ArticlePDF Available

Attachment to God and Psychological Distress: Evidence of a Curvilinear Relationship

Authors:

Abstract and Figures

Previous religion/spirituality (R/S) research on attachment to God and mental well‐being has relied entirely on linear models. Scholars, however, have called for more nuanced analysis of religious beliefs and dispositions relative to mental health, and several studies using a nonlinear approach have yielded fruitful results with a handful of R/S predictors. Relying on national data from the 2010 Baylor Religion Survey, this study investigates potential nonlinear associations between attachment to God and psychological distress by fitting curvilinear models of avoidant attachment to God and multiple measures of general and psychological distress. For conceptual reasons, linear models of anxious attachment are also employed. Results reveal a nonlinear relationship between avoidant/secure attachment and distress and a deleterious linear relationship between anxious attachment to God and distress. This supports the overall hypothesis that anxiety or a lack of certainty about one's relationship with the divine represents a threat to psychological well‐being.
Content may be subject to copyright.
Attachment To God and Psychological Distress: Evidence of A Curvilinear Relationship
W. MATTHEW HENDERSON
Department of Sociology and Family Studies
Union University
BLAKE VICTOR KENT
Department of Sociology and Anthropology
Westmont College
Previous religion/spirituality (R/S) research on attachment to God and mental well-being has
relied entirely on linear models. Scholars, however, have called for more nuanced analysis of
religious beliefs and dispositions relative to mental health, and several studies using a non-linear
approach have yielded fruitful results with a handful of R/S predictors. Relying on national data
from the 2010 Baylor Religion Survey, this study investigates potential non-linear associations
between attachment to God and psychological distress by fitting curvilinear models of avoidant
attachment to God and multiple measures of general and psychological distress. For conceptual
reasons, linear models of anxious attachment are also employed. Results reveal a non-linear
relationship between avoidant/secure attachment and distress and a deleterious linear relationship
between anxious attachment to God and distress. . This supports the overall hypothesis that
anxiety or a lack of certainty about one’s relationship with the divine represents a threat to
psychological well-being.
Keywords: attachment to god, psychological distress, religious beliefs, mental health
Acknowledgements: The authors would like to acknowledge Matt Bradshaw and Carson
Mencken for guidance in early drafts, as well as helpful suggestions from JSSR reviewers.
Correspondence should be addressed to Matt Henderson, Department of Sociology and Family
Studies, Union University, 1050 Union University Dr, Jackson, TN 38305. E-Mail:
mhenderson@uu.ed
***This is not the version of record. Please see final version, copyright Wiley, published in the
Journal for the Scientific Study of Religion. https://doi.org/10.1111/jssr.12767
Introduction
Religion is a central dimension of many people’s lives and a more complete
understanding of its contingent effects on mental health is valuable to both scholars and the
public (Schieman and Bierman 2011). Contrary to early critiques (Freud 1927; Leuba 1925;
Skinner 1953), the bulk of empirical findings suggest a salutary association between religious
participation and mental health (Ellison and Henderson 2011; Greenfield, Vaillant, and Marks
2009; Rosmarin and Koenig 2020). Systematic reviews demonstrate consistent associations
between multiple dimensions of religious life, such as salience and service attendance, and
desirable mental health outcomes such as lower levels of depression and distress and increased
life satisfaction and happiness (Ellison and Levin 1998; George, Ellison, and Larson 2002;
Rosmarin and Koenig 2020; Schieman, Bierman, and Ellison 2013). Religion is multi-faceted,
however, and other important aspects of religion, such as subjective perceptions of the divine,
are less studied in association with mental health (Kent 2020). This dearth contributes to an
incomplete understanding of religion and well-being (Schieman and Bierman 2011).
Attachment to God represents a promising advance in this area of inquiry (Granqvist and
Kirkpatrick 2013). This research relies on insights from Attachment Theory (Bowlby 1969) to
investigate how people’s perceptions of—and emotional dispositions towardthe divine impact
aspects of human relationships and outcomes. Extant findings suggest a straightforward linear
relationship between attachment to God and mental well-being (Bradshaw, Ellison, and Marcum
2010; Bradshaw and Kent 2018; Ellison et al. 2012, 2014; Kent, Bradshaw, and Dougherty 2016;
Kent, Bradshaw, and Uecker 2018). However, this arena is fairly fresh and needs to contend
further with the contingent nature of attachment styles indicated in reviews of the literature. For
example, religious involvement is typically associated with positive well-being, but the opposite
appears true for those whose perception of the divine is fraught (Ellison and Henderson 2011).
Moreover, a handful of studies have identified nonlinear relationships between some measures of
religion/spirituality (R/S) and mental health (e.g., Galen and Kloet 2011; Stroope et al. 2020).
These studies indicate the possibility that religion may not meaningfully associate with mental
health for religious believers who are nonetheless largely disengaged from the divine.
Our study builds on this prior work by investigating non-linear associations between
attachment to God and psychological distress. To do so, we fit curvilinear models to the
avoidance/security
1
subscale of attachment to God and six measures of general and
psychological distress. We expect to find that those on the low end of this subscale (i.e.,
avoidant) will have similarly low levels of psychological distress as those on the high end of the
scale (i.e., secure), but for differing reasons. We expect that that those somewhere in the middle
of the scale (i.e., the anxious) will evince higher levels of distress.
Additionally, we examine the linear relationship between the anxiety subscale and
psychological distress as well as the curvilinear relationship between the two subscales to
confirm that those in the middle of the avoidance/security scale are likely to be high on the
anxiety subscale.
Attachment Theory
Attachment theory examines child-caretaker bonding as a central motivator of human
behavior and a primer for future interpersonal relationships. Young children engage in
proximity-seeking behavior, drawing close to primary caregivers to feel emotionally comforted,
supported, and safe. In this capacity, caregivers provide infants with a “secure base” from which
1
In Rowatt and Kirkpatrick’s (2002) measure of attachment to God, “security” is used to refer to the opposite end of
the avoidance subscale. This is grounded in earlier work using secure, avoidant, and anxious-ambivalent as three
ideal types of attachment (Hazan and Shaver 1987). Many studies currently use a four-part typology, secure,
anxious-ambivalent, avoidant, and disorganized (these terms can vary), and in this typology, “secure” refers to those
who score low on both the avoidance and anxiety subscales. We do not use this typology but rather examine the two
subscales separately, and in conjunction (see Granqvist 2020). For lack of a better term, our use of the words
secure and “security” are meant to illustrate “non-avoidance,” or the opposite pole of the avoidance subscale.
to explore the world (Ainsworth et al. 2015). The style of attachment a child develops with the
caregiver serves as an “internal working model” (IWM), a collection of neurological, biological,
emotional, and social stimuli that coalesce to prime expectations for future relationships
(Johnson et al. 2010).
Attachment Theory proposes that IWMs influence the nature of relationships throughout
childhood and into adulthood. Researchers applying this approach have provided insight into the
dynamics of many relational contexts such as caregiving (Collins and Feeney 2000), romance
(Fraley and Shaver 2000; Hazan and Shaver 1987), friendship (Saferstein, Neimeyer, and
Hagans 2005), and employee/employer relationships (Frazier et al 2014), among others. They
have also demonstrated salutary relationships between secure attachment style and measures of
depression, distress, coping, psychological functioning, and other mental health outcomes
(Kirkpatrick and Davis 1994; Kirkpatrick and Hazan 1994; Mikulincer and Florian 1998;
Murphy and Bates 1997).
Attachment to God
The dynamics people experience in connection with God, such as feelings of closeness,
anxiety, alienation, love, care, trust, and uncertainty meet the criteria of an attachment
relationship (Granqvist and Hagekull 1999, 2003; Kirkpatrick 2004). In fact, some scholars
suggest that compared to the variability of human relationships, God represents the ultimate
attachment figure (Cicirelli 2004; Kirkpatrick 2004). The sub-field of attachment to God
analyzes how emotional connections to God are related to childhood attachments (Birgegard and
Granqvist 2004; Granqvist and Hagekull 1999, 2003; McDonald et al. 2005) which then
condition a variety of social, psychological, and organizational outcomes (Bradshaw et al. 2010;
Ellison et al. 2012, 2014; Kent 2017; Kent et al. 2016; Kent and Henderson 2017; Krause and
Hayward 2016).
Studies investigating the impact of attachment to God on well-being have found a few
consistent patterns (Bradshaw et al. 2010; Ellison et al. 2012, 2014; Rowatt and Kirkpatrick
2002). First, secure attachment is consistently associated with positive mental health outcomes,
including greater life satisfaction, optimism, and agreeableness as well as lower depressed affect,
lower anxiety, decreased psychological distress, and decreased loneliness. This pattern has been
confirmed with longitudinal data (Bradshaw and Kent 2017; Ellison et al. 2012). Second,
deleterious relationships have been observed between anxious attachment and positive affect,
distress, neuroticism, psychiatric symptoms, and distress over time. Third, while these
associations appear straightforward on their own, findings involving the interactive effects of
non-organizational practices, such as the frequency of prayer and feeling forgiven by God,
complicate this interpretation. Despite sound reasons for hypothesizing mental health benefits
vis-à-vis regular religious practice, empirical findings have been mixed: multiple interaction
models demonstrate that religious practice contributes to improved mental health outcomes for
those who are securely attached to God, but poorer outcomes for those who are insecurely
attached (Bradshaw and Kent 2017; Ellison et al. 2014; Kent et al. 2016).
These observations suggest an intriguing possibility in light of attachment theory, that of
a curvilinear (or U-shaped) association between religion and well-being. Prayer, for example,
represents a type of proximity-seeking behavior that should theoretically contribute to improved
well-being for believers since it often involves drawing close to God. However, much of the
research investigating the main effects of prayer on mental health is mixed, indicating positive,
negative, and null associations (Bradshaw et al. 2010; Ellison, Burdette, and Hill 2009).
Incorporating divine relationship quality (the context in which prayer occurs) helps parse these
findings by demonstrating that divine relationships affect these associations (Bradshaw and Kent
2017; Schieman et al. 2013). Some people believe in God yet carry no expectation of divine
involvement in their lives (Froese and Bader 2010). For them, we should not expect divine
relationship quality to be particularly salient for mental well-being since God is perceived to be
distant and relatively uninvolved with human affairs. These individuals are likely to utilize other
sources for addressing psychological health, such as personal relationships, exercise, hobbies,
etc. (Huppert 2009). People who are securely attached to God use these other sources, of course,
but in addition they find God a ready source of comfort and strength, which correlates with lower
levels of distress (Bradshaw et al. 2010). In both the avoidant and secure (relatively settled
positions) we should expect to see lower levels of distress compared to those who are unsettled
in their relationship with God in other words, those in the middle of the security/avoidance
scale. These unsettled believers desire relational connection with God, but struggle to both
receive it and consistently embrace it, as attachment anxiety is characterized by continual shifts
and uncertainty, correlating to greater distress (Ainsworth et al. 2015; Bradshaw et al. 2010).
The basic proposition that attachment anxiety is associated with psychological distress is
not novel, but to our knowledge no studies have utilized curvilinear models and the
avoidance/security subscale to investigate the question. Rather, attachment anxiety is typically
measured as a separate subscale (e.g., Rowatt and Kirkpatrick 2002). Identifying curvilinear
associations with the avoidance/security scale, as we do here, helps to characterize the subscale
as accurately as possible vis-à-vis psychological characteristics. Additionally, the task is made
relevant given that both subscales are not always fielded (e.g., several recent iterations of the
Baylor Religion Survey include only the avoidance/security subscale). This leads to the first
formal hypothesis of our study:
H1: There will be a curvilinear association between avoidant/secure attachment to God
2
and psychological distress, such that those who are avoidant or secure will experience
lower levels of distress than those in the center of the subscale.
This curvilinear hypothesis is specific to the avoidance/security subscale. The anxiety
subscale is unlikely to correspond with psychological distress in a similar fashion. This has
everything to do with the nature of the construct since the anxiety subscale directly measures
relational uncertainty with God. Anxious attachment to God should therefore be reflected in a
positive linear association with measures of psychological distress. This leads to the second
formal hypothesis:
H2: Anxious attachment to God will be positively and linearly associated with
psychological distress.
Lastly, we propose a hypothesis regarding the association between the avoidant/secure
and anxious subscales of attachment to God. There is broad consensus that avoidant/secure and
anxious attachment styles represent relatively independent constructs (Ellison et al. 2012; Rowatt
and Kirkpatrick 2002). However, our first hypothesis is predicated on the view that attachment
anxiety is reflected in the center of the avoidance/security subscale. Both subscales therefore
measure relational certainty and uncertainty, albeit in different forms. This suggests that both
attachment to God subscales represent overlapping assessments of relational uncertainty with
God. Rather than accept this at face value, we test the potential overlap by examining the
2
Measurement note: Security/avoidance is measured on a continuum with one set of questions
while anxiety is measured with a separate set of questions. See methods section for further
details.
relationship between these two measures. If the poles of the avoidant/secure subscale represent
greater relational certainty (i.e., engaged or disengaged), we should see this reflected by a
curvilinear relationship vis-à-vis anxious attachment. Verifying this relationship should
strengthen our case that these constructs overlap in capturing relational uncertainty, and that
relational uncertainty with God explains the relationship between attachment style and mental
health. Thus, the third formal hypothesis:
H3: There will be a non-linear association between avoidant/secure attachment to God
and anxious attachment, with the highest levels of anxious attachment being observed for
individuals who are at the center of the avoidant/secure distribution.
Data, Methods, and Sample Characteristics
Data come from the 2010 Baylor Religion Survey (BRS), a national random sample of
adults in the contiguous United States aged 18 years or older collected by the Gallup
Organization. Random Digit Dialing was used to contact a sample of 7,000 potential
respondents. Of these, 2,556 agreed to receive a mailed questionnaire and 1,714 were returned,
resulting in a total response rate of 24.5%. Potential non-response bias was addressed using
Groves’ (2006) “gold standard” approach, comparing key demographic means and proportions
with the 2010 General Social Survey. We found no apparent bias in the BRS toward religious
respondents; further, atheists and non-affiliates may be more accurately represented in the BRS.
3
Of the 1,714 cases, three respondents reported ages either less than 18 or beyond the
fence for statistical outliers; these were excluded from all analyses. Also, because atheists were
3
Over a dozen demographic and religion variables were compared between BRS and GSS.
Comparisons are available by request. For more information about the Baylor Religion Survey,
see Bader, Mencken, and Froese (2007).
instructed to skip the attachment to God questionnaire they were also withheld from analysis,
reducing the sample to 1,624. Other selected variables contained some missing values, but in
these cases multiple imputation was used to recover them (see Analytic Strategy).
Dependent Variables
Psychological distress was assessed using six dependent variables measuring mental
health and psychopathology among the general population. General distress was measured from
an additive scale of three Likert items, adapted from the Centre for Disease Control Health
Related Quality of Life Instrument. The first item asked respondents: “Now thinking about your
mental health, which includes stress, depression, and problems with emotions, for how many
days during the past 30 days was your mental health not good?”; the second asked During the
past 30 days, for about how many days have you felt sad, blue, or depressed?” and the third
asked During the past 30 days, for about how many days have you felt worried, tense, or
anxious?” For each item, respondents chose between the following responses: None = 0, 1-10
days = 1, 11-20 days = 2, 21-29 days = 3, and All 30 days = 4. Items were coded 0-4 and
responses were added together to provide a general distress score (α = 0.878).
The remaining five measures were constructed via scales adapted from the Diagnostic
and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Each measure relies on
three items, scaled 0-4, and summed. All items begin with the prompt “Over the past month, how
often have you…” followed by a specific symptom. Available responses were Never = 0, Rarely
= 1, Sometimes = 2, Often = 3, and Very Often = 4. Items measuring generalized anxiety
(Ellison et al. 2014; Kroenke et al. 2010) include: “…felt nervous, anxious, or on edge”; “…not
been able to stop or control worrying; and “…worried too much about different things” (α =
0.841). Items measuring social anxiety (Ellison et al. 2014; Moore and Gee 2003) include: “…
feared that you might do something embarrassing in social situations”; “… became anxious
doing things because people were watching”; and “… endured intense anxiety in social
performance situations” (α = 0.824). Items measuring paranoia (Fenigstein and Vanable 1992)
include: “… felt like you were being watched or talked about by others”; “… felt that it is not
safe to trust anyone”; “… felt that people were taking advantage of you” (α = 0.763). Items
measuring obsession (Ellison et al. 2014; Kaplan 1994) include: “… been plagued by thoughts or
images cannot get out of your from mind”; “… thought too much about things that would not
bother other people”; and “… thought too much about pointless matters” (α = 0.764). Items
measuring compulsion (Ellison et al. 2014; Kaplan 1994) include: “… felt compelled to perform
certain actions, for no justifiable reason”; “… repeated simple actions that realistically did not
need not be repeated”; and “… been afraid something terrible would happen if you did not
perform certain rituals” (α = 0.763).
Independent Variables Attachment to God
Attachment to God was assessed using Rowatt and Kirkpatrick’s (2002) nine item
multidimensional measure identifying two dimensions of attachment, secure/avoidant and
anxious. Security/avoidant was gauged by respondent agreement with six items (α = 0.914): (a)
“I have a warm relationship with God,” (b) “God knows when I need support,” (c) “I feel that
God is generally responsive to me,” (d) “God seems impersonal to me” (reverse coded), (e) “God
seems to have little or no interest in my personal problems” (reverse coded), and (f) “God seems
to have little or no interest in my personal affairs” (reverse coded). Available responses included
0=Strongly Disagree; 1=Disagree; 2=Agree, and 3=Strongly Agree. Anxious attachment (α =
0.786) was assessed from the following three items and was scaled identically to secure
attachment items: (a) “God sometimes seems responsive to my needs, but sometimes not,” (b)
“God’s reactions to me seem to be inconsistent,” (c) “God sometimes seems warm and other
times very cold to me.”
Covariates
Several religious and demographic control variables were included. Religious affiliation
was measured from a series of discrete religious tradition categories based on RELTRAD
(Dougherty, Johnson, and Polson 2007; Steensland et al. 2000). Categories included Mainline
Protestant, Black Protestant, Catholic, Other, and No Religious Tradition as a reference category.
A continuous measure of religious service attendance was also included (0=never; 1=less than
once a year; 2=once or twice a year; 3=several times a year; 4=once a month; 5=2-3 times a
month; 6=about weekly; 7=weekly; 8=several times a week). Demographic measures include
discrete measures of gender (1=male), marital status (1=married), whether the respondent is
currently living with one or more child under the age of 18 (1=yes), and full-time employment
status (1=yes). Also controlled are geographic region (East, Midwest, West, with reference
category South) and education (some college, BA, greater than BA, with reference category high
school diploma or less). Finally included were a continuous measure of age (18-100) and an
interval measure of annual household income (1=$10,000 or less, 2=$10,001 to $20,000,
3=$20,001 to $35,000, 4=35,001 to $50,000, 5=50,001 to $100,000, 6=$100,001 to $150,000
and 7 = 150,001 or more).
Table 1 presents unweighted descriptive statistics for study variables. On average,
psychological distress index values are low, ranging from 1.098 for Compulsion to 3.537 for
General Anxiety, all on a scale of 0-12. Because the BRS is a national random sample, these
values are certainly lower than they would be among samples of clinically referred patients.
Means of secure and anxious attachment to God styles are 13.20 (on a scale of 0-18) and 2.99
(on a scale of 0-9), respectively.
[Table 1 about here]
Analytic Strategy
Data analysis was performed using SAS 9.4. Preliminary diagnostic tests of normality
indicated positively skewed dependent variable distributions. To reduce potential bias, square
root transformations were performed on these variables. Initial modeling using listwise deletion
resulted in an 11% sample loss. The potential biases attributable to missing cases are well
documented (Acock 2005). Once sample restrictions were made (see above), PROC MI and MI
ANALYZE, using the Markov Chain Monte Carlo (Zhang 2003) method of data imputation,
were used to recover incomplete or missing cases. Results are based on analysis of 25 imputed
datasets and similar to results obtained through listwise deletion. Because transformed dependent
variables are continuous and normally distributed, Ordinary Least Squares regression was used
for all analyses; the data were also weighted to achieve a more representative sample of blacks
and non-white Hispanics. In non-linear analyses, key predictors were mean centered for ease of
interpretation.
Four sets of results are presented: the first presents the results of both attachment to God
measures and the singular measure of general distress. The second presents the respective results
of avoidant/secure attachment and anxious attachment and five psychological symptom scales
(general anxiety, social anxiety, paranoia, obsession, and compulsion). A final set of analyses is
included presenting the results of avoidant/secure attachment and anxious attachment. Two sets
of prediction lines are also included, one presenting the results of secure attachment and
psychological symptom scales, the other presenting results of secure attachment and anxious
attachment.
Results
Hypothesis 1 is investigated via six tests. The firstgeneral distressshows no evidence of
a curvilinear relationship for secure attachment. We present both the linear and non-linear model
of general distress in Table 2. Table 3, Models 1-5 report results of OLS regression of
psychological symptom scales on avoidant/secure attachment to God. Each of the five models
includes the lower order measure of secure attachment, along with its quadratic square. In each
model, both the lower order measure of attachment and its square term are significant. Prediction
lines from each model are presented in Figure 1, Panel A. The curves for general anxiety and
social anxiety are nearly symmetric and the respective inflection points of -4.619 and -4.225
4
are
either close or virtually equal to the midpoint of the secure attachment range (-4.203). The
prediction line and inflection point of -3.318 for paranoia, nearly a full unit past the midpoint,
predicts lower levels for those with a highly avoidant attachment style. For obsession, prediction
lines suggest the obverse: highly avoidant attachment predicts greater obsession than highly
secure attachment, with an inflection point -5.384 just over a full unit shy of the midpoint.
Finally, the prediction line and inflection point for compulsion is comparable to those for general
and social anxiety, with an inflection point of -4.705, just short of the midpoint.
[Table 2 about here]
[Table 3 about here]
[Figure 1 about here]
4
Inflection points and midpoints calculated after mean centering.
H2 also consists of six tests. In Table 2 we present the results of OLS regression of
general distress on anxious attachment, where a one-unit increase in anxious attachment style is
associated with a 0.69 increase in the square root of reported general distress. Table 3, Models 6-
10 reports results of OLS regression of psychological symptom scales on anxious attachment to
God. Anxious attachment is associated with significantly higher levels of psychological distress
for all five measures.
Table 4 presents results of OLS analysis of secure attachment to God as a predictor of
anxious attachment to God. Models 1 and 2 present linear and non-linear results, respectively.
Models also include the measure of general distress, as this could account for a significant
amount of the variance in anxious attachment (see Table 2, Model 3). Here, general distress
predicts an increase in anxious attachment style while avoidant/secure attachment predicts a
0.208-unit decrease in the square root of anxious attachment. Model 2 adds a quadratic measure
of avoidant/secure attachment. Here, both the lower order measure of secure God and its
quadratic term are significant. Figure 1, Panel B displays the prediction lines observed in both
models. The dotted line represents the linear prediction line reported in Model 1. As presented,
results suggest those with an avoidant attachment style may also be the most anxious in their
attachment to God. However, the solid line representing results from Model 2 suggest much
lower anxious attachment for those with an avoidant attachment style. Rather, the highest levels
of anxious attachment are predicted for those who appear in the middle of the avoidant/secure
continuum.
[Table 4 about here]
Discussion
While studies have generally linked religion with positive mental well-being (Ellison and
Hummer 2010; Ellison and Levin 1998; Kent et al. 2020), scholars have called for more nuanced
analysis of religious beliefs and dispositions relative to mental health outcomes (Ellison and
Hummer 2010; Schieman and Bierman 2011). The present study addresses this gap by
investigating the relationship between attachment to God and psychological distress in a national
sample. We hypothesize that greater anxiety or uncertainty in one’s attachment style is
associated with greater distress. Results overwhelmingly support our hypotheses. Curvilinear
relationships are observed between secure attachment and all five measures of psychological
distress, as well as positive linear associations between anxious attachment and all six dependent
measures of distress. The relationship between anxious attachment to God and psychological
distress appears straightforward and robust. Summarily, these results suggest that anxiety or a
lack of certainty about one’s relationship with the divine represents a threat to psychological
well-being.
Results also support our hypothesis of a curvilinear relationship between secure
attachment and anxious attachment. This finding is intriguing, given the broad consensus among
adult attachment scholars that multi-item self-report measures, like those used here, measure
relatively independent dimensions of attachment style (Ellison et al. 2012; Rowatt and
Kirkpatrick 2002). Our findings do not necessarily undermine conventional use of these
measures but do raise the possibility that mischaracterization can occur when the avoidant/secure
subscale is applied in a linear fashion, introducing “difference” between avoidant and secure
respondents that may not be present.
Higher distress among those in the middle of the secure/avoidant spectrum suggests it
may be important to revisit previous studies reporting a generally deleterious effect of avoidant
attachment to God (Bradshaw et al. 2010; Bradshaw and Kent 2017; Ellison et al. 2012, 2014;
Kirkpatrick and Shaver 1992; Kirkpatrick et al. 1999; Rowatt and Kirkpatrick 2002). The
implicit conclusion drawn from those studies is that both anxious and avoidant attachment
correlates with psychological maladjustment. From the point of view of general attachment
theory this makes intuitive sense. However, this rationale may be inappropriately applied to
one’s relationship with God as it implicitly assumes proximity to God is uniformly desirous.
Conceptions of God and the degree we should seek divine proximity vary according to religious
socialization, and a growing number simply aren’t socialized to believe a close relationship with
God is useful or desirable (Smith and Cooperman 2016). Thus, avoidant attachment to God may
reflect theological views of God as appropriately distal, rather than problematically so.
Meanwhile, findings for anxious attachment to God reinforce the conclusion that fraught
dimensions of religious practice uniquely threaten well-being (Ellison et al. 2009; Ellison and
Lee 2010; Exline 2002). One worthy line of inquiry is whether anxious attachment to God
inhibits access to the psychosocial resources that religious commitment often provides (Ellison
and George 1994; Krause et al. 2001; Levin, Chatters, and Taylor 1995). Recent studies indicate
both avoidant and anxious attachment to God styles erode commitment to religious
congregations where one would find networks of social support and positive appraisals (Freeze
and DiTommaso 2014; Kent and Henderson 2017). However, our analyses suggest this may not
be especially problematic for the avoidant, as the results may fit a larger pattern of proximal
ambivalence that may not correlate with distress.
Two further lines of inquiry may also be valuable. First, we suspect anxiety about one’s
relationship with God is coterminous with other life anxieties. Unfortunately, the cross-sectional
nature of these data prevents us from examining the durability of attachment to God styles or
whether attachment is affected by peoples fluctuating emotional responses to life stress. While
attachment styles are usually presumed durable, it is possible that people express greater anxiety
about their relationship with God during times of acute distress and may settle into more secure
or avoidant attachment styles when times are more stable. More research is needed to investigate,
not only attachment styles across the life course, but also their trajectories for psychological
well-being over time. Second, our results suggest that the relationship between religious beliefs
and mental health is strongly contingent on the uncertainty people experience regarding their
beliefs. This is consistent with studies of uncertainty or anxiety about the absence of God among
professed atheists which exhibit similar results as those presented here (Baker, Stroope, and
Walker 2018; May 2018; Weber et al. 2012). Though beyond the scope of this study, future work
should continue comparing the differential effects of different beliefs about the divine.
Limitations And Conclusion
We must note that although our theoretical assumption is that dispositions about God
predict mental health, it is also likely that mental distress destabilizes those dispositions (Kent et
al. 2020). Previous findings demonstrate that some religious behaviors, such as frequency of
prayer, are associated with greater distress (Bradshaw, et al. 2008) and people may revise or
refine their religious beliefs in response to periods of crisis, similar to how they may augment
their religious practices. The cross-sectional nature of these data prevent us from definitively
demonstrating the direction of this relationship. We should also note that the BRS is limited to
the U.S. population and is predominantly Christian (86%). Though attachment to God measures
are not explicitly based on Christian conceptions, and have been validated in non-Christian
contexts (Bonab, Miner, and Proctor 2013), we cannot address whether uncertain or fraught
relationships with the divine associate similarly in different cultural settings. Lastly, our results
are drawn from a sample of the general population, and do not speak to experiences of clinically
referred patients. The relationships detailed here may be different among samples with more
acute psychiatric symptoms.
To conclude, our study makes several contributions to the study of religion and health.
First, we confirm the documented salutary association between secure (non-avoidant) attachment
and psychological well-being, while at the same time identifying a similar association for the
avoidant. Second, we identify a consistent deleterious relationship between anxiety about one’s
relationship with the divine and psychological well-being. Results suggest anxious attachment
represents a unique source of spiritual struggle. Third, findings demonstrating a non-linear
relationship between secure attachment and anxious attachment further suggest applying the
avoidant/secure subscale in a linear fashion mischaracterizes the relationship between attachment
to God and mental health. Finally, we conclude that avoidance and security vis-à-vis relationship
with God represent stable ontological dispositions associated with lower psychological distress.
Rather, relationships with the divine characterized by anxiety evince a greater degree of distress.
References
Acock, Alan C. 2005. “Working with missing values.” Journal of Marriage and Family
67(4):101228.
Ainsworth, Mary D. Salter, Mary C. Blehar, Everett Waters, and Sally N. Wall. 2015. Patterns of
Attachment: A Psychological Study of the Strange Situation. Psychology Press.
Bader, Christopher D., F. Carson Mencken, and Paul Froese. 2007. “American piety 2005:
Content and methods of the Baylor Religion Survey.” Journal for the Scientific Study of
Religion 46(4):44763.
Baker, Joseph O., Samuel Stroope, and Mark H. Walker. 2018. “Secularity, religiosity, and
health: Physical and mental health differences between atheists, agnostics, and
nonaffiliated theists compared to religiously affiliated individuals.” Social Science
Research 75:4457.
Baylor University. 2010. The Baylor Religion Survey, Wave III. Waco, TX: Baylor Institute for
Studies of Religion.
Birgegard, Andreas, and Pehr Granqvist. 2004. “The correspondence between attachment to
parents and god: Three experiments using subliminal separation cues.” Personality &
Social Psychology Bulletin 30(9):112235.
Bonab, Bagher Ghobary, Maureeen Miner, and Marie-Therese Proctor. 2013. “Attachment to
god in Islamic spirituality.” Journal of Muslim Mental Health 7(2):77-104.
Bowlby, John. 1969. Attachment and loss: Vol. 1. attachment. 2nd edition. New York: Basic
Books.
Bradshaw, Matt, Christopher G. Ellison, and Kevin J. Flannelly. 2008. “Prayer, god imagery, and
symptoms of psychopathology.” Journal for the Scientific Study of Religion 47(4):644
59.
Bradshaw, Matt, Christopher G. Ellison, and Jack 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(2):13047.
Bradshaw, Matt, and Blake Victor Kent. 2016. “Prayer, attachment to god, and changes in
psychological well-being in later life.” Journal of Aging and Health 30(5) 667-691.
Cicirelli, Victor. 2004. “God as the ultimate attachment figure for older adults.” Attachment &
Human Development 6(4):37188.
Collins, Nancy L. and Brooke C. Feeney. 2000. A safe haven: An attachment theory perspective
on support seeking and caregiving in intimate relationships. Journal of Personality and
Social Psychology 78(6):1053-73.
Dougherty, Kevin D., Byron R. Johnson, and Edward C. Polson. 2007. “Recovering the lost:
Remeasuring U.S. religious affiliation.” Journal for the Scientific Study of Religion
46(4):48399.
Ellison, Christopher. G., and Jeffrey. S. Levin. 1998. “The religion-health connection: Evidence,
theory, and future directions.” Health Education & Behavior 25(6):700720.
Ellison, Christopher. G., Wei Zhang, Neil Krause, and John. P. Marcum. 2009. “Does negative
interaction in the Church increase psychological distress? Longitudinal findings from the
Presbyterian Panel Survey.” Sociology of Religion 70(4):40931.
Ellison, Christopher G., and Andrea K. Henderson. 2011. “Religion and mental health: Through
the lens of the stress process.” in Religion and the social order : Toward a sociological
theory of religion and health, edited by A. Blasi. Leiden, NL: Brill.
Ellison, Christopher G., Matt Bradshaw, Kevin J. Flannelly, and Kathleen C. Galek. 2014.
“Prayer, attachment to God, and symptoms of anxiety-related disorders among U.S.
adults.” Sociology of Religion 75(2):20833.
Ellison, Christopher G., Matt Bradshaw, Nilay Kuyel, and Jack P. Marcum. 2012. “Attachment
to God, stressful life events, and changes in psychological distress.” Review of Religious
Research 53(4):493511.
Ellison, Christopher G., Amy M. Burdette, and Terrence D. Hill. 2009. “Blessed assurance:
Religion, anxiety, and tranquility among US adults.” Social Science Research 38(3):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(1):4661.
Ellison, Christopher G., and Robert A. Hummer. 2010. “Future directions in population-based
research on religion, family life, and health in the United States.” Pp. 431–54 in Religion,
families, and health: Population-based research in the United States, edited by C. G.
Ellison and R. A. Hummer. Rutgers University Press.
Ellison, Christopher G., and Jinwoo Lee. 2010. “Spiritual struggles and psychological distress: Is
there a dark side of religion?” Social Indicators Research 98(3):50117.
Exline, Julie Juola. 2002. “Stumbling blocks on the religious road: Fractured relationships,
nagging vices, and the inner struggle to believe.” Psychological Inquiry 13(3):18289.
Fraley, R. Chris, and Phillip R. Shaver. 2000. Adult romantic attachment: Theoretical
developments, emerging controversies, and unanswered questions. Review of General
Psychology 4(2):13254.
Frazier, M. Lance, Janaki Gooty, Laura M. Little, and Debra L. Nelson. 2015. Employee
attachment: implications for supervisor trustworthiness and trust. Journal of Business and
Psychology 30(2):37386.
Fenigstein, Allan., and Peter. A. Vanable. 1992. Paranoia and self-consciousness. Journal of
Personality and Social Psychology 62(1):12938.
Freeze, Tracy A., and Enrico DiTommaso. 2014. “An examination of attachment, religiousness,
spirituality and well-being in a Baptist faith sample.” Mental Health, Religion & Culture
17(7):690702.
Freud, Sigmund. 1927. The future of an illusion. W. W. Norton & Company.
Froese, Paul, and Christopher Bader. 2010. America’s four gods: What we say about God-- &
what that says about us. New York: Oxford University Press.
Galen, Luke William, and James D. Kloet. 2011. “Mental well-being in the religious and the
non-religious: Evidence for a curvilinear relationship.” Mental Health, Religion &
Culture 14(7):67389.
George, Linda K., Christopher G. Ellison, and David B. Larson. 2002. “Explaining the
relationships between religious involvement and health.” Psychological Inquiry
13(3):190200.
Granqvist, Pehr. 2020. Attachment in religion and spirituality: A wider view. New York: The
Guilford Press.
Granqvist, Pehr, and Berit Hagekull. 1999. “Religiousness and perceived childhood attachment:
Profiling socialized correspondence and emotional compensation.” Journal for the
Scientific Study of Religion 38(2):25473.
Granqvist, Pehr, and Berit Hagekull. 2003. “Longitudinal predictions of religious change in
adolescence: Contributions from the interaction of attachment and relationship status.”
Journal of Social and Personal Relationships 20(6):793817.
Granqvist, Pehr, and Lee A. Kirkpatrick. 2013. “Religion, spirituality, and attachment.” Pp. 139–
55 in APA handbook of psychology, religion, and spirituality (Vol 1): Context, theory,
and research, APA handbooks in psychology., edited by K. I. Pargament, J. J. Exline, and
J. W. Jones. Washington, DC, US: American Psychological Association.
Greenfield, Emily A., George E. Vaillant, and Nadine F. Marks. 2009. “Do formal religious
participation and spiritual perceptions have independent linkages with diverse dimensions
of psychological well-being?” Journal of Health and Social Behavior 50(2):196212.
Groves, Robert M. 2006. “Nonresponse rates and nonresponse bias in household surveys.”
Public Opinion Quarterly 70(5):64675.
Hazan, Cindy, and Phillip Shaver. 1987. “Romantic love conceptualized as an attachment
process.” Journal of Personality and Social Psychology 52(3):51124.
Huppert, Felicia A. 2009. “Psychological well-being: Evidence regarding its causes and
consequences.” Applied Psychology: Health and Well-Being 1(2):13764.
Johnson, Susan C., Carol S. Dweck, Frances S. Chen, Hilarie L. Stern, Su-Jeong Ok, and Maria
Barth. 2010. “At the intersection of social and cognitive development: Internal working
models of attachment in infancy.” Cognitive Science 34(5):80725.
Kaplan, Stephen L. 1994. A self-rated scale for obsessive-compulsive disorder. Journal of
Clinical Psychology 50(4):56474.
Kent, Blake Victor. 2017. “Attachment to God, religious tradition, and firm attributes in
workplace commitment.” The Journal of Social Psychology 157(4):485501.
Kent, Blake Victor. 2020. “Religion/spirituality and gender-differentiated trajectories of
depressive symptoms age 13-34.” Journal of Religion and Health 59(4):206481.
Kent, Blake Victor, Matt Bradshaw, and Kevin D. Dougherty. 2016. “Attachment to God,
vocational calling, and worker contentment.” Review of Religious Research 58(3):343
64.
Kent, Blake Victor, Matt Bradshaw, and Jeremy E. Uecker. 2018. “Forgiveness, attachment to
God, and mental health outcomes in older U.S. adults: A longitudinal study.” Research
on Aging 40(5):456-479.
Kent, Blake Victor, and W. Matthew Henderson. 2017. “Attachment to God, tenure, race, and
participation in congregational life.” Review of Religious Research 59:319-340.
Kent, Blake Victor, W. Matthew Henderson, Matt Bradshaw, Christopher G. Ellison, and
Bradley R. E. Wright. 2020. “Do daily spiritual experiences moderate the effect of
stressors on psychological well-being? A smartphone-based experience sampling study of
depressive symptoms and flourishing.” The International Journal for the Psychology of
Religion 31(2): 5778.
Kirkpatrick, Lee A. 2004. Attachment, evolution, and the psychology of religion. New York: The
Guilford Press.
Kirkpatrick, Lee A., and Keith E. Davis. 1994. “Attachment style, gender, and relationship
stability: A longitudinal analysis.” Journal of Personality and Social Psychology
66(3):50212.
Kirkpatrick, Lee A., and Cindy Hazan. 1994. “Attachment styles and close relationships: A four-
year prospective study.” Personal Relationships 1(2):12342.
Kirkpatrick, Lee A., and Philip R. Shaver. 1992. “An attachment-theoretical approach to
romantic love and religious belief.” Personality and Social Psychology Bulletin
18(3):26675.
Kirkpatrick, Lee A., Daniel J. Shillito, and Susan L. Kellas. 1999. “Loneliness, social support,
and perceived relationships with God.” Journal of Social and Personal Relationships
16(4):51322.
Krause, Neal, Christopher G. Ellison, Benjamin A. Shaw, John P. Marcum, and Jason D.
Boardman. 2001. “Church-based social support and religious coping.” Journal for the
Scientific Study of Religion 40(4):63756.
Krause, Neal, and R. David Hayward. 2016. “Anxious attachment to God, spiritual support, and
obesity: Findings from a recent nationwide survey.” Journal for the Scientific Study of
Religion 55(3):48597.
Kroenke, Kurt, Robert L. Spitzer, Janet B. W. Williams, and Bernd Löwe. 2010. The patient
health questionnaire somatic, anxiety, and depressive symptom scales: A systematic
review. General Hospital Psychiatry 32(4):34559.
Leuba, James H. 1925. The psychology of religious mysticism. Kegan Paul, Trench, Trubner And
Company., Limited.
Levin, Jeffrey S., Linda M. Chatters, and Robert Joseph Taylor. 1995. “Religious effects on
health status and life satisfaction among black Americans.” The Journals of Gerontology
Series B: Psychological Sciences and Social Sciences 50B(3):S15463.
May, Matthew. 2018. “Should I stay or should I go? Religious (dis)affiliation and depressive
symptomatology.” Society and Mental Health 8(3):21430.
McDonald, Angie, Richard Beck, Steve Allison, and Larry Norswortby. 2005. “Attachment to
God and parents: Testing the correspondence vs. compensation hypotheses.” Journal of
Psychology & Christianity 24(1).
Mikulincer, Mario, and Victor Florian. 1998. “The relationship between adult attachment styles
and emotional and cognitive reactions to stressful events.” Pp. 143–65 in Attachment
theory and close relationships, edited by J. A. Simpson and W. S. Rholes. New York,
NY, US: Guilford Press.
Moore, Kathleen A. and Donna L. Gee. 2003. The reliability, validity, discriminant and
predictive properties of the social phobia inventory (SoPhI). Anxiety, Stress, & Coping
16(1):10917.
Murphy, Barbara, and Glen W. Bates. 1997. “Adult attachment style and vulnerability to
depression.” Personality and Individual Differences 22(6):83544.
Rosmarin, David H., and Harold G. Koenig, eds. 2020. Handbook of spirituality, religion, and
mental health. 2nd edition. Waltham: Academic Press.
Rowatt, Wade C., and Lee A. Kirkpatrick. 2002. “Two dimensions of attachment to God and
their relation to affect, religiosity, and personality constructs.” Journal for the Scientific
Study of Religion 41(4):63751.
Saferstein, Jocelyn A., Greg J. Neimeyer, and Chad L. Hagans. 2005. Attachment as a predictor
of friendship qualities in college youth. Social Behavior and Personality 33(8):767.
Schieman, Scott, and Alex Bierman. 2011. “The role of divine beliefs in stress processes.” in
Religion and the social order: Toward a sociological theory of religion and health, edited
by A. Blasi. Leiden, NL: Brill.
Schieman, Scott, Alex Bierman, and Christopher G. Ellison. 2013. “Religion and mental health.”
Pp. 45778 in Handbook of the sociology of mental health, edited by C. S. Aneshensel, J.
C. Phelan, and A. Bierman. Dordrecht: Springer Netherlands.
Skinner, B. F. 1953. Science And human behavior. Simon and Schuster.
Smith, Gregory A., and Alan Cooperman. 2016. “The factors driving the growth of religious
‘Nones’ in the U.S.” Pew Research Center. available at http://www.pewresearch.org/fact-
tank/2016/09/14/the-factors-driving-the-growth-of-religious-nones-in-the-u-s/, accessed
May 26, 2017
Steensland, Brian, Lynn D. Robinson, W. Bradford Wilcox, Jerry Z. Park, Mark D. Regnerus,
and Robert D. Woodberry. 2000. “The measure of American religion: Toward improving
the state of the art.” Social Forces 79(1):291318.
Stroope, Samuel, Blake Victor Kent, Ying Zhang, Namratha R. Kandula, Alka M. Kanaya, and
Alexandra E. Shields. 2020. “Self-rated religiosity/spirituality and four health outcomes
among US south Asians: Findings from the study on stress, spirituality, and health.” The
Journal of Nervous and Mental Disease 208(2):16568.
Weber, Samuel R., Kenneth I. Pargament, Mark E. Kunik, James W. Lomax, and Melinda A.
Stanley. 2012. “Psychological distress among religious nonbelievers: A systematic
review.” Journal of Religion and Health 51(1):7286.
Zhang, Paul. 2003. “Multiple imputation: theory and method.” International Statistical Review /
Revue Internationale de Statistique 71(3):58192.
Table 1. Unweighted Descriptive Statistics of Study Variables
Variable
N
SD
Range
Alpha
Dependent Variables
General Distress
1538
2.573
0-12
.878
General Anxiety
1573
2.678
0-12
.841
Social Anxiety
1564
2.289
0-12
.824
Paranoia
1567
2.356
0-12
.763
Obsession
1566
2.326
0-12
.764
Compulsion
1564
1.778
0-12
.763
Attachment to God Variables
Secure Attachment
1464
4.29
0-18
.914
Anxious Attachment
1461
2.10
0-9
.786
Religious Control Variables
Evangelical Protestant
1575
-
0,1
-
Mainline Protestant
1575
-
0,1
-
Black Protestant
1575
-
0,1
-
Catholic
1575
-
0,1
-
Religious Other
1575
-
0,1
-
Non-Religious a
1575
-
0,1
-
Religious Attendance
1612
-
0-8
-
Control Variables
Age
1577
16.133
18-100
-
White
1624
-
0,1
-
Female
1604
-
0,1
-
South a
1624
-
0,1
-
East
1624
-
0,1
-
Midwest
1624
-
0,1
-
West
1624
-
0,1
-
Married
1580
-
0,1
-
Raising Minor Child/ren
1622
-
0,1
-
Ed: HS or Less a
1580
-
0,1
-
Ed: Some College
1580
-
0,1
-
Ed: Bachelor’s Degree
1580
-
0,1
-
Ed: Beyond Bachelor’s
1580
-
0,1
-
Income Bracket
1495
-
1-7
-
Employed
1570
-
0,1
-
Notes: Means and standard deviations are unweighted and recorded prior to imputation. a Indicates reference
category. Source: Baylor Religion Survey (2010).
Table 2. Results of OLS Analysis of General Distress on Attachment to God Styles
Secure Attachment
Linear Model
Secure Attachment
Non-Linear Model
Anxious Attachment
Linear Model
Parameter
b
SE
b
SE
b
SE
Intercept
2.166
***
.16
2.212
***
.16
2.007
***
.15
Secure Attachment
-.011
.01
-.015
.01
-
-
Secure Attachment2
-
-
-.001
.00
-
-
Anxious Attachment
-
-
-
-
.069
***
.01
Evangelical Protestant a
-.088
.11
-.119
.10
-.165
.10
Mainline Protestant a
-.020
.10
-.060
.10
-.114
.10
Black Protestant a
-.312
.17
-.334
*
.17
-.411
*
.17
Catholic a
.031
.11
-.001
.10
-.071
.10
Religious Other a
-.017
.12
-.031
.12
-.095
.12
Attendance
-.034
***
.01
-.033
***
.01
-.030
***
.01
r2
.114
.115
.133
Notes: n = 1,624; *p <. 05; **p < .01; ***p < .001; † p < .10; r2 taken from models using non-imputed data; non-
linear model includes mean centered measure of Secure Attachment; all models include control measures of Age,
Race, Gender, Region, Parental Status, Education, Income, and Employment. Full models available on request.
Source: Baylor Religion Survey (2010)
Table 3. Results of OLS Analysis of Psychological Symptom Scales on Avoidant/Secure Attachment to God (1-5) and Anxious Attachment to God (6-10)
General Anxiety (1)
Social Anxiety (2)
Paranoia (3)
Obsession (4)
Compulsion (5)
Parameter
b
SE
b
SE
b
SE
b
SE
b
SE
Intercept
2.363
***
.15
1.710
***
.16
2.036
***
.16
2.048
***
.14
1.320
***
.15
Secure Attachment
-.030
***
.01
-.031
***
.01
-.032
***
.01
-.030
***
.01
-.028
***
.01
Secure Attachment²
-.003
***
.00
-.004
***
.00
-.005
***
.00
-.003
**
.00
-.003
**
.00
Evangelical Protestant a
.032
.09
.054
.11
.124
.10
-.089
.09
.028
.09
Mainline Protestant
.009
.09
.067
.11
.151
.10
-.003
.09
.065
.10
Black Protestant
-.145
.15
.202
.18
.191
.17
-.297
.15
.107
.17
Catholic
.041
.09
.109
.11
.134
.10
-.008
.09
.081
.09
Religious Other
-.056
.11
-.060
.13
.075
.12
-.135
.11
-.011
.11
Attendance
-.017
*
.01
.009
.01
-.022
*
.01
-.006
.01
-.006
.01
r2
.101
.073
.116
.088
.087
General Anxiety (6)
Social Anxiety (7)
Paranoia (8)
Obsession (9)
Compulsion (10)
Parameter
b
SE
b
SE
b
SE
b
SE
b
SE
Intercept
2.112
***
.13
1.450
***
.16
1.700
***
.15
1.838
***
.13
1.113
***
.14
Anxious Attachment
.073
***
.01
.073
***
.01
.083
***
.01
.074
***
.01
.066
***
.01
Evangelical Protestant a
-.005
.09
.028
.10
.109
.10
-.143
.09
-.022
.09
Mainline Protestant
-.042
.09
.032
.10
.128
.10
-.058
.09
.010
.09
Black Protestant
-.176
.15
.184
.17
.164
.17
-.357
*
.15
.034
.16
Catholic
-.025
.09
.072
.10
.106
.10
-.077
.09
.020
.09
Religious Other
-.099
.11
-.084
.12
.055
.12
-.195
.10
-.077
.11
Attendance
-.021
**
.01
.005
.01
-.023
**
.01
-.011
.01
-.009
.01
r2
.122
.084
.135
.117
.109
Notes: n = 1,624; *p <. 05; **p < .01; ***p < .001; † p < .10; r2 taken from models using non-imputed data; Models 1-5 include mean centered measure of
Secure Attachment; all models include control measures of Age, Race, Gender, Region, Parental Status, Education, Income, and Employment. Full models
available on request.
Source: Baylor Religion Survey (2010)
Table 4. Results of OLS Analysis of Anxious Attachment on Avoidant/Secure Attachment to God
Parameter
b
SE
b
SE
Intercept
1.227
**
.38
2.195
***
.35
Secure Attachment
-.208
***
.01
-.342
***
.02
Secure Attachment²
-
-
-.032
***
.00
Evangelical Protestant a
1.406
***
.22
.911
***
.21
Mainline Protestant
1.500
***
.22
.881
***
.21
Black Protestant
1.549
***
.38
1.035
**
.35
Catholic
1.569
***
.22
1.040
***
.21
Religious Other
1.409
***
.26
.896
***
.25
Attendance
-.018
.02
.027
.02
General Distress
.124
***
.02
.119
***
.02
r2
.229
.339
Notes: n = 1,624; *p <. 05; **p < .01; ***p < .001; † p < .10; r2 taken from models using non-imputed data; non-
linear model includes mean centered measure of Secure Attachment; all models include control measures of Age,
Race, Gender, Region, Parental Status, Education, Income, and Employment. Full models available on request.
Source: Baylor Religion Survey (2010)
Figure 1.
Prediction Lines for Non-Linear Models
0.5
1
1.5
2
2.5
3
-13.2
-12.2
-11.2
-10.2
-9.2
-8.2
-7.2
-6.2
-5.2
-4.2
-3.2
-2.2
-1.2
-0.2
0.8
1.8
2.8
3.8
4.8
Psychological Distress
Secure Attachment to God (centered data)
Generalized Anxiety
Social Anxiety
Paranoia
Obsession
Compulsion
-0.5
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
-13.2
-12.2
-11.2
-10.2
-9.2
-8.2
-7.2
-6.2
-5.2
-4.2
-3.2
-2.2
-1.2
-0.2
0.8
1.8
2.8
3.8
4.8
Anxious Attachment to God
Secure Attachment to God (Centered Data)
Linear Model
Non-Linear Model
A
B
Article
Full-text available
The “spiritual but not religious” (SBNR) are a growing group in the religious landscape of the United States. Thousands of studies to date have been devoted to the study of religion and health, but far less attention has been given to the study of the “spiritual but not religious.” In this study, we address this gap by using two waves of longitudinal data from the National Study of Youth and Religion (2005–2008). We assess whether within-person changes in SBNR identity are associated with health and mental health in emerging adulthood and consider several pathways that may account for observed differences. Results suggest that consistently identifying as SBNR was associated with worse physical and mental health relative to youth that were consistently religious. Using parametric mediation analyses, we found evidence that three of our four proposed mediators (religious attendance, sense of closeness to God, and religious doubt, but not life meaning) partially explained these mental health differences. This study therefore makes an important advance in assessing the health implications of (non)-religion/spirituality early in the life course.
Article
Full-text available
Much of the survey research on religion/spirituality and mental health in the U.S. must be assumed to capture stable respondent traits when it is possible temporary states are actually being recorded. The smartphone-based experience sampling method (ESM) in the SoulPulse Study, which collected data twice a day for two weeks, allows an opportunity to examine this problem further by assessing state (single moment) and trait (two-week average) scores of daily spiritual experiences (DSE) as moderators of a daily stressor checklist, depressive symptoms, and flourishing (a well-being indicator that addresses happiness, life satisfaction, meaning and purpose, virtue, close social relationships, etc.). Findings indicate robust direct associations between stressors, DSE, and well-being, as well as substantial support for the moderating role of state and trait daily spiritual experiences. The study: 1) demonstrates that DSE may serve as a buffer against daily stressors at both the trait and state levels, 2) provides further evidence for flourishing as a holistic indicator of well-being, and 3) indicates that ESM methodologies can add to our understanding of human well-being.
Book
Full-text available
Synthesizing diverse strands of theory and research, this compelling book explores the psychology of religion and spirituality through an innovative attachment lens. Pehr Granqvist examines the connections between early caregiving experiences, attachment patterns, and individual differences in religious cognition, experience, and behavior. The function of a deity as an attachment figure is analyzed, as are ways in which attachment facilitates the intergenerational transmission of religion. The book also shows how the attachment perspective can aid in understanding mystical experiences, connections between religion and mental health, and cultural differences between more and less religious societies. Granqvist's conversational writing style, concrete examples, and references to popular culture render complex concepts accessible. For further info and order, see: https://www.guilford.com/books/Attachment-in-Religion-and-Spirituality/Pehr-Granqvist/9781462542680/reviews For podcast interview: https://newbooksnetwork.com/pehr-granqvist-attachment-in-religion-and-spirituality-a-wider-view-guilford-press-2020/
Article
Full-text available
Studies examining religion/spirituality (R/S) and depressive symptoms report divergent findings, often depending on the types of variables considered. This study assessed whether subjective and experiential R/S variables were associated with increased depressive symptom burden from adolescence to young adulthood. Variations by gender were also assessed. Using group-based trajectory modeling with a cohort-sequential design, four distinct symptom trajectories were identified for women and five for men. 27.4% of women and 10.2% of men were classified on peak trajectory groups. Religious attendance was protective for men and women. Prayer was protective for women but linked to risk for men. Born-again and life-changing spiritual experiences, along with belief in supernatural leading and angelic protection, were broadly associated with increased classification on elevated symptom trajectories. In one exception, belief in supernatural leading was associated among some men with decreased risk of depressive symptoms during adolescence. Researchers must take a variety of R/S variables into account when assessing depressive symptoms, not simply religious attendance, prayer frequency, or affiliation as is commonly practiced. Religion and spirituality are multidimensional and in some cases may operate differently for men than women vis-à-vis mental well-being.
Article
Full-text available
Extensive literature in the social and medical sciences link religiosity to positive health outcomes. Conversely it is often assumed that secularity carries negative consequences for health; however, recent research outlining different types of secular individuals complicates this assumption. Using a national sample of American adults, we compare physical and mental health outcomes for atheists, agnostics, religiously nonaffiliated theists, and theistic members of organized religious traditions. Results indicate better physical health outcomes for atheists compared to other secular individuals and members of some religious traditions. Atheists also reported significantly lower levels of psychiatric symptoms (anxiety, paranoia, obsession, and compulsion) compared to both other seculars and members of most religious traditions. In contrast, physical and mental health were significantly worse for nonaffiliated theists compared to other seculars and religious affiliates on most outcomes. These findings highlight the necessity of distinguishing among different types of secular individuals in future research on health.
Article
Full-text available
We analyze a sample of older US adults with religious backgrounds in order to examine the relationships among two types of divine forgiveness and three indicators of psychological well-being (PWB), as well as the moderating role of attachment to God. Results suggest that: (a) feeling forgiven by God and transactional forgiveness from God are not associated with changes in PWB over time, (b) secure attachment to God at baseline is associated with increased optimism and self-esteem, (c) feeling forgiven by God and transactional forgiveness from God are more strongly associated with increased PWB among the securely attached, and (d) among the avoidantly attached, PWB is associated with consistency in one's beliefs, i.e., a decreased emphasis on forgiveness from God. Findings underscore the importance of subjective beliefs about God in the lives of many older adults in the United States.
Article
Full-text available
The vitality of religious congregations is dependent on how effectively they attract congregants and mobilize participation, and people are more likely to participate when they share similar characteristics with other congregants. This study suggests attachment to God is a fundamental “behavioral and intrapersonal characteristic” which distinguishes participants from one another, contributing to varying levels of participation in congregational and religious life beyond service attendance. Using a national sample, we test several hypotheses related to this theoretical claim. Findings suggest that: (1) secure attachment to God is positively associated with congregational participation, (2) anxious and avoidant attachment are negatively associated with participation, (3) these relationships are moderated by length of time attending a particular congregation, and (4) secure attachment is associated with higher levels of participation in congregational life for blacks than it is for whites. By demonstrating a link between attachment to God and participation in congregational and religious life, this study affirms that intrapersonal characteristics structure the dynamics of religious congregations.
Article
Full-text available
Objective: The aim of this study was to investigate the effects of prayer and attachment to God on psychological well-being (PWB) in later life. Method: Using data from two waves of the nationwide Religion, Aging, and Health Survey, we estimate the associations between frequency of prayer and attachment to God at baseline with cross-wave changes in three measures of PWB: self-esteem, optimism, and life satisfaction. Results: Prayer does not have a main effect on PWB. Secure attachment to God is associated with improvements in optimism but not self-esteem or life satisfaction. The relationship between prayer and PWB is moderated by attachment to God; prayer is associated with improvements in PWB among securely attached individuals but not those who are insecurely attached to God. Discussion: These findings shed light on the complex relationship between prayer and PWB by showing that the effects of prayer are contingent upon one's perceived relationship with God.
Article
Religious affiliation is generally associated with better mental health. The nonreligious, however, currently constitute one of the fastest-growing religious categories in the United States. Since most of the nonreligious were raised in religious homes, their growth raises important questions about the mental health of those who consider dropping out of religion. In this article, I use longitudinal data from the Portraits of American Life Study to examine the impact of religious affiliation on mental health. Specifically, I compare individuals who dropped out of religion (leavers) with individuals who considered dropping out (stayers) and individuals who are more consistent in their religious (stable affiliates) and nonreligious (stable Nones) affiliations. I find that stayers experience more depressive symptoms than any other group and that they experience a greater increase in depressive symptoms over time. My findings are consistent with identity theories in sociology, and they provide evidence that a strong religious or secular identity is an important contributor to mental health.