Religious Locus of Control Scale

Abstract

The Religious Locus of Control Scale (RLCS; Gabbard, Howard, & Tageson, 1986) is a specialized psychometric revision of Julian B. Rotter’s (1966) classic Internal versus External Control of Reinforcement Scale. Designed to overcome systemic construct underrepresentation and confounding measurement artifact within religious populations, the instrument assesses generalized expectancies regarding whether life outcomes are governed by individual agency (internal control) or transcendent spiritual and supernatural entities (external control). In Rotter’s original 29-item forced-choice inventory, external locus of control was operationalized predominantly through the conceptual lens of arbitrary chance, blind luck, and impersonal fate. Psychometric researchers discovered that devout and fundamentalist believers systematically rejected external items not out of high personal self-efficacy or internal agency, but due to theological dissonance with the concept of random chance, leading to artificially elevated internal scores. To resolve this measurement bias, Gabbard, Howard, and Tageson modified eight critical item pairs to incorporate terminology denoting divine agency, providence, spiritual assistance, and supernatural forces, alongside a parallel “neutral” revision. The complete RLCS comprises 29 forced-choice dyadic items (23 scored items and 6 filler items designed to obscure test intent), scored additively in the direction of external control (ranging from 0 to 23). Empirical validation across religious fundamentalist adults (N = 115) and university undergraduates (N = 184) demonstrated strong convergent validity with Levenson’s Multidimensional Locus of Control Scales and high intercorrelations with the original Rotter scale (r = .84 to .91), while effectively capturing theological external attributions that standard generalized scales miss entirely.

Keywords

Religious Locus of Control Scale, Internal-External Control, Julian Rotter, Gabbard Howard Tageson, God Control, Spiritual Attribution, Religious Coping, Measurement Bias, Forced-Choice Scale, Psychometrics

Authors

The Religious Locus of Control Scale was developed and validated by:

• Clinton E. Gabbard, Ph.D. — Department of Psychology, University of Notre Dame. Dr. Gabbard’s empirical scholarship has focused on the intersection of clinical personality assessment, cognitive belief structures, and the psychological dimensions of religious orthodoxies.
• George S. Howard, Ph.D. — Professor Emeritus of Psychology, University of Notre Dame. Dr. Howard is a prominent methodologist and theoretical psychologist known for his extensive contributions to narrative psychology, self-determination, research methodology, and philosophical issues in behavioral science.
• C. W. Tageson, Ph.D. — Department of Psychology, University of Notre Dame. Dr. Tageson’s scholarly agenda centered on humanistic counseling psychology, existential dimensions of personality, and religious integration in therapeutic contexts.

Purpose

The primary purpose of the Religious Locus of Control Scale is to provide an ontologically sensitive, psychometrically valid assessment of generalized control expectancies in individuals holding deep religious or spiritual worldviews. For decades, locus of control has functioned as one of the most widely researched constructs in personality, social, and clinical psychology. Derived from Rotter’s (1954, 1966) social learning theory, locus of control reflects a person’s generalized expectancy regarding whether life outcomes, reinforcements, and consequences are contingent upon their own behaviors, effort, and stable internal characteristics (internal locus of control) or are determined by external factors outside personal mastery (external locus of control).

However, the operationalization of external locus of control in Rotter’s 1966 inventory heavily relied on secularized notions of stochastic randomness, encompassing terms such as “accidental happenings,” “luck,” “chance,” “fate,” and “flipping a coin.” Gabbard, Howard, and Tageson (1986) recognized that this specific phrasing produces severe measurement distortion when administered to devout individuals, such as adherents of conservative Christian, Orthodox, or fundamentalist traditions. From a strict theistic framework, the cosmos is neither arbitrary nor chaotic; it is orchestrated by an omnipotent, sovereign deity exercising purposeful providence. When confronted with a forced-choice dichotomy between personal agency (e.g., studying hard produces good grades) and arbitrary chance (e.g., grades are a matter of luck), devout respondents consistently choose the internal option—not because they discount divine sovereignty, but because their theology explicitly repudiates random chance.

Consequently, orthodox religious populations routinely obtained spuriously “internal” scores on the original Rotter scale, obscuring genuine external spiritual surrender. The RLCS was formulated to rectify this systemic measurement artifact by re-operationalizing external control through constructs of spiritual guidance, divine providence, supernatural forces, and God’s assistance. In clinical and health psychology contexts, the RLCS enables researchers and practitioners to differentiate between maladaptive passivity (fatalistic external control) and adaptive religious coping (surrendering outcomes to God while maintaining moral or behavioral diligence), providing nuanced predictive power for mental health outcomes, treatment compliance, and existential resilience.

Psychological Construct

The psychological construct evaluated by the Religious Locus of Control Scale is the multidimensional continuum of generalized control expectancies, specifically delineating internal personal agency from divine and supernatural external attribution. The scale operationalizes control across two fundamental domains:

1. Internal Agency (Personal Causality)

Internal control represents the generalized expectancy that reinforcements, life achievements, leadership efficacy, interpersonal conflicts, and socio-political dynamics are the direct consequences of one’s personal choices, moral responsibility, skill acquisition, and deliberate effort. In the RLCS, internal agency is captured across academic, interpersonal, and civic arenas. For instance, endorsing statements such as “People’s misfortunes result from the mistakes they make” (Item 2b) or “Becoming a success is a matter of hard work; no other powerful forces are at work” (Item 11a) reflects an ontological posture wherein human beings are viewed as the primary architects of their temporal trajectories. Internal agency reflects personal accountability, cognitive self-efficacy, and perceived behavioral mastery.

2. Transcendent External Control (God and Spiritual Forces)

External locus of control in the RLCS is conceptualized not as passive surrender to blind chaos, but as an acknowledgment of transcendent, supernatural, and divine governance over human affairs. This dimension encapsulates several distinct theological and spiritual attributions:

• Divine Governance and Spiritual Assistance: The belief that human leadership, personal triumph, and purposeful direction require active divine empowerment (e.g., “Without God’s help, one cannot be an effective leader,” Item 6a).
• Providential Causality: The perspective that historical events, personal fortunes, and existential occurrences are governed by supernatural designs beyond human intellectual mastery (e.g., “Most people can’t realize the extent to which their lives are controlled by supernatural happenings which man can’t understand,” Item 18a).
• Spiritual Interdependence versus Self-Sufficiency: The explicit rejection of autonomous self-reliance in favor of reliance on higher powers, directly contrasting secular self-determination against spiritual yielding (e.g., evaluating whether “trusting to spiritual assistance has never turned out as well for me as making a decision to take a definite course of action,” Item 9b).

By classifying divine intervention as a legitimate form of external reinforcement control, the RLCS provides a granular psychometric lens capable of capturing how theistic surrender coexists with, or diverges from, generalized beliefs about personal efficacy.

Theoretical Framework

The foundational architecture of the Religious Locus of Control Scale rests upon Julian Rotter’s Social Learning Theory (1954), later expanded by multidimensional personality models (Levenson, 1974) and the psychology of religion (Pargament, 1997). Rotter posited that the potential for a given behavior to occur in a specific psychological situation is a function of the expectancy that the behavior will lead to a particular reinforcement, multiplied by the psychological value of that reinforcement:

BP_{x, s1, Ra} = f(E_{x, Ra, s1} & RV_{a, s1})

Where BP is behavior potential, E is expectancy, and RV is reinforcement value within situation s1. When individuals face ambiguous, novel, or complex situations, their behavior is largely directed by generalized expectancies abstracted from past learning histories. Generalized expectancies for control fall along an internal-external axis. Rotter recognized that external control was heterogeneous, but his primary measurement tool treated externality as a single aggregate construct heavily saturated with chance and luck.

Levenson (1974) advanced this theoretical landscape by demonstrating that external control must be disaggregated into at least two distinct dimensions: Chance (unpredictable, chaotic forces) and Powerful Others (orderly, predictable, but external human agents). Gabbard, Howard, and Tageson (1986) expanded Levenson’s critique into the theological sphere. They argued that within the psychological ecology of theistic believers, “God” represents the ultimate “Powerful Other.” However, unlike mundane powerful others (such as authoritarian employers or political figures), God is perceived as benevolent, omniscient, and perfectly just.

Consequently, attributing control to God does not carry the same psychological connotations of helplessness, depression, or external passivity that characterizes attribution to secular chance or oppressive human others. This aligns closely with Pargament’s (1988, 1997) tripartite model of religious coping and locus of control, which delineates: (a) Collaborative Coping, where the individual and God share active responsibility; (b) Deferring Coping, where active agency is completely relinquished to God; and (c) Self-Directing Coping, where God gives humans the freedom and capacity to solve their own problems. The RLCS operates directly at this conceptual crossroads, providing the empirical apparatus needed to test whether religious external attributions mirror the psychological vulnerabilities typically associated with secular externality.

Validity

Empirical validation of the Religious Locus of Control Scale was systematically detailed by Gabbard, Howard, and Tageson (1986) across two distinct psychodemographic cohorts: a religious fundamentalist sample (N = 115 adults from two Midwestern Christian fundamentalist congregations) and a general university sample (N = 184 undergraduates from a public Midwestern university). The validation battery comprised Rotter’s original scale, the Religious Revision (RLCS), the Neutral Revision, Levenson’s (1974) Multidimensional Locus of Control Scales (Internal, Powerful Others, and Chance subscales), the Nowicki-Duke Locus of Control Scale (1974), and the Dies Projective Measure of Locus of Control (1968).

Construct and Known-Groups Validity

Construct validity was demonstrated through marked known-groups divergence. On a brief pre-experimental religiosity index, the fundamentalist cohort scored significantly higher in theological orthodoxy and religious practice than the university student cohort (p < .001). When administered Rotter’s original scale, the fundamentalist group displayed a markedly low external mean (M = 7.23, SD = 3.39), which would traditionally categorize them as highly “internal.” However, when evaluated using the RLCS (Religious Revision), their external control mean shifted substantially upward to M = 11.09 (SD = 3.05)—an absolute mean elevation of 3.86 points. Conversely, the general college sample exhibited minimal change between Rotter’s original scale (M = 10.40, SD = 4.32) and the RLCS (M = 11.26, SD = 4.08), an increase of merely 0.86 points. This interaction confirms that the RLCS successfully unmasked latent external expectancies in devout individuals that were systematically suppressed by the chance-laden vocabulary of the original inventory.

Convergent and Discriminant Validity

Because the RLCS, the Neutral Revision, and the original Rotter scale share 15 identical scored items, baseline inter-scale correlations were inevitably high:

• Rotter Original vs. Religious Revision (RLCS): r = .84
• Rotter Original vs. Neutral Revision: r = .90
• Religious Revision vs. Neutral Revision: r = .91

Convergent validity was further corroborated through correlations with Levenson’s (1974) multidimensional scales. The RLCS exhibited moderate, statistically significant positive correlations with Levenson’s Powerful Others scale and moderate negative correlations with Levenson’s Internal scale, confirming that the religious modifications preserve core locus of control variance. In contrast, the RLCS demonstrated near-zero correlations with the Dies (1968) projective locus of control measure and low associations with the Nowicki-Duke (1974) instrument. This divergent pattern was interpreted by the authors as reflecting the psychometric limitations and poor construct saturation of projective and non-college generalized scales, rather than a failure of RLCS validity.

Reliability

In the original validation study by Gabbard, Howard, and Tageson (1986), formal internal consistency coefficients (Cronbach’s alpha, Kuder-Richardson Formula 20) and test-retest temporal stability coefficients were not explicitly reported for the religious sample. This omission represents a acknowledged limitation in the initial psychometric literature, arising primarily from the study’s analytical focus on cross-scale mean comparisons and structural validity modeling.

However, the psychometric architecture of the RLCS directly mirrors Rotter’s (1966) original instrument, which historically yields KR-20 reliability coefficients ranging between .65 and .79 across diverse populations, and one-month test-retest reliability coefficients ranging from .72 to .83. Given that 15 of the 23 scored items in the RLCS are drawn verbatim from Rotter’s validated pool, and considering the exceptionally high parallel-forms correlation between the original instrument and the Religious Revision (r = .84), the expected internal consistency of the RLCS aligns closely with Rotter’s baseline parameters (estimated α ≈ .70–.78). Subsequent psychometric investigations utilizing religious control adaptations have affirmed that forced-choice dichotomous formats typically yield lower internal consistency estimates than continuous multi-point Likert formats due to restricted item variance, but retain solid rank-order temporal stability.

Factor Analysis

Gabbard, Howard, and Tageson (1986) conducted confirmatory factor analyses (CFA) to examine whether the Religious Revision, the Neutral Revision, and the original Rotter inventory shared an underlying unidimensional control latent space or diverged into multi-factor configurations across samples. The authors attempted to specify measurement models linking individual item subsets to latent internal and external factors.

However, the analytical utility of their confirmatory factor model was compromised by methodological constraints. Because the three scale variants share 15 identical scored items, the residual covariance matrices contained substantial artificial collinearity and correlated measurement error. When item overlap is high, standard maximum likelihood CFA algorithms generate inflated goodness-of-fit statistics or fail to reach structural convergence unless error covariances are explicitly freed and modeled. Despite these modeling limitations, exploratory analyses revealed that:

• Item Rewriting Integrity: The eight altered items (Items 2, 6, 9, 11, 15, 16, 18, and 25) maintained strong primary factor loadings (> .40) onto the generalized external control factor when evaluated within the fundamentalist religious cohort.
• Construct Separation: In the religious sample, the rewritten items loaded distinctively away from secular chance items, confirming that “God control” and “chance control” operate as distinct latent dimensions in orthodox populations rather than interchangeable indicators of a single general external factor.

Instrument / Measurement Tool

The operational characteristics and administration parameters of the Religious Locus of Control Scale are structured as follows:

• Instrument Name: Religious Locus of Control Scale (RLCS; Gabbard, Howard, & Tageson, 1986).
• Alternative Title: Internal vs. External Control Religious Revision.
• Test Format: Paper-and-pencil or computer-administered forced-choice psychometric inventory.
• Item Composition: 29 paired statements (dyads labeled ‘a’ and ‘b’).
• Scoring Items: 23 scored dyads assessing internal versus external expectancies.
• Distractor / Filler Items: 6 non-scored dyads (Items 1, 8, 14, 19, 24, and 27) included to mask the test’s intent and mitigate demand characteristics.
• Altered Items: 8 specific items (Items 2, 6, 9, 11, 15, 16, 18, and 25) modified from Rotter’s original scale to substitute chance/luck terminology with spiritual, divine, and supernatural constructs.
• Response Mechanism: Forced-choice dichotomous selection. For each dyad, the respondent must choose the single statement with which they more strongly agree.
• Scoring Procedure: Additive sum of external choices endorsed across the 23 scored items. Total scores range from 0 (maximum internal locus of control) to 23 (maximum external locus of control).
• Item Orientation Balance: To mitigate mechanical response bias, the external alternative appears first (‘a’) in 12 pairs and second (‘b’) in 11 pairs.

Permissions & Fee and Test Year

The Religious Locus of Control Scale was published in 1986 in the Journal of Research in Personality. The instrument is considered an open-access psychometric instrument for non-commercial academic research and clinical evaluation, subject to standard scholarly attribution. The intellectual property rights of the scale’s modifications reside with the original authors (Clinton E. Gabbard, George S. Howard, and C. W. Tageson) and the publisher, Academic Press (now Elsevier Inc.). No licensing fees are required for educational or dissertation research. Researchers conducting empirical investigations may utilize and reproduce the items provided appropriate formal citation to the 1986 seminal publication is maintained.

References

• Dies, R. R. (1968). Development of a projective measure of perceived locus of control. Journal of Projective Techniques and Personality Assessment, 32(5), 487–490. https://doi.org/10.1080/0091651X.1968.10120525
• Gabbard, C. E., Howard, G. S., & Tageson, C. W. (1986). Assessing locus of control with religious populations. Journal of Research in Personality, 20(3), 292–308. https://doi.org/10.1016/0092-6566(86)90137-0
• Levenson, H. (1974). Activism and powerful others: Distinctions within the concept of internal-external control. Journal of Personality Assessment, 38(4), 377–383. https://doi.org/10.1080/00223891.1974.10119988
• Nowicki, S., & Duke, M. P. (1974). A locus of control scale for noncollege as well as college students. Journal of Personality Assessment, 38(2), 136–137. https://doi.org/10.1080/00223891.1974.10119952
• Pargament, K. I. (1997). The psychology of religion and coping: Theory, research, practice. Guilford Press.
• Pargament, K. I., Kennell, J., Hathaway, W., Grevengoed, N., Newman, J., & Jones, W. (1988). Religion and the problem-solving process: Three styles of coping. Journal for the Scientific Study of Religion, 27(1), 90–104. https://doi.org/10.2307/1387404
• Rotter, J. B. (1954). Social learning and clinical psychology. Prentice-Hall. https://doi.org/10.1037/10788-000
• Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), 1–28. https://doi.org/10.1037/h0092976

Items of the Scale