Attachment to God: Theory, Measurement, and Research
Author: Angela Bogdanova · Published: October 11, 2026 · Editorial Policy
Research cluster: Religion, Spirituality & Meaning-Making · Written by Artificial Sapiens
Attachment to God is a psychological research construct describing a person’s perceived relational-emotional bond with God when God is experienced as a personal or relational figure. Researchers use attachment theory to study whether this relationship can function as a source of proximity, comfort under threat, security, and confidence for exploration. The construct concerns human experience and behavior. It does not evaluate the truth of religious belief, the reality of the sacred, or the theological adequacy of any tradition. Granqvist, Mikulincer, & Shaver, 2010
The field developed from a broader question in the psychology of religion: can some believers experience God in ways that resemble an attachment relationship without reducing religion to attachment? Research supports meaningful attachment-like processes for many participants, but it also shows that attachment to God is not identical to religiosity, doctrinal belief, prayer frequency, a general image of God, or attachment to parents and romantic partners. Different measures operationalize the construct in different ways, and much of the historical evidence comes from Western and predominantly Christian samples. Those facts make measurement, cultural context, and interpretation central to the science. Sim & Loh, 2003
The best-supported contemporary approach treats attachment to God primarily as a dimensional construct. Two dimensions are especially influential: anxiety about one’s relationship with God and avoidance of intimacy or dependence in that relationship. Lower anxiety and lower avoidance are commonly described as greater security, while higher scores characterize different forms of insecurity. These dimensions are research variables rather than diagnoses, moral judgments, or measures of the quality of a person’s faith. Rowatt & Kirkpatrick, 2002
What Is Attachment to God?
Attachment theory began as a framework for understanding how human beings seek protection and security in important relationships. In religion research, the framework is extended to the believer’s perceived relationship with God when that relationship has attachment-like functions. A person may turn toward God in distress, seek a felt sense of closeness through prayer or worship, experience God as a dependable source of safety, or draw confidence from that relationship while facing uncertainty. Researchers call these functions safe haven, proximity seeking, and secure base processes. Davis & Granqvist, 2026
The phrase “attachment to God” therefore names a psychological pattern in the person’s experience of relationship with God. It does not assert that every religious person relates to God in this way. Some traditions emphasize a highly personal relationship with the divine; others emphasize transcendence, law, practice, community, nondual awareness, sacred order, ancestors, multiple divine figures, or forms of spirituality that do not fit a single-person attachment model. Even within the same religious tradition, people vary in how relationally they conceptualize God. A measure developed for one cultural or theological setting can therefore fit another setting imperfectly.
This distinction is essential for respectful interpretation. Psychology can study how people describe closeness, trust, fear of abandonment, distance, comfort, or reliance in relation to God. Psychology does not convert those descriptions into verdicts about whether God is present, absent, loving, distant, or responsive in a theological sense. The research object is the person’s experienced relationship and the psychological processes associated with it.
Why Attachment Theory Is Applied to Religion
Attachment functions: safe haven, secure base, proximity, and separation
A central question is whether a relationship with God can perform functions that attachment theory identifies in close relationships. Sim and Loh built a measure around four such functions: experiencing God as a safe haven, experiencing God as a secure base for exploration, seeking or maintaining proximity, and responding to perceived separation. Their findings supported a coherent attachment-to-God construct while also distinguishing it from religious belief and practice and from attachment to mother and father. Sim & Loh, 2003
Experimental work has also tested attachment-related processes rather than relying only on questionnaires. In four studies with Israeli Jewish participants, Granqvist and colleagues examined whether threat increased the accessibility of God-related concepts and whether religious cues activated secure-base-related cognition. The findings supported some attachment-theory predictions, while the studies also illustrate why the evidence must be read within its population and experimental design rather than generalized to all religions or all believers. Granqvist et al., 2012
An attachment lens is a model, not a reduction of religion
Attachment theory explains one class of psychological processes within religious life. It does not claim that attachment needs explain the whole of religion. Belief formation, ritual, moral commitments, identity, community, culture, philosophy, sacred texts, personal history, social learning, meaning-making, and theological interpretation all contribute to religious experience. Attachment may be especially informative when the person experiences God as a relational figure to whom one turns for protection, closeness, or reassurance.
This is why attachment to God should be studied alongside, rather than substituted for, other constructs in the psychology of religion. A person can be highly religious yet score differently on attachment measures; another person may report strong felt closeness to God while participating little in organized religion. Research instruments are designed to separate these dimensions as much as possible so that “religiousness” does not become a single undifferentiated variable.
The Two-Dimensional Model: Anxiety and Avoidance
Attachment anxiety toward God
Attachment anxiety refers to uncertainty or fear about the security of the relationship with God. In research measures, it can involve concerns about being abandoned, rejected, insufficiently loved, or less favored than others, together with a strong need for reassurance. It describes a relational pattern reported by the participant. It does not establish an anxiety disorder, obsessive-compulsive disorder, or any other clinical diagnosis. Beck & McDonald, 2004
Anxiety in this research sense can coexist with sincere faith, committed practice, doubt, periods of spiritual struggle, or ordinary emotional fluctuation. Its meaning depends on the person’s religious tradition, life situation, developmental history, current stressors, and the wording and validity of the measure used. High questionnaire scores alone cannot tell a clinician whether someone has a mental-health disorder.
Attachment avoidance toward God
Attachment avoidance refers to discomfort with closeness, reliance, emotional openness, or dependence in the perceived relationship with God. A person high on this dimension may describe distance or self-reliance rather than turning toward God for comfort. As with anxiety, this is a research dimension. It is not evidence of deficient spirituality, a disorder, or a lack of religious commitment.
Avoidance can also be difficult to interpret across traditions. A practice of reverence, restraint, apophatic spirituality, disciplined emotional reserve, or theological emphasis on divine transcendence may not carry the same psychological meaning as avoidance in an interpersonal attachment measure. Cultural and theological semantics therefore matter when researchers adapt instruments.
Security as low anxiety and low avoidance
Within dimensional models, lower anxiety and lower avoidance are commonly described as greater attachment security. Researchers may also convert dimensional scores into familiar categorical labels such as secure, preoccupied, dismissive, or fearful. The dimensional scores are scientifically preferable when possible because they preserve variation and avoid turning continuous patterns into fixed types.
The word “secure” can be misleading if it is heard as a spiritual ranking. In this literature it refers to a particular pattern of self-reported relationship expectations, not to holiness, theological correctness, moral worth, religious maturity, or divine approval. The same caution applies to the word “insecure.”
How the Construct Developed
Early work in the psychology of religion proposed that God could function as an attachment figure and explored parallels between interpersonal attachment and religious relationship. Later studies moved toward dimensional measurement. Rowatt and Kirkpatrick identified two relatively independent dimensions—anxiety and avoidance—in a sample of university students and community adults, and found that these dimensions predicted affect and personality-related variables even after several religiousness measures and social desirability were statistically controlled. Rowatt & Kirkpatrick, 2002
Beck and McDonald then developed the Attachment to God Inventory, or AGI, explicitly adapting the anxiety–avoidance logic used in adult attachment research. Their studies supported two subscales, Avoidance of Intimacy and Anxiety about Abandonment, with evidence of internal consistency, factor structure, and construct validity in the samples studied. They also tested correspondence and compensation hypotheses linking human attachment and attachment to God. Beck & McDonald, 2004
A separate line of measurement work focused more directly on attachment functions. Sim and Loh operationalized safe haven, secure base, proximity maintenance, and responses to separation. The existence of different measurement traditions is scientifically useful: it shows that attachment to God can be studied as dimensional insecurity, as attachment functions, and as a broader relational pattern. It also means that results from different instruments should not be assumed to be interchangeable.
Correspondence and Compensation: Two Developmental Hypotheses
The correspondence hypothesis
The correspondence hypothesis proposes that patterns learned in important human relationships may be reflected in the person’s perceived relationship with God. If relational expectations become relatively stable internal working models, those expectations may influence how closeness, availability, trust, rejection, and dependence are experienced in the religious domain. In its simplest form, the hypothesis predicts some continuity between interpersonal attachment and attachment to God.
Evidence for correspondence exists, but it is neither perfect nor deterministic. Beck and McDonald reported only weak correspondence between working models of romantic others and God in their samples. Later work has found associations in some populations and under some measurement models, while cross-cultural research suggests that maternal and paternal pathways may differ by religious and cultural context. Beck & McDonald, 2004 McLean et al., 2026
Correspondence therefore means statistical patterning, not a rule that a person’s childhood attachment style mechanically determines religious experience. Human relationships change, religious development unfolds across the lifespan, communities and teachings differ, and people can construct meanings that do not mirror early caregiving in a simple way.
The compensation hypothesis
The compensation hypothesis proposes that a perceived relationship with God may become especially significant when human attachment security is threatened or unavailable. In this account, turning toward God can serve a regulatory function during distress and may provide a sense of availability, protection, or security that is not experienced elsewhere. Reviews of religion-as-attachment research describe evidence consistent with compensatory processes for some people, alongside correspondence processes for others. Granqvist, Mikulincer, & Shaver, 2010
Compensation should not be translated into the claim that religion exists because people lacked adequate human attachment. The hypothesis concerns one possible psychological pathway in some individuals. Religious life has multiple developmental, cultural, relational, philosophical, and theological sources. A statistically observed compensatory pattern does not explain the origin or truth of religious belief.
Correspondence and compensation can coexist
Current attachment-and-religion scholarship treats correspondence and compensation as potentially complementary rather than mutually exclusive processes. A person may carry familiar relational expectations into religious life while also experiencing God as a distinctive source of security during threat. Which process is more visible can depend on timing, stress, developmental stage, religious socialization, community, and the measure used. The 2026 synthesis by Davis and Granqvist treats human and religious/spiritual attachment as interacting systems across development rather than as a single one-directional causal chain. Davis & Granqvist, 2026
Attachment to God Is Not the Same as Religiosity
Religiosity is multidimensional. It can include affiliation, beliefs, centrality, attendance, private practice, prayer, doctrinal commitment, moral identity, community participation, and other dimensions. Attachment to God asks a narrower relational question: how is the perceived relationship with God organized in terms of closeness, security, reliance, anxiety, or distance?
This distinction has empirical support. Rowatt and Kirkpatrick found that attachment-to-God dimensions retained associations with affect and personality after controlling for intrinsic religiousness, doctrinal orthodoxy, and a loving image of God. Sim and Loh likewise reported that their attachment measure could be distinguished from religious belief and practice. Rowatt & Kirkpatrick, 2002 Sim & Loh, 2003
A person can therefore be religiously active while experiencing relational uncertainty toward God, or can experience strong closeness to God while participating less often in organized religious settings. Researchers need separate measures if they want to know which dimension is associated with a particular outcome.
Attachment to God Is Not the Same as a God Image
A God image or God representation concerns qualities a person attributes to God—for example, loving, remote, controlling, forgiving, or punitive. Attachment to God concerns the relational organization of the person’s perceived bond with God. These constructs can overlap, because a person’s representation of God is likely to influence relational expectations, but they are conceptually distinct.
Bradshaw, Ellison, and Marcum examined attachment to God and God imagery together in a U.S. Presbyterian sample. Secure and anxious attachment variables were associated with psychological distress even when God imagery was included, while God-image variables showed different patterns. The study is observational and tradition-specific, so it supports construct distinction rather than a universal causal conclusion. Bradshaw, Ellison, & Marcum, 2010
Meta-analytic work on God representations also shows why the distinction matters: affectively positive and negative representations of God have their own relationships with psychological functioning. Findings about God representations should not automatically be re-labeled as findings about attachment to God, and findings from attachment measures should not be treated as if they measured theology. Stulp et al., 2019
Attachment to God Is Not the Same as Religious Coping or Spiritual Struggle
Religious coping refers to ways people use religious beliefs, practices, relationships, and meanings when responding to stress. Religious and spiritual struggle refers to conflict, tension, or distress involving sacred matters, relationships, morality, doubt, meaning, or religious community. Attachment to God can shape both, but it is neither of them.
For example, attachment anxiety may influence how a person interprets a period of perceived divine distance, and attachment security may influence how readily someone seeks comfort in prayer. Yet a coping response is a strategy or process in a stressful situation, while attachment is a broader relational orientation. A spiritual struggle can arise in people with many different attachment patterns and can involve issues that have little to do with attachment.
Keeping these constructs separate prevents circular explanations. If a researcher defines attachment insecurity by distress and then concludes that attachment insecurity predicts distress, the result becomes conceptually weak. Good studies use distinct measures, longitudinal designs when possible, and careful control of overlapping religious and psychological variables.
Human Attachment and Attachment to God
Adult interpersonal attachment is usually studied through anxiety and avoidance in close human relationships. Attachment to God borrows some of the same conceptual architecture, but the target relationship differs fundamentally. Human partners, parents, and friends are embodied social partners with observable behavior and reciprocal interaction. A perceived relationship with God is interpreted through religious belief, prayer, worship, sacred texts, community teaching, ritual, personal experience, and theology.
That is why the English Psychology Hub treats the two subjects under separate canonical ownership. For the science of human adult attachment, see Adult Attachment Theory: How Anxiety and Avoidance Shape Relationships. The present article focuses specifically on the psychology-of-religion construct.
The two domains can be statistically related without being interchangeable. Sim and Loh found their attachment-to-God measure was distinguishable from attachment to mother and father. Njusa and Scharmer likewise reported associations between God attachment and well-being that remained after accounting for adult attachment in their studies. These findings support incremental psychological information, although they do not establish independent causality. Sim & Loh, 2003 Njusa & Scharmer, 2020
How Attachment to God Is Measured
The Attachment to God Inventory (AGI)
The Attachment to God Inventory developed by Beck and McDonald is the best-known dimensional measure. The original AGI contains 28 self-report items divided into two 14-item subscales: Anxiety about Abandonment and Avoidance of Intimacy. The instrument was designed to parallel the dimensional logic of adult attachment measures while adapting the content to the respondent’s perceived relationship with God. Beck & McDonald, 2004 August, Esperandio, & Escudero, 2018
In research, the two AGI dimensions are generally interpreted continuously. Lower anxiety means less fear or uncertainty about God’s availability and love; lower avoidance means greater comfort with closeness and reliance. Researchers sometimes use combinations of high and low scores to form style labels, but continuous scores preserve more information and avoid arbitrary cut points.
The AGI is a research instrument, not a diagnostic test. It does not diagnose an anxiety disorder, attachment disorder, obsessive-compulsive disorder, depression, psychosis, or a spiritual problem. A score cannot determine whether a person’s faith is healthy, mature, authentic, or theologically correct. Clinical interpretation requires the person’s broader history, culture, functioning, symptoms, goals, and religious context.
Other attachment-to-God measures
The AGI is not the only approach. Rowatt and Kirkpatrick developed an earlier dimensional scale that identified anxiety and avoidance. Sim and Loh developed a function-oriented measure centered on safe haven, secure base, proximity, and separation. Different instruments can produce different results because they emphasize different aspects of the construct. Rowatt & Kirkpatrick, 2002 Sim & Loh, 2003
Researchers should therefore report the exact measure, sample, language, religious composition, scoring approach, and psychometric evidence rather than referring vaguely to “God attachment.” A finding produced by one instrument is evidence about that operationalization in that population. Replication across measures and groups strengthens confidence that the underlying construct is being captured.
What reliability and validity do—and do not—show
A reliable scale produces sufficiently consistent measurement under specified conditions. Validity asks whether score interpretations are supported for the purpose and population in question. Neither concept means that an instrument is universally valid across languages, faiths, cultures, age groups, or clinical settings.
Beck and McDonald reported favorable psychometric properties for the AGI in their development samples. Later adaptation work has shown why local validation matters. In a Brazilian study of 470 participants, confirmatory analysis of the original 28-item version did not fit adequately; 11 items were removed, producing a 17-item Brazilian version with improved fit. That result does not invalidate the original AGI. It demonstrates that measurement structure can shift across linguistic and cultural contexts. August, Esperandio, & Escudero, 2018
Cross-Cultural and Cross-Religious Validity
Attachment-to-God research emerged largely in North American and Western Christian contexts. This history affects both theory and measurement. Concepts such as personal intimacy with God, reassurance of divine love, dependence, and emotional disclosure may carry different meanings across Christian denominations and across Jewish, Muslim, Hindu, and other religious contexts. They may fit less well in traditions or individuals whose sacred orientation is less personal or less centered on a single divine attachment figure.
Miner and colleagues compared conceptual issues in Islam and Christianity and argued that attachment language can be useful across both traditions while requiring attention to theological and cultural differences. Their analysis is especially important methodologically: an item that appears psychologically equivalent in English may invoke different assumptions about divine-human relationship in another religious framework. Miner et al., 2014
Recent cross-religious evidence expands the field but does not solve the generalizability problem. McLean and colleagues studied Christian, Muslim, and Hindu participants from the United States and India and found patterns consistent with parental correspondence, while maternal and paternal pathways differed by religious group. This is valuable contemporary evidence that the relation between human and divine attachment can vary across contexts; it is not a basis for treating three religions as internally uniform or for generalizing to all believers. McLean et al., 2026
Cross-cultural research therefore needs more than translation. Strong comparisons require evidence that items have similar meaning, factor structure, and measurement properties across groups. Without measurement invariance, apparent group differences may partly reflect how the questionnaire functions rather than genuine differences in attachment to God.
What Does Research Show About Attachment to God and Mental Health?
The broad pattern in observational research is that more secure attachment to God tends to correlate with better psychological functioning, while higher anxiety and avoidance often correlate with greater distress or lower well-being. The strength and form of these associations vary by outcome, sample, measure, and analytic model. The evidence supports association more strongly than causation. For the broader evidence base on protective and adverse associations, confounding, heterogeneity, and causal limits, see Religion, Spirituality, and Mental Health: What Research Shows.
In a nationwide U.S. Presbyterian sample, Bradshaw and colleagues found secure attachment to God was inversely associated with psychological distress and anxious attachment was positively associated with distress. The study also separated attachment from God imagery, but its cross-sectional observational design and denominational sample limit causal and cross-religious conclusions. Bradshaw, Ellison, & Marcum, 2010
Njusa and Scharmer reported across three studies that attachment to God contributed information about psychological well-being beyond adult attachment. Their outcomes included self-esteem, depressive symptoms, flourishing, and anxiety-related measures. Because the designs rely heavily on self-report and correlational associations, the results support distinct predictive relationships rather than a claim that attachment to God directly causes mental-health outcomes. Njusa & Scharmer, 2020
Henderson and Kent demonstrated another important limitation of simple conclusions: the relationship between attachment to God and distress may not always be linear. Using national Baylor Religion Survey data, they reported a curvilinear relationship for avoidant/secure attachment and distress, while anxious attachment showed a more straightforward positive association with distress. This finding cautions against assuming that every one-unit change on an attachment scale has the same psychological meaning across the entire score range. Henderson & Kent, 2022
Systematic-review evidence
A 2026 systematic review by Facchino and colleagues examined attachment to God in relation to positive orientation indicators such as self-esteem, life satisfaction, and optimism. Fourteen eligible records covering 16 studies were included. Secure attachment to God was generally associated with more favorable indicators, while anxious and avoidant attachment tended to show less favorable associations. The review also emphasized heterogeneity, uncontrolled variables, the predominance of Christian participants, and high risk of bias in the longitudinal or cohort studies it assessed. Facchino et al., 2026
That evidence is informative but narrower than the phrase “attachment to God improves mental health.” The review did not establish a universal treatment effect, and the literature contains substantial design limitations. Better well-being could influence the way people experience God; religious community support could influence both; personality or life circumstances could affect both; and attachment scores may overlap with distress-related perceptions. Longitudinal, experimental, and intervention research is needed to clarify direction and mechanism.
Coping may be one pathway
One proposed mechanism is coping. Parenteau and colleagues found associations between insecure attachment to God and psychological adjustment that were statistically mediated by coping patterns in their cross-sectional study. Their authors explicitly noted that the design could not establish causality. The result is best interpreted as evidence that coping may be one pathway linking perceived relationship with God and adjustment, not as proof of a causal sequence. Parenteau et al., 2019
Other plausible pathways include perceived support, hope, meaning, emotion regulation, religious community relationships, interpretation of stressful events, and patterns of prayer or worship. These mechanisms overlap but should be measured separately. Treating every positive association as an “attachment effect” would overstate the evidence.
Attachment to God, Prayer, and Worship
Prayer and worship can be contexts in which attachment processes become psychologically salient. A person may seek proximity through prayer, turn toward God under threat, experience comfort during worship, or interpret a lack of felt connection as separation. Experimental and theoretical attachment research makes these processes plausible, and some studies directly test attachment-related activation. Granqvist et al., 2012
The psychological functions of prayer and the mental-health evidence for prayer are broader topics than attachment to God. Frequency of prayer does not by itself reveal attachment security, and an attachment score does not tell researchers what a prayer means to the person. Prayer may express gratitude, confession, petition, contemplation, lament, obligation, communal identity, or other religious purposes. Attachment is one interpretive lens among several.
Can Attachment to God Change?
Attachment-to-God patterns are not necessarily fixed. Theories of attachment allow for both continuity and revision as relationships, contexts, and expectations change. Religious conversion or deconversion, major life events, periods of suffering, community change, pastoral relationships, psychotherapy, family transitions, and new interpretations of faith can all coincide with changes in how a person experiences God.
The 2026 Cambridge synthesis describes religious/spiritual attachment as developing across the lifespan and discusses potential transformation through relationships and psychospiritual intervention. The existence of change does not imply that therapy should aim to produce a particular religious attachment style. Change is an empirical and clinical possibility whose desirability depends on the person’s own values, goals, and religious context. Davis & Granqvist, 2026
Longitudinal evidence remains thinner than cross-sectional evidence. Researchers should therefore avoid strong claims about what causes attachment to God to become more secure or insecure. A change in questionnaire score can reflect a durable relational shift, a stressful period, altered beliefs, changes in religious participation, measurement effects, or several processes at once.
Clinical Interpretation and Culturally Competent Psychotherapy
Attachment to God can be clinically relevant when a client considers religion or spirituality important and wants it included in therapy. It may help a clinician ask how the client experiences trust, closeness, reassurance, distance, anger, uncertainty, or support in relation to God. The purpose is to understand the client’s psychological and cultural world, not to evaluate doctrine or prescribe a religious conclusion.
A competent clinician should not infer pathology from ordinary religious commitment, prayer, worship, experiences of divine presence, doubt, or changes in felt closeness to God. Clinical assessment depends on broader evidence: distress, duration, impairment, reality testing where relevant, symptom patterns, risk, cultural and religious norms, and the person’s own interpretation. A research questionnaire cannot replace that assessment.
The same principle applies in the opposite direction. A therapist should not assume that greater “security” on an attachment-to-God measure is always the client’s goal, or that lower attachment anxiety automatically means clinical improvement. Religious traditions differ in the meanings they assign to dependence, reverence, fear, surrender, intimacy, lament, silence, and uncertainty. Psychotherapy should work within the client’s informed goals and values rather than imposing a standardized spiritual ideal.
When attachment concepts are useful, they can support questions rather than conclusions. A clinician might explore what the client experiences during stress, how religious community relationships interact with the perceived relationship with God, whether spiritual struggle intensifies or relieves distress, and whether the client wants religious resources included in treatment. Collaboration and cultural humility are more important than fitting every religious experience into an attachment label.
What Attachment-to-God Scores Cannot Tell You
• They cannot diagnose a mental disorder. Anxiety and avoidance on an attachment-to-God questionnaire are dimensions of a research construct, not psychiatric diagnoses.
• They cannot measure the truth or validity of a religious belief. Psychology can study reported experience, cognition, emotion, and behavior without adjudicating sacred claims.
• They cannot rank faith, holiness, spiritual maturity, or divine favor. Psychometric scores answer operationalized research questions.
• They cannot be assumed to mean the same thing in every religious or cultural setting. Translation, theology, language, and measurement invariance matter.
• They cannot establish childhood causation from a single adult questionnaire. Correspondence is probabilistic and developmental evidence is more complex.
• They cannot prove that attachment to God caused well-being or distress. Much of the literature is correlational and vulnerable to reverse causation and confounding.
Major Research Limitations
Predominantly Christian and Western samples
A large share of foundational attachment-to-God research was conducted with Christian participants in the United States or other Western contexts. The Brazilian AGI validation notes that most U.S. studies it identified used Christian institutions or churches. The 2026 systematic review likewise highlighted the predominance of Christian participants. The field is becoming more cross-religious, but the historical sampling imbalance still limits universal claims. August, Esperandio, & Escudero, 2018 Facchino et al., 2026
Self-report and common-method bias
Most studies ask the same participant to report attachment to God and psychological outcomes at one time. Shared method, mood, response style, social desirability, or current stress can inflate associations. Some studies statistically control relevant variables, but statistical adjustment cannot fully substitute for independent measures, repeated assessment, prospective designs, or experimental evidence.
Cross-sectional designs and reverse causality
If distress and attachment anxiety are measured at the same time, several causal stories remain possible. Insecurity could contribute to distress; distress could make the relationship with God feel less secure; a third factor could influence both; or the relationship could be reciprocal. Cross-sectional correlations cannot decide among these possibilities.
Construct overlap
Attachment anxiety to God can overlap conceptually with spiritual struggle, negative God representations, reassurance seeking, or religious coping. Avoidance can overlap with low salience of a personal relationship with God. Researchers need discriminant-validity tests and models that include neighboring constructs rather than assuming each questionnaire captures a completely independent process.
Theological and semantic variation
Words such as intimacy, dependence, abandonment, closeness, fear, or love can have different meanings across communities. A person may reject the premise of “intimacy with God” for theological reasons while remaining deeply religious. Another may understand divine closeness as central to faith. Psychometric interpretation is strongest when the wording makes sense inside the participant’s religious framework.
Categorizing continuous dimensions
Turning anxiety and avoidance scores into four styles can be convenient for communication, but thresholds are often sample-dependent. A person just above and just below a cut point may be psychologically very similar. Dimensional analyses generally preserve more information and reduce the risk of turning probabilistic research patterns into identity labels.
How to Read the Evidence Responsibly
The construct is most useful when it is treated as one layer of a multidimensional religious life. A person’s perceived relationship with God can interact with human attachment, religious identity, community, stress, coping, religious meaning-making, prayer, and life history. Each construct should remain distinguishable enough to be tested rather than folded into a single story.
Evidence is strongest for three broad conclusions. First, many believers describe God in ways that meet important psychological criteria for an attachment figure. Second, anxiety and avoidance provide useful dimensions for measuring individual differences in this perceived relationship. Third, these dimensions show recurring associations with well-being and distress, while the causal direction, universality, and mechanisms remain incompletely established. Granqvist, Mikulincer, & Shaver, 2010 Facchino et al., 2026
The evidence is weakest when findings are converted into universal spiritual prescriptions. Research does not justify telling all people that they should experience God as an attachment figure, that one attachment score represents the ideal religious life, or that changing the score is a guaranteed route to mental health. These claims exceed what the measurements and study designs establish.
Practical Implications for Researchers
Researchers studying attachment to God should select measures whose assumptions fit the target population, report the religious and cultural composition of the sample, and test whether the measure functions adequately in the language and group being studied. A translated questionnaire should not be presumed equivalent simply because the words are grammatically accurate.
Studies should separate attachment to God from religiosity, God representations, coping, spiritual struggle, and interpersonal attachment whenever those constructs matter to the research question. Longitudinal designs are especially valuable for testing whether attachment predicts later adjustment, adjustment predicts later attachment, or both change together.
Cross-religious comparisons require particular care. Researchers should establish whether the underlying construct is meaningful in each tradition, test measurement invariance when possible, and avoid treating religions as homogeneous categories. Differences between denominations, schools, communities, levels of observance, personal theology, and cultural settings can matter as much as broad affiliation labels.
Practical Implications for Readers
For an individual reader, attachment-to-God language can provide a vocabulary for reflecting on a perceived relationship with God, especially around trust, comfort, closeness, reliance, fear of rejection, or distance. It works best as a descriptive framework. It becomes misleading when treated as a spiritual grade, a self-diagnosis, or a prediction of one’s future.
Periods of doubt, anger, grief, silence, perceived distance, or increased need for reassurance can occur within ordinary religious life. Their psychological meaning depends on context. If such experiences are accompanied by significant distress or impaired functioning, a culturally competent mental-health professional can assess the broader picture while respecting the person’s religious framework.
If someone belongs to a tradition in which personal attachment language for God feels inappropriate, the mismatch may be in the model rather than in the person. Psychological constructs are tools for describing patterns. They should fit the phenomenon they are used to study.
Frequently Asked Questions
What does attachment to God mean in psychology?
It means the person’s perceived relational-emotional bond with God analyzed through attachment theory. Researchers ask whether God is experienced as a source of closeness, security, comfort under threat, or confidence for exploration, and how anxiety or avoidance characterize that perceived relationship. The construct concerns human psychological experience rather than theological truth.
Is attachment to God the same as faith?
No. Faith can include belief, trust, commitment, practice, identity, values, community, and theological meaning. Attachment to God is narrower: it focuses on relational security, anxiety, avoidance, closeness, and attachment functions. The constructs can correlate without being identical.
What is the Attachment to God Inventory?
The Attachment to God Inventory is a self-report research measure developed by Beck and McDonald. The original version contains 28 items organized into Anxiety about Abandonment and Avoidance of Intimacy subscales. It was designed to operationalize attachment-to-God differences for research; it is not a psychiatric diagnostic instrument. Beck & McDonald, 2004
Does a high attachment-anxiety score mean someone has an anxiety disorder?
No. Attachment anxiety toward God is a construct-specific score about fears and uncertainty in a perceived relationship with God. Anxiety disorders are clinical diagnoses based on defined symptom patterns, duration, impairment, context, and professional assessment. The two uses of the word anxiety should not be conflated.
Does insecure human attachment cause insecure attachment to God?
Research supports some correspondence between interpersonal attachment and attachment to God, but the relationship is probabilistic rather than deterministic. Evidence also supports compensatory processes in which the relationship with God can take on a distinct security-regulating role. Cultural context, religious socialization, life events, later relationships, and development all matter. Granqvist, Mikulincer, & Shaver, 2010
Can someone have secure attachment to God and insecure attachment in human relationships?
Yes. The constructs are related in some studies but are not identical. A person can show different patterns across relationship domains, and attachment representations can be relationship-specific. Findings that God attachment contributes information beyond adult or parental attachment are one reason researchers measure the domains separately. Njusa & Scharmer, 2020
Can attachment to God change over time?
It can. Attachment theory allows for continuity and change, and religious/spiritual attachment can develop across the lifespan. The evidence base for exactly how and why it changes is less mature than the cross-sectional literature, so strong causal formulas are premature. Davis & Granqvist, 2026
Is secure attachment to God always better for mental health?
Research often finds more favorable well-being associations with security and more distress with anxiety or avoidance, but the literature is heterogeneous and largely observational. Associations can differ by outcome and may be nonlinear. A score is not a treatment target by itself, and individual clinical meaning depends on context. Facchino et al., 2026 Henderson & Kent, 2022
Is attachment to God studied outside Christianity?
Yes, including studies with Jewish, Muslim, and Hindu participants. However, the historical evidence base is still disproportionately Christian and Western. Cross-religious research is growing, and recent findings indicate that developmental pathways can differ across traditions and cultural settings. McLean et al., 2026
Can attachment-to-God scales be used with atheists or people who do not conceive of a personal God?
Often the construct will have poor conceptual fit if the respondent does not experience or endorse a relational divine figure presupposed by the items. Researchers should not force a score from a measure whose basic premise does not apply. Studies of nonreligious identity, spirituality without a personal deity, or other sacred orientations require measures designed for those phenomena.
Should psychotherapy try to make a client’s attachment to God more secure?
Psychotherapy should follow the client’s goals, culture, values, and informed preferences. Attachment-to-God concepts can be explored when relevant, but a therapist should not impose a theological model or treat one questionnaire profile as the universal spiritual goal. Clinical care should distinguish religious meaning from psychiatric assessment while respecting both.
Related Articles
References
August, H., Esperandio, M. R. G., & Escudero, F. T. (2018). Brazilian validation of the Attachment to God Inventory (IAD-Br). Religions, 9(4), 103. https://doi.org/10.3390/rel9040103
Beck, R., & McDonald, A. (2004). Attachment to God: The Attachment to God Inventory, tests of working model correspondence, and an exploration of faith group differences. Journal of Psychology and Theology, 32(2), 92–103. https://doi.org/10.1177/009164710403200202
Bradshaw, M., Ellison, C. G., & Marcum, J. P. (2010). Attachment to God, images of God, and psychological distress in a nationwide sample of Presbyterians. The International Journal for the Psychology of Religion, 20(2), 130–147. https://doi.org/10.1080/10508611003608049
Davis, E. B., & Granqvist, P. (2026). Attachment, religion, and spirituality. Cambridge University Press. https://doi.org/10.1017/9781009501019
Facchino, A. P., D’Ettorre, M., Verrocchio, M. C., Fontanesi, L., & Marchetti, D. (2026). Attachment to God and positive orientation: A systematic review. Archive for the Psychology of Religion, 48(1), 91–109. https://doi.org/10.1177/00846724251338431
Granqvist, P., Mikulincer, M., & Shaver, P. R. (2010). Religion as attachment: Normative processes and individual differences. Personality and Social Psychology Review, 14(1), 49–59. https://doi.org/10.1177/1088868309348618
Granqvist, P., Mikulincer, M., Gewirtz, V., & Shaver, P. R. (2012). Experimental findings on God as an attachment figure: Normative processes and moderating effects of internal working models. Journal of Personality and Social Psychology, 103(5), 804–818. https://doi.org/10.1037/a0029344
Henderson, W. M., & Kent, B. V. (2022). Attachment to God and psychological distress: Evidence of a curvilinear relationship. Journal for the Scientific Study of Religion, 61(1), 161–177. https://doi.org/10.1111/jssr.12767
McLean, A., Amin, S., Poojar, R., & Kamble, S. V. (2026). Attachment to God across religions: Fathers and mothers matter differently in Christianity, Islam, and Hinduism. Religions, 17(7), 853. https://doi.org/10.3390/rel17070853
Miner, M., Ghobary, B., Dowson, M., & Proctor, M.-T. (2014). Spiritual attachment in Islam and Christianity: Similarities and differences. Mental Health, Religion & Culture, 17(1), 79–93. https://doi.org/10.1080/13674676.2012.749452
Njusa, D. M., & Scharmer, A. (2020). Evidence that God attachment makes a unique contribution to psychological well-being. The International Journal for the Psychology of Religion, 30(3), 178–201. https://doi.org/10.1080/10508619.2020.1723296
Parenteau, S. C., Hurd, K., Wu, H., & Feck, C. (2019). Attachment to God and psychological adjustment: God’s responses and our coping strategies. Journal of Religion and Health, 58(4), 1286–1306. https://doi.org/10.1007/s10943-019-00765-2
Rowatt, W., & Kirkpatrick, L. A. (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), 637–651. https://doi.org/10.1111/1468-5906.00143
Sim, T. N., & Loh, B. S. M. (2003). Attachment to God: Measurement and dynamics. Journal of Social and Personal Relationships, 20(3), 373–389. https://doi.org/10.1177/0265407503020003006
Stulp, H. P., Koelen, J., Schep-Akkerman, A., Glas, G. G., & Eurelings-Bontekoe, L. (2019). God representations and aspects of psychological functioning: A meta-analysis. Cogent Psychology, 6(1), 1647926. https://doi.org/10.1080/23311908.2019.1647926