Attitudes Toward God Scale -9 (ATGS-9)

Abstract

The Attitudes Toward God Scale -9 (ATGS-9) is a brief, psychometrically robust self-report instrument developed to assess an individual’s emotional and cognitive orientations toward the divine. Originating from the broader psychometric tradition of measuring relational spirituality and religious coping, the ATGS-9 was refined by Benjamin T. Wood, Everett L. Worthington Jr., Julie Juola Exline, Ann Marie Yali, Jamie D. Aten, and Mark R. McMinn (2010) as an abbreviated version of earlier multi-item inventories. The scale conceptualizes attitudes toward God along two distinct, orthogonal dimensions: Positive Attitudes toward God (often characterized as divine comfort, perceived benevolence, and relational closeness; 5 items) and Negative Attitudes toward God (characterized by anger, perceived abandonment, and relational strain; 4 items). Administered using an 11-point response format ranging from 0 (“Not at all”) to 10 (“Extremely”), the ATGS-9 minimizes respondent burden while preserving strong psychometric integrity. Extensive validation studies demonstrate that the two-factor structure exhibits high internal consistency, with Cronbach’s alpha coefficients typically exceeding .90 for Positive Attitudes and .80 to .88 for Negative Attitudes across diverse samples, including university undergraduates, community adults, medical patients, and psychiatric populations. Confirmatory factor analyses consistently support the two-dimensional model over a unidimensional continuum, affirming that positive and negative feelings toward God can co-occur within the same individual. The scale demonstrates profound convergent validity with measures of spiritual struggle, attachment to God, positive and negative religious coping, depressive symptomatology, and subjective well-being. Consequently, the ATGS-9 serves as an essential tool in clinical psychology, counseling, behavioral medicine, and empirical studies exploring the psychological determinants of spiritual health and distress.

Keywords

Attitudes Toward God Scale, ATGS-9, religious coping, spiritual struggle, anger at God, relational spirituality, attachment to God, divine comfort, psychometrics, religious strain

Authors

The development and empirical validation of the Attitudes Toward God Scale -9 (ATGS-9) represents a collaborative effort among prominent researchers in the fields of clinical psychology, counseling psychology, and the psychology of religion and spirituality:

• Benjamin T. Wood, Ph.D. — Department of Psychology, Virginia Commonwealth University, Richmond, Virginia. Dr. Wood’s research focuses on the intersection of religious coping, interpersonal forgiveness, and relational spirituality.
• Everett L. Worthington Jr., Ph.D. — Commonwealth Professor Emeritus of Psychology, Virginia Commonwealth University, Richmond, Virginia. A world-renowned expert on forgiveness, humility, and the integration of psychology and Christian faith, Dr. Worthington has authored over 40 books and hundreds of peer-reviewed articles.
• Julie Juola Exline, Ph.D. — Professor of Psychological Sciences, Case Western Reserve University, Cleveland, Ohio. Dr. Exline is a leading authority on spiritual struggles, supernatural beliefs, and religious anger, having pioneered foundational conceptual and empirical models on anger directed toward God.
• Ann Marie Yali, Ph.D. — Associate Professor of Psychology, City University of New York (CUNY), Hunter College, New York, New York. Dr. Yali’s scholarly work investigates religious and spiritual coping mechanisms, health psychology, and the psychological sequelae of medical crises.
• Jamie D. Aten, Ph.D. — Founder and Co-Director of the Humanitarian Disaster Institute, Wheaton College, Wheaton, Illinois. Dr. Aten specializes in disaster psychology, spiritual trauma, and faith-based interventions in humanitarian contexts.
• Mark R. McMinn, Ph.D., ABPP — Professor Emeritus of Psychology, George Fox University, Newberg, Oregon. Dr. McMinn is an expert on the integration of psychology and theology, positive psychology, and clergy mental health.

Purpose

The primary purpose of the Attitudes Toward God Scale -9 (ATGS-9) is to provide an efficient, psychometrically sound, and conceptually nuanced measurement of individuals’ affective and cognitive appraisals of God. Throughout history, psychological and religious scholarship often treated religious sentiment as a monolithic, uniformly positive construct, frequently equating religious commitment with psychological well-being. However, clinical and empirical observations reveal that relational engagement with the divine is inherently multidimensional, encompassing experiences of profound comfort alongside acute episodes of conflict, disillusionment, and distress.

Prior to the establishment of the ATGS-9, existing measures of religious attitudes either conflated positive and negative dimensions onto a single bipolar continuum (erroneously presuming that the presence of positive sentiment implies the absence of distress) or relied on extensive inventories that imposed substantial respondent burden. Grounded in earlier work by Exline, Yali, and Sanderson (2000), which identified spiritual and religious strains as salient predictors of depression and suicidality, the authors designed the ATGS-9 to capture both valences simultaneously while streamlining the administration process for clinical and field settings.

From an applied perspective, the ATGS-9 serves critical functions across diverse research and clinical domains:

• Clinical and Counseling Assessment: Mental health professionals utilize the ATGS-9 to assess spiritual struggles in clients facing acute grief, trauma, moral injury, or life-threatening medical diagnoses. Identifying unspoken anger toward God or feelings of divine abandonment allows therapists to address religious strain that may otherwise hinder psychological recovery.
• Medical and Palliative Care: In behavioral medicine, oncology, and end-of-life care, negative attitudes toward God are documented risk factors for diminished treatment adherence, spiritual pain, and poor quality of life. The ATGS-9 provides healthcare chaplains and palliative care teams with a rapid screening tool to identify patients who require targeted spiritual care.
• Empirical Research in the Psychology of Religion: The scale enables researchers to investigate how attachment security, personality traits (such as neuroticism and agreeableness), and life stressors shape internal working models of the divine. Its brief 9-item format makes it exceptionally well-suited for inclusion in longitudinal, epidemiological, and battery-heavy survey studies.

Psychological Construct

The ATGS-9 operationalizes attitudes toward God as an affective-cognitive evaluation reflecting an individual’s perceived relational bond with the divine. Rather than assessing abstract theological beliefs, dogmatic orthodoxy, or institutional religious behaviors (such as church attendance or prayer frequency), the instrument assesses personal, experiential, and relational dynamics. The construct comprises two correlated yet distinct dimensions:

1. Positive Attitudes toward God / Comfort (5 Items)

This subscale captures the extent to which an individual experiences God as a source of emotional solace, benevolent protection, personal validation, and relational support. It taps into positive affective states (feeling loved, nurtured, and supported) and affirming cognitive attributions (viewing God as trustworthy, all-powerful, and all-knowing). In psychological terms, this dimension reflects:

• Perceived Divine Benevolence: The attribution that the divine operates with goodwill, compassionate intention, and tender regard toward the self.
• Secure Relational Anchoring: The psychological experience of God as a protective figure capable of providing shelter and coping resources during adverse life events.
• Existential Reassurance: A sense of trust rooted in the belief that an omnipotent and omniscient entity oversees personal and cosmic events, fostering hope and reducing existential dread.

Item examples include acknowledging that one can “Trust God to protect and care for you” and experiencing feelings of being “loved by God” and “nurtured or cared for by God.” High scores on this subscale reflect robust positive religious coping, spiritual serenity, and secure spiritual attachment.

2. Negative Attitudes toward God / Anger (4 Items)

This subscale measures the affective and cognitive expressions of divine relational strain, characterized by subjective feelings of grievance, mistreatment, desertion, and resentment directed toward God. Historically marginalized in clinical psychology due to religious taboos against expressing anger toward the sacred, this dimension captures critical vulnerabilities:

• Direct Divine Anger: Conscious feelings of hostility, exasperation, or rage targeted at God, typically sparked by unexplained suffering, loss, or unanswered prayers.
• Perceived Divine Betrayal: The cognitive appraisal that God has failed to uphold an implicit covenant of protection, leading to the subjective conclusion that God has “let the individual down.”
• Perceived Divine Malevolence or Indifference: The appraisal of God as harsh, uncaring, punitive, or fundamentally “unkind.”
• Divine Abandonment: The painful phenomenological state of feeling deserted, isolated, or ignored by God during moments of existential crisis.

Item examples include feeling “angry at God,” perceiving that God has “let you down,” and feeling “abandoned by God.” High scores on this subscale are systematically linked to spiritual distress, internal moral conflicts, depressive symptoms, and existential alienation.

Theoretical Framework

The conceptual architecture of the ATGS-9 is grounded in several interrelated psychological frameworks, primarily attachment theory, cognitive appraisal theory, and the psychology of religious coping.

Attachment Theory Applied to the Divine

Pioneered by John Bowlby and expanded into the psychology of religion by Lee A. Kirkpatrick, attachment theory posits that humans possess an innate psychobiological system that seeks proximity to protective attachment figures under conditions of threat or distress. For believers, God frequently functions as an idealized primary attachment figure, fulfilling the four core attachment criteria:

1. Proximity Maintenance: Seeking cognitive and emotional closeness through prayer, meditation, and religious rituals.
2. Safe Haven: Turning to God for comfort, reassurance, and emotional stabilization during crises.
3. Secure Base: Deriving confidence and psychological equilibrium from the divine presence to explore the world and engage in social life.
4. Protest at Separation: Experiencing distress, anxiety, or anger when proximity is disrupted or perceived as severed.

Within this framework, the Positive Attitudes subscale corresponds directly to a secure attachment to God, wherein the divine is perceived as reliably accessible, responsive, and loving. Conversely, the Negative Attitudes subscale reflects insecure attachment dynamics—specifically anxious and avoidant attachment manifest as attachment protest. When an attachment figure fails to alleviate distress, the individual experiences anger and protest behaviors aimed at re-establishing connection or expressing grief over relational rupture.

Cognitive Appraisal and Attribution Theory

Grounded in the transactional stress model of Richard Lazarus and Susan Folkman, attitudes toward God are shaped by primary and secondary cognitive appraisals. When confronting traumatic or stressful life events, individuals appraise whether the event constitutes harm, loss, or threat, and subsequently make causal attributions regarding God’s role in the event:

• If God is appraised as an active ally providing internal strength, individuals generate positive attitudes (comfort, trust).
• If God is appraised as the sovereign cause of the tragedy, as negligent, or as willfully withholding protection, individuals experience negative attitudes characterized by moral outrage, perceived injustice, and affective anger.

Dual-System Relational Model

A foundational theoretical assumption underpinning the ATGS-9 is the non-mutually exclusive nature of positive and negative emotions. Borrowing from bivariate models of emotion (e.g., Cacioppo & Berntson), positive and negative evaluations of God do not occupy opposite poles of a single continuum. Rather, individuals can simultaneously experience profound love for God alongside deep disappointment or anger—a paradox frequently observed in clinical grieving processes and biblical laments (e.g., Job, the Psalms). The two-factor structure of the ATGS-9 provides the empirical fidelity required to map this psychological complexity.

Validity

The psychometric validity of the ATGS-9 has been established across diverse empirical studies involving college students, community populations, medical patients, and military veterans.

Construct and Structural Validity

Construct validity is evidenced by confirmatory factor analyses demonstrating that the hypothesized two-factor structure provides an exceptional fit to the data, consistently outperforming one-factor alternatives. Across validation cohorts (Wood et al., 2010), the two latent factors exhibit a modest to moderate negative correlation (ranging from r = -.20 to -.40), confirming that while positive and negative feelings are related, they operate as distinct psychometric dimensions rather than reciprocal endpoints.

Convergent Validity

The subscales demonstrate robust convergent correlations with established psychometric instruments:

• Positive Attitudes toward God: Correlates strongly and positively with the Brief RCOPE Positive Religious Coping subscale (r = .70 to .82), the Attachment to God Inventory (AGI) Secure Attachment scale, intrinsic religious motivation, frequency of personal prayer, and measures of life satisfaction and subjective well-being.
• Negative Attitudes toward God: Correlates substantially with the Brief RCOPE Negative Religious Coping subscale (r = .55 to .68), the Religious and Spiritual Struggles (RSS) Scale Divine Struggle subscale (r = .72 to .85), and anxious/avoidant attachment styles.

Discriminant Validity

Discriminant validity has been confirmed through differential relationships with psychological symptomatology and general affectivity:

• Negative Attitudes toward God uniquely account for variance in depressive symptoms, generalized anxiety, and hostility, even after controlling for demographic variables, global religiousness, and general trait negative affect (Wood et al., 2010; Exline et al., 2000).
• Positive Attitudes toward God demonstrate distinct pathways, correlating positively with psychological resilience, optimism, and post-traumatic growth, while showing weak or non-significant direct correlations with psychiatric symptom distress when negative attitudes are held constant.
• Both subscales diverge appropriately from social desirability scales, confirming that scores do not merely reflect tendencies to provide socially acceptable responses.

Predictive and Criterion-Related Validity

Longitudinal and cross-sectional investigations show that the Negative Attitudes subscale prospectively predicts worsening health outcomes, prolonged grief, and emotional exhaustion in clinical populations, including cancer patients, individuals recovering from natural disasters, and combat veterans suffering from moral injury. Conversely, high scores on the Positive Attitudes subscale predict adaptive coping, lower mortality risk in chronically ill cohorts, and accelerated recovery from bereavement.

Reliability

The ATGS-9 exhibits exceptional reliability across various psychometric parameters, indicating high internal consistency and temporal stability:

Internal Consistency

In the initial psychometric validation across multiple samples reported by Wood et al. (2010):

• Positive Attitudes toward God / Comfort (5 items): Cronbach’s alpha coefficients ranged from α = .92 to .96 across undergraduate cohorts (N = 1,024) and adult community samples. McDonald’s omega (ω) values similarly exceed .93, reflecting exceptional item homogeneity and common factor variance.
• Negative Attitudes toward God / Anger (4 items): Cronbach’s alpha coefficients ranged from α = .82 to .88. Despite containing only four items, the subscale maintains high internal consistency without redundancy.

Subsequent independent studies among diverse cultural, clinical, and religious populations have corroborated these findings, reporting alpha coefficients routinely exceeding .85 for both dimensions.

Test-Retest Reliability

Evaluation of temporal stability over intervals ranging from two to six weeks demonstrates strong test-retest reliability coefficients:

• Positive Attitudes: r = .84 to .89, indicating that positive relational sentiments toward God represent a relatively enduring, trait-like orientation under stable environmental conditions.
• Negative Attitudes: r = .71 to .78, reflecting both a stable underlying vulnerability to spiritual struggle and sensitivity to situational stressors, personal suffering, and life crises.

Factor Analysis

The structural dimensionality of the ATGS-9 was rigorously evaluated during its derivation from longer preliminary versions (such as the original 14-item and 22-item iterations examined by Exline and colleagues). Psychometricians employed both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) to verify the instrument’s structural integrity.

Exploratory Factor Analysis

In exploratory analyses using principal axis factoring with promax (oblique) rotation, scree plot evaluations and parallel analysis consistently revealed two distinct factors possessing eigenvalues greater than 1.0:

• Factor 1 (Positive Attitudes / Comfort): Accounted for approximately 48% to 55% of the total variance, with factor loadings ranging from .78 to .94 across items 1, 5, 6, 7, and 8. Cross-loadings onto the negative factor were negligible (all < .15).
• Factor 2 (Negative Attitudes / Anger): Accounted for an additional 14% to 20% of the total variance, with factor loadings ranging from .68 to .86 across items 2, 3, 4, and 9. Cross-loadings onto the positive factor were similarly minimal.

Confirmatory Factor Analysis

Confirmatory factor analyses conducted on independent validation samples verified the superiority of the two-factor oblique model over alternative structural configurations (e.g., unidimensional models or orthogonal two-factor models). Representative fit indices for the correlated two-factor model reported in the psychometric literature (Wood et al., 2010) include:

• Chi-Square / Degrees of Freedom: χ²/df ≤ 2.85
• Comparative Fit Index (CFI): ≥ .97 to .99
• Tucker-Lewis Index (TLI): ≥ .96 to .98
• Root Mean Square Error of Approximation (RMSEA): .042 to .058 (90% CI [.031, .069])
• Standardized Root Mean Square Residual (SRMR): .029 to .038

By comparison, a single-factor model forced through all 9 items resulted in severely degraded fit indices (CFI < .65, RMSEA > .18), mathematically confirming the essential multidimensionality of the construct. Measurement invariance testing across gender and diverse Christian denominations (Protestant vs. Catholic) confirmed metric and scalar invariance, supporting the validity of cross-group comparisons.

Instrument / Measurement Tool

The operational features and structural parameters of the ATGS-9 are summarized below:

• Instrument Name: Attitudes Toward God Scale -9 (ATGS-9)
• Authors: Benjamin T. Wood, Everett L. Worthington Jr., Julie Juola Exline, Ann Marie Yali, Jamie D. Aten, and Mark R. McMinn (2010; foundational work by Exline, Yali, & Sanderson, 2000)
• Construct Measured: Relational and affective attitudes toward God (positive comfort vs. negative anger/strain)
• Number of Items: 9 self-report items
• Subscales:
  • Positive Attitudes toward God / Comfort (5 items): Items 1, 5, 6, 7, 8
  • Negative Attitudes toward God / Anger (4 items): Items 2, 3, 4, 9
• Response Format: 11-point response scale from 0 (Not at all) to 10 (Extremely)
• Administration Time: Approximately 2 to 3 minutes
• Target Population: Adolescents and adults (ages 16+) who hold a concept of God or a supreme divine being (applicable to monotheistic traditions including Christianity, Judaism, and Islam, as well as individuals describing themselves as “spiritual but not religious”)
• Scoring Protocol:
  • Scores are computed separately for each subscale by calculating the mean or the sum of the constituent items.
  • There are no reverse-scored items; positive and negative items must not be combined into a single composite score, as doing so obscures critical clinical and empirical variance.
  • Positive Attitudes Score: Mean of items 1, 5, 6, 7, and 8 (range: 0 to 10) or sum (range: 0 to 50).
  • Negative Attitudes Score: Mean of items 2, 3, 4, and 9 (range: 0 to 10) or sum (range: 0 to 40).
• Interpretation Guidelines:
  • High Positive / Low Negative: Reflects secure spiritual attachment, effective positive religious coping, and adaptive spiritual well-being.
  • Low Positive / High Negative: Indicates severe spiritual alienation, acute religious strain, perceived abandonment, and elevated risk for depressive symptoms and moral distress.
  • High Positive / High Negative: Reflects an ambivalent or conflicted spiritual state where an individual values their relationship with God but is actively struggling through pain, grief, or theological dissonance.
  • Low Positive / Low Negative: Represents spiritual indifference, religious apathy, or low personal salience of the divine.

Permissions & Fee and Test Year

The Attitudes Toward God Scale -9 was formally published in its current 9-item format in 2010 through the American Psychological Association (APA) journal Psychology of Religion and Spirituality (with preliminary psychometric development dating to 2008 and earlier foundations in Exline et al., 2000). The instrument is widely accessible in the public academic domain for non-commercial research, academic inquiry, educational purposes, and clinical assessment.

No fee is required for non-commercial scholarly research. Clinicians and investigators are expected to credit the original authors through standard academic citation (Wood et al., 2010). For commercial applications, inclusion in copyrighted proprietary software, or broad institutional reprinting, permission should be sought from the authors and the American Psychological Association.

References

• Exline, J. J., Yali, A. M., & Sanderson, W. C. (2000). Guilt, discord, and alienation: The role of religious strain in depression and suicidality. Journal of Clinical Psychology, 56(12), 1481–1496. https://doi.org/10.3390/rel2010051
• Wood, B. T., Worthington, E. L., Jr., Exline, J. J., Yali, A. M., Aten, J. D., & McMinn, M. R. (2010). Development, refinement, and psychometric properties of the Attitudes Toward God Scale (ATGS-9). Psychology of Religion and Spirituality, 2(3), 148–167. https://doi.org/10.1037/a0018753

Items of the Scale