musick, Christopher G. Ellison, Linda K. George, Neal Krause, Marcia G. Ory, et al. 2003.
Measuring multiple dimensions of religion and spirituality for health research: Conceptual background and ﬁndings
from the 1998 General Social Survey. Research on Aging 25:327–65.
SLEEP QUALITY AND THE STRESS-BUFFERING ROLE OF RELIGIOUS INVOLVEMENT 17
Johnson, David R. and Rebekah Young. 2011. Toward best practices in analyzing datasets with missing data: Comparisons
and recommendations. Journal of Marriage and Family 73:926–45.
Jung, Jong Hyun. 2018. Country-level differences in the effects of ﬁnancial hardship on life satisfaction: The role of
religious context and age-contingent buffering. Society and Mental Health 8(2):123–40.
Khoramirad, Ashraf, Maryam Mousavi, Tahmineh Dadkhahtehrani, and Davoud Pourmarzi. 2015. Relationship between
sleep quality and spiritual well-being/religious activities in Muslim women with breast cancer. Journal of Religion
and Health 54(6):2276–85.
Kim, Eui-Joong and Joel E. Dimsdale. 2007. The effect of psychosocial stress on sleep: A review of polysomnographic
evidence. Behavioral Sleep Medicine 5:256–78.
Kirkpatrick, Lee A. 2005. Attachment, evolution, and the psychology of religion. New York: Guilford.
Koenig, Harold G., Dana King, and Verna B. Carson. 2012. Handbook of religion and health, 2nd ed. New York: Oxford
Kopp, Antonia T., Barbara A. Chini, Sophia M. Dimitriou, and Daniel H. Grossoehme. 2017. Body sanctiﬁcation and
sleep in adolescents with cystic ﬁbrosis: A pilot study. Journal of Religion and Health 56(5):1837–45.
Krause, Neal. 2005. God mediated control and psychological well-being in late life. Research on Aging 27:136–64.
———. 2006a. Exploring the stress-buffering effects of church-based and secular social support on self-rated health in
late life. Journal of Gerontology: Social Sciences 61B:S35–S43.
———. 2006b. Gratitude toward God, stress, and health in late life. Research on Aging 28(2):163–83.
———. 2008. Aging in the church: How social relationships affect health. West Conshohocken, PA: Templeton Press.
———. 2009. Core religious beliefs and providing support to others in late life. Mental Health, Religion, and Culture
Krause, Neal and Christopher G. Ellison. 2003. Forgiveness by God, forgiveness of others, and psychological well-being
in late life. Journal for the Scientiﬁc Study of Religion 42:77–93.
Krause, N., C. G. Ellison, B. A. Shaw, J. P. Marcum, and J. D. Boardman. 2001. Church-based social support and religious
coping. Journal for the Scientiﬁc Study of Religion 40(4):637–56.
Krause, Neal, Peter C. Hill, Robert Emmons, Kenneth I. Pargament, and Gail Ironson. 2017. Assessing the relationship
between religious involvement and health behaviors. Health Education and Behavior 44:278–84.
Krause, Neal and Kenneth I. Pargament. 2018. Reading the Bible, stressful life events, and hope: Assessing an overlooked
coping resource. Journal of Religion and Health 57:1428–39.
Krystal, Andrew D. 2012. Psychiatric disorders and sleep. Neurologic Clinics 30:1389–1413.
Lin, Steven Y. and Michael B. Stevens. 2014. The symptom cluster-based approach to individualize patient-centered
treatment for major depression. Journal of the American Board of Family Medicine 27:151–59.
Mai, Quan, Terrence D. Hill, Luis Vila-Henninger, and Michael Grandner. 2018. Employment insecurity
and sleep disturbance: Evidence from 31 European countries. Journal of Sleep Research. Online ﬁrst.
Maselko, Joanna, Stephen E. Gilman, and Stephen Buka. 2009. Religious service attendance and spiritual well-being are
differentially associated with risk of major depression. Psychological Medicine 29(6):1009–1107.
McConnell, Kelly, Kenneth I. Pargament, Christopher G. Ellison, and Kevin J. Flannelly. 2006. Examining the links
between spiritual struggles and symptoms of psychopathology in a national sample. Journal of Clinical Psychology
Mezick, Elizabeth J., Karen A. Matthews, Patrick J. Strollo, Jr., Daniel J. Buysse, Thomas W. Kamarck, Jane F. Owens,
and Steven E. Reis. 2008. Inﬂuence of race and socioeconomic status on sleep: The Pittsburgh Sleep Score project.
Psychosomatic Medicine 70:410–16.
Middlekoop, Huub, Dini A. Smilde-van de Doel, Arie Kniustingh Neven, Hilbert A.C. Kamphulsen, and Chiel P. Springer.
1996. Subjective sleep characteristics of 1,485 males and females aged 50–93: Effects of sex and age, and factors
related to self-evaluated quality of sleep. Journal of Gerontology: Medical Sciences 51A:M108–15.
Morin, Charles M., Sylvie Rodrigue, and Hans Ivers. 2003. Role of stress, arousal, and coping skills in primary insomnia.
Psychosomatic Medicine 65:259–67.
Nutt, David, Sue Wilson, and Louise Paterson. 2008. Sleep disorders as core symptoms of depression. Dialogues in
Clinical Neuroscience 10:329–36.
Pargament, Kenneth I. 1997. The psychology of religion and coping. New York: Guilford.
Park, Crystal L. 2017. Religious cognitions and well-being: A meaning perspective. In The happy mind: Cognitive
contributions to well-being, edited by Michael D. Robinson and Michael Eid, pp. 443–58. New York: Springer.
Pilcher, June J., Douglas R. Ginter, and Brigitte Sadowsky. 1997. Sleep quality versus sleep quantity: Relationships
between sleep and measures of health, well-being, and sleepiness in college students. Journal of Psychosomatic
Pollner, Melvin L. 1989. Divine relations, social relations, and well-being. Journal of Health and Social Behavior
Poloma, Margaret and George H. Gallup, Jr. 1991. Varieties of prayer: A survey report. Philadelphia: Trinity Press
18 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION
Radloff, Lenore S. 1977. The CES-D scale: A self-report depression scale for research in the general population. Applied
Psychological Measurement 1(3):385–401.
Riemann, Dieter, Mathias Berger, and Ulrich Voderholzer. 2001. Sleep and depression—Results from psychobiological
studies: An overview. Biological Psychology 57:67–103.
Schieman, Scott, Alex Bierman, and Christopher G. Ellison. 2010. Religious involvement, beliefs about God, and sense
of mattering among older adults. Journal for the Scientiﬁc Study of Religion 49:517–35.
———. 2013. Religion and mental health. In Handbook of the sociology of mental health, 2nd ed., edited by Carol S.
Aneshensel, Jo C. Phelan, and Alex Bierman, pp. 457–78. New York: Springer.
Schieman, Scott, Alex Bierman, Laura Upenieks, and Christopher G. Ellison. 2017. Love thy self? How belief in a
supportive God shapes self-esteem. Review of Religious Research 59:293–318.
Schieman, Scott, Tetyana Pudrovska, Leonard I. Pearlin, and Christopher G. Ellison. 2006. Sense of divine control and
psychological distress in late life: Moderating effects of race and socioeconomic status. Journal for the Scientiﬁc
Study of Religion 45:529–49.
Schieman, Scott, Tetyana Pudrovska, and Melissa A. Milkie. 2005. The sense of divine control and the self-concept: A
study of race difference in late life. Research on Aging 27(2):165–96.
Sethi, Sheena and Martin E. Seligman. 1993. Optimism and fundamentalism. Psychological Science 4(4):256–59.
Sharp, Shane. 2010. How does prayer help manage emotions? Social Psychology Quarterly 73:417–37.
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 129:614–36.
Sobel, Michael. 1982. Asymptotic intervals for indirect effects in structural equation models. In Sociological methodology,
edited by S. Leinhart, pp. 290–312. San Francisco: Jossey-Bass.
Strawbridge, William J., Sarah J. Shema, Richard D. 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
Stroope, Samuel, Scott Draper, and Andrew L. Whitehead. 2013. Images of a loving God and sense of meaning in life.
Social Indicators Research 111:25–44.
Vidal, Hila and Tamar Shochat. 2017. Early to bed, early to rise: Sleep perceptions, patterns and related behaviors in
ultra-orthodox Jewish adolescents in Israel. Sleep Health 3(6):458–64.
Vishkin, Allon, Yochana E. Bigman, Roni Porat, Nevin Solak, Eran Halperin, and Maya Tamir. 2016. God rest our hearts:
Religiosity and cognitive reappraisal. Emotion 16:252–62.
White, James, Xiaohe Xu, Christopher G. Ellison, Reed T. DeAngelis, and T. S. Sunil. 2018. Religion, combat casualty
exposure, and sleep disturbance in the U.S. military. Journal of Religion and Health 57(6):2362–77.
Wiks tr ¨
om, Owe. 1987. Attribution, roles, and religion: A theoretical analysis of Sunden’s role theory of religion and the
attributional approach to religious experience. Journal for the Scientiﬁc Study of Religion 26(3):390–400.
Williams, David R., David B. Larson, Robert E. Buckler, Richard C. Heckmann, and Caroline M. Pyle. 1991. Religion
and psychological distress in a community sample. Social Science and Medicine 32:1257–62.
Williamson, Ann M. and Anne-Marie Feyer. 2000. Moderate sleep deprivation produces impairments in cognitive
and motor performance equivalent to legally prescribed levels of alcohol intoxication. BMJ: Occupational and
Environmental Medicine 57:649–55.
Williamson, Douglas E., Ronald E. Dahl, Boris Birmaher, Raymond R. Goetz, Beverly Nelson, and Neal D. Ryan. 1995.
Stressful life events and EEG sleep in depressed and normal control adolescents. Behavioral Psychiatry 37:859–65.