Religious Orientation Scale- Revised (ROS-R)

1. Abstract

The Religious Orientation Scale-Revised (ROS-R) is an extensively utilized psychometric instrument developed by Richard L. Gorsuch and Susan E. McPherson in 1989. Designed to refine, psychometrically strengthen, and adapt Gordon W. Allport and J. Michael Ross’s original 1967 Religious Orientation Scale, the ROS-R addresses critical methodological limitations, such as linguistic complexity and acquiescence response bias. The instrument comprises 14 self-report items calibrated across a 5-point Likert scale ranging from “I strongly disagree” to “I strongly agree” (with intermediate anchors: “I disagree”, “I’m not sure”, and “I tend to agree”). The scale assesses two foundational, multidimensional psychological orientations toward religion: Intrinsic Religiosity (I), which reflects religion as an internalized, master motive wherein faith is lived for its own sake, and Extrinsic Religiosity (E), in which religion serves utilitarian, instrumental, or social functions such as status, security, or fellowship.

Extensive psychometric investigations substantiate the structural validity and reliability of the ROS-R across diverse collegiate, community, and clinical cohorts. Confirmatory and exploratory factor analyses consistently isolate the intrinsic factor from extrinsic subcomponents, often revealing a further bifurcation of the extrinsic domain into Extrinsic-Personal (Ep) (comfort, safety, protection) and Extrinsic-Social (Es) (fellowship, interpersonal networking). The ROS-R demonstrates strong internal consistency, with Cronbach’s alpha coefficients routinely exceeding .80 for the Intrinsic scale and falling between .65 and .75 for the Extrinsic dimensions. Its balanced keying structure—incorporating reverse-scored items across both dimensions—effectively mitigates response set distortions, rendering it an indispensable standard in the psychology of religion, health psychology, and cross-cultural personality assessment.

2. Keywords

Religious Orientation Scale-Revised, ROS-R, Intrinsic Religiosity, Extrinsic Religiosity, Psychology of Religion, Richard L. Gorsuch, Psychometrics, Gordon Allport, Extrinsic-Personal, Extrinsic-Social, Religious Motivation, Measurement Validity

3. Authors

The Religious Orientation Scale-Revised was developed by Richard L. Gorsuch and Susan E. McPherson, based upon the theoretical foundation and earlier psychometric instruments created by Gordon W. Allport and J. Michael Ross (1967), as well as the Age-Universal scale introduced by Gorsuch and G. Daniel Venable (1983).

• Richard L. Gorsuch, Ph.D. (1937–2016): Senior Professor of Psychology in the Graduate School of Psychology at Fuller Theological Seminary, Pasadena, California. A premier psychometrician and leading authority on factor analysis and the psychology of religion, Dr. Gorsuch authored seminal textbooks including Factor Analysis (1983) and contributed profoundly to the empirical measurement of religious behavior and spirituality.
• Susan E. McPherson, Ph.D.: Clinical psychologist and researcher affiliated with the Graduate School of Psychology, Fuller Theological Seminary, who collaborated with Dr. Gorsuch on psychometric refinements and the development of single-item and revised multi-item measurement batteries for religious orientation.

4. Purpose

The primary purpose of the Religious Orientation Scale-Revised (ROS-R) is to provide a brief, psychometrically sound, and linguistically accessible assessment of motivational orientations toward religious commitment and practice. Although religion has historically played an ambiguous role in mental health and social behavior—at times correlating with prejudice, authoritarianism, and dogmatism, while at other times predicting altruism, psychological resilience, and existential well-being—psychological theory posited that these contradictory findings stemmed from conflating fundamentally disparate religious motivations. Gordon Allport formulated the intrinsic versus extrinsic paradigm to disentangle these conflicting dynamics, yet early empirical operationalizations suffered from severe psychometric flaws.

Specifically, the original 20-item Allport and Ross (1967) Religious Orientation Scale suffered from four pronounced limitations: (a) several items contained complex, archaic, or double-barreled vocabulary that precluded accurate administration to younger or less educated respondents; (b) the intrinsic scale contained zero negatively keyed items, rendering it exceptionally vulnerable to positive response bias and acquiescence; (c) the items failed to cleanly differentiate personal utilitarian motives from social utilitarian motives; and (d) overall scale length created fatigue in comprehensive survey batteries. Gorsuch and McPherson (1989) formulated the ROS-R to eliminate these structural barriers while maintaining theoretical fidelity to Allport’s paradigm.

In clinical and counseling psychology, the ROS-R is employed to examine how a patient’s core belief architecture influences coping styles, existential crises, therapeutic alliance, and vulnerability to psychological distress. For instance, high intrinsic religiosity frequently serves as a cognitive buffer against depression, existential anxiety, and trauma, fostering internal locus of control and moral meaning-making. Conversely, extrinsic religiosity has been repeatedly tied to heightened anxiety, cognitive dissonance, interpersonal dependency, and maladaptive coping strategies during adverse life events. In empirical research, the ROS-R functions as a standard control or independent predictor in investigations spanning social psychology (e.g., studies of racial prejudice, outgroup hostility, pro-social behavior), behavioral medicine (e.g., cardiovascular reactivity, immune functioning, oncology recovery), and developmental psychology.

5. Psychological Construct

The ROS-R evaluates religious commitment through a motivational lens, operating on the foundational assumption that people do not simply possess varying amounts of religiosity, but rather engage with religion in qualitatively distinct ways. The scale delineates two overarching dimensions: Intrinsic Religiosity and Extrinsic Religiosity, with the latter frequently categorized into distinct interpersonal and intrapersonal subcomponents.

Intrinsic Religiosity (I)

Intrinsic religiosity conceptualizes faith as an ultimate end in itself. An intrinsically oriented individual perceives religion as the master motive of existence, integrating doctrinal teachings into everyday decision-making, moral evaluations, and personal identity. Rather than exploiting religious institutions for secondary gains, the intrinsic individual conforms their life, desires, and objectives to their spiritual framework. Key markers of high intrinsic orientation include:

• Internalized Motivation: Spiritual growth, private prayer, and doctrinal reflection are pursued voluntarily due to genuine subjective value rather than external social pressure (e.g., Item 1: “I enjoy reading about my religion”; Item 4: “It is important to me to spend time in private thought and prayer”).
• Experiential Awareness: An intense, subjective awareness of the divine or transcendent presence in daily life (e.g., Item 5: “I have often had a strong sense of God’s presence”).
• Holistic Life Organization: The alignment of all secular domains—work, relationships, ethics—beneath a coherent religious framework (e.g., Item 7: “I try hard to live all my life according to my religious beliefs”; Item 12: “My whole approach to life is based on my religion”).
• Counter-Attitudinal Vigilance: Active rejection of moral compartmentalization, measured through reversed items (e.g., Item 10: “Although I am religious, I don’t let it affect my daily life”; Item 14: “Although I believe in my religion, many other things are more important”).

Extrinsic Religiosity (E)

In contrast, Extrinsic religiosity conceptualizes religion as an instrumental mechanism employed to attain external, secular goals. Extrinsically oriented individuals participate in religious rites, identify with denominations, or attend worship primarily because doing so provides immediate pragmatic utility. The individual “uses” religion rather than “living” it. Psychometrically and theoretically, this construct branches into two operational facets:

• Extrinsic-Social (Es): Using religious communities and worship settings as vehicles for social networking, interpersonal validation, community status, and peer companionship (e.g., Item 2: “I go to church because it helps me to make friends”; Item 11: “I go to church mostly to spend time with my friends”; Item 13: “I go to church mainly because I like seeing people I know there”).
• Extrinsic-Personal (Ep): Utilizing religious rituals and beliefs as an emotional tranquilizer, security mechanism, or crisis buffer to alleviate subjective discomfort and psychological vulnerability (e.g., Item 6: “I pray mainly to gain relief and protection”; Item 8: “What religion offers me most is comfort in times of trouble and sorrow”; Item 9: “Prayer is peace and happiness”).

6. Theoretical Framework

The theoretical bedrock of the ROS-R is rooted in Gordon Allport’s personality theory and his conceptualization of mature versus immature religious sentiment, formulated across his classic texts The Individual and His Religion (1950) and The Nature of Prejudice (1954). Allport was deeply concerned by empirical observations indicating that regular churchgoers frequently demonstrated greater ethnic prejudice, xenophobia, and authoritarian tendencies than non-churchgoers. To resolve this paradox, Allport posited a fundamental dichotomy: mature religion is integral, warm, comprehensive, and productive (intrinsic), whereas immature religion is defensive, magic, self-serving, and comfort-seeking (extrinsic).

Allport and Ross (1967) formally codified this paradigm, stating: “The extrinsically motivated person uses his religion, whereas the intrinsically motivated lives his.” In Allport’s functional view, extrinsic religiosity corresponds to early developmental stages wherein religion serves basic safety, ego-defense, and belonging needs. Intrinsic religiosity, by contrast, mirrors self-actualization and autonomous cognitive integration, transforming religion into a unifying philosophy of life that motivates altruism and universal regard.

Psychometrically, however, early implementations treated intrinsic and extrinsic orientations as mutually exclusive polar opposites on a single continuum. Subsequent theoretical and empirical work by Bernard Spilka, Richard L. Gorsuch, and Peter C. Hill established that intrinsic and extrinsic orientations are independent, orthogonal dimensions rather than polar opposites. An individual can score high on both (the “indiscriminately pro-religious” typology), low on both (the “indiscriminately anti-religious” or non-religious typology), or high on one and low on the other (pure intrinsic or pure extrinsic). Gorsuch and McPherson (1989) refined this framework by applying cognitive readability theory and structural equation modeling, showing that the extrinsic dimension inherently fractures into self-focused functional utility (Ep) and social belonging (Es).

7. Validity

The ROS-R has undergone extensive psychometric validation across thousands of participants in numerous cross-sectional, longitudinal, and experimental paradigms, showing robust construct, convergent, and discriminant validity.

Construct and Factorial Validity

Gorsuch and McPherson’s (1989) validation study demonstrated that the 14-item revised scale significantly outperformed the original 20-item Allport-Ross scale. Confirmatory factor analyses demonstrated that the ROS-R maintains a clear, interpretable factor structure. Factorial validity was further reinforced by demonstrating that reversed items loaded appropriately on their designated conceptual factors without generating the spurious method factors that frequently plague un-reversed survey instruments.

Convergent Validity

The Intrinsic scale demonstrates strong convergent correlations with alternative measures of religious commitment, including the Santa Clara Strength of Religious Faith Questionnaire (correlations ranging from $r = .72$ to $.84$), Duke University Religion Index (DUREL) intrinsic subscale ($r > .80$), frequency of private prayer ($r = .55$ to $.68$), and frequency of voluntary text reading. The Extrinsic scale, particularly the Extrinsic-Social component, correlates positively with organizational involvement metrics and social affiliation measures, while Extrinsic-Personal correlates moderately with religious coping measures designed around tension reduction, such as the Brief RCOPE’s passive religious deferral subscales ($r = .40$ to $.55$).

Discriminant and Predictive Validity

The ROS-R effectively differentiates distinct psychological outcomes:

• Prejudice and Social Tolerance: In line with Allport’s foundational hypothesis, meta-analytic reviews (e.g., Trimble, 1997) and subsequent studies show that Extrinsic Religiosity correlates positively with racial, ethnic, and outgroup prejudice ($r = .20$ to $.35$), whereas Intrinsic Religiosity correlates either neutrally or negatively with overt prejudice when social desirability is statistically controlled.
• Mental Health and Coping: Intrinsic religiosity consistently demonstrates inverse relationships with depressive symptomatology, generalized anxiety, and suicidal ideation, while predicting higher life satisfaction, optimism, and post-traumatic growth. In contrast, extrinsic religiosity exhibits positive associations with state-trait anxiety, neuroticism, external locus of control, and existential guilt (Jones, 2014).
• Social Desirability: While intrinsic items carry an inherent risk of socially desirable responding, Trimble’s (1997) meta-analysis confirmed that social desirability accounts for only a minor fraction of the variance in ROS-R scores, demonstrating that the intrinsic scale captures genuine motivational disposition rather than pure impression management.

8. Reliability

The ROS-R displays consistently high internal consistency and temporal stability across diverse research populations.

Internal Consistency

In the seminal psychometric evaluation conducted by Gorsuch and McPherson (1989), the 8-item Intrinsic Religiosity scale yielded a Cronbach’s alpha of .83, indicating high internal coherence among items measuring prayer, experiential awareness, and lifestyle integration. The 6-item Extrinsic Religiosity scale demonstrated an alpha coefficient of .65. When extrinsic items are parsed into their theoretical subcomponents, the Extrinsic-Social (Es; 3 items) and Extrinsic-Personal (Ep; 3 items) subscales generally yield alpha values between .60 and .72, which is considered acceptable given the brief, 3-item length of each subscale.

Replication studies across student samples, adult churchgoers, and clinical populations have consistently corroborated these coefficients. For example, Maltby and Day (1998) reported alpha coefficients of .81 for the Intrinsic scale, .70 for Extrinsic-Personal, and .68 for Extrinsic-Social in non-US cohorts, confirming stable internal structure across Western populations.

Temporal Stability

Test-retest reliability assessments over periods ranging from two weeks to six months show high stability coefficients. Intrinsic religiosity demonstrates test-retest correlations between $r = .82$ and $r = .89$, reflecting its nature as an enduring, trait-like motivational structure. The Extrinsic subscales display slightly lower but stable coefficients ranging between $r = .71$ and $r = .78$, mirroring minor fluctuations in environmental and social conditions.

9. Factor Analysis

Psychometric evaluations of the ROS-R using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have provided critical empirical validation for its structural integrity.

Exploratory Factor Analyses

Early principal axis factoring and principal component analyses with varimax and oblimin rotations confirmed that the 14 items cleanly separate into intrinsic and extrinsic domains. In Gorsuch and McPherson’s (1989) analysis, items 1, 3, 4, 5, 7, 10, 12, and 14 loaded predominantly onto the primary Intrinsic factor, with factor loadings ranging from .48 to .78. Crucially, reverse-scored items (Item 10: “Although I am religious, I don’t let it affect my daily life”; Item 14: “Although I believe in my religion, many other things are more important”) loaded robustly on the intrinsic dimension rather than coalescing into an uninterpretable response-artifact factor.

Confirmatory Factor Analyses & The Two- vs. Three-Factor Debate

Subsequent CFA investigations by researchers such as Kirkpatrick (1989), Genia (1993), and Maltby (1999) evaluated competing structural models:

• Two-Factor Model: Intrinsic (8 items) and Extrinsic (6 items) as correlated or orthogonal factors. While fit indices are generally acceptable ($\chi^2/df < 3.0$, $\text{CFI} \approx .90$, $\text{RMSEA} \approx .065$), residual covariance between specific extrinsic items frequently indicates local strain.
• Three-Factor Model: Intrinsic (8 items), Extrinsic-Personal (3 items: 6, 8, 9), and Extrinsic-Social (3 items: 2, 11, 13). This model routinely provides superior fit statistics across diverse samples: Comparative Fit Index ($\text{CFI}$) exceeding $.94$, Tucker-Lewis Index ($\text{TLI}$) exceeding $.93$, and Root Mean Square Error of Approximation ($\text{RMSEA}$) between $.042$ and $.055$.

These findings substantiate Gorsuch and McPherson’s theoretical expectation that extrinsic religiosity is not a strictly monolithic construct, but rather incorporates two distinct, moderately correlated utilitarian expressions ($r \approx .35$ to $.50$).

10. Instrument / Measurement Tool

• Instrument Name: Religious Orientation Scale-Revised (ROS-R)
• Alternative Designations: Revised I/E Scale; Gorsuch and McPherson I/E-Revised
• Target Population: Adolescents and adults (designed with reading levels suitable for individuals aged 12 and older)
• Administration Format: Self-report questionnaire; paper-and-pencil or secure computer/online administration
• Item Count: 14 items
• Estimated Completion Time: 3 to 5 minutes
• Response Scale: 5-point Likert-type scale with the following anchors:
  • 1 = I strongly disagree
  • 2 = I disagree
  • 3 = I’m not sure
  • 4 = I tend to agree
  • 5 = I strongly agree
• Subscale Breakdown:
  • Intrinsic Religiosity (I): 8 items — Items 1, 3, 4, 5, 7, 10, 12, 14
  • Extrinsic Religiosity (E): 6 items — Items 2, 6, 8, 9, 11, 13
    • Extrinsic-Social (Es): Items 2, 11, 13
    • Extrinsic-Personal (Ep): Items 6, 8, 9
• Scoring Procedures:
  • Reverse Scoring: Items marked with a plus sign (+) or identified as reverse-keyed must be inverted prior to scoring (i.e., $1 = 5$, $2 = 4$, $3 = 3$, $4 = 2$, $5 = 1$). Note that in the original source documentation, items 2, 10, and 14 are flagged for reversal in specific scoring typologies (items 10 and 14 in the Intrinsic scale, and item 3 in alternative formulations; in standard Gorsuch & McPherson coding, reverse-scored items are inverted to maintain positive conceptual alignment).
  • Scale Totals: Compute mean or sum scores for Intrinsic and Extrinsic scales. High scores reflect higher levels of the respective religious orientation.
  • Typological Categorization: Researchers may split distributions along sample medians to assign respondents into Allport’s four traditional categories: Intrinsic (High I, Low E), Extrinsic (Low I, High E), Indiscriminately Pro-religious (High I, High E), and Indiscriminately Anti-religious (Low I, Low E).

11. Permissions & Fee and Test Year

The Religious Orientation Scale-Revised was published in 1989 by Richard L. Gorsuch and Susan E. McPherson in the peer-reviewed Journal for the Scientific Study of Religion. As an academic assessment tool published in a scholarly journal, the ROS-R is considered to be within the public domain for non-commercial educational, clinical, and scientific research purposes, provided that appropriate bibliographic citation is extended to the original authors.

No licensing fees or formal commercial royalties are required to use the ROS-R in academic research or clinical settings. Organizations or commercial assessment developers seeking to incorporate the instrument into proprietary, fee-based diagnostic platforms or mobile applications should review copyright permissions directly via the JSTOR repository or through the rights-holder of the Journal for the Scientific Study of Religion (Wiley-Blackwell on behalf of the Society for the Scientific Study of Religion).

12. References

• Allport, G. W. (1950). The individual and his religion: A psychological interpretation. Macmillan.
• Allport, G. W. (1954). The nature of prejudice. Addison-Wesley.
• Allport, G. W., & Ross, J. M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5(4), 432–443. https://doi.org/10.1037/h0021212
• Genia, V. (1993). A psychometric evaluation of the Allport-Ross I/E Scale and the Venice Age Universal Scale. The International Journal for the Psychology of Religion, 3(2), 99–106. https://doi.org/10.1207/s15327582ijpr0302_3
• Gorsuch, R. L. (1994). Toward motivational theories of intrinsic religious commitment. Journal for the Scientific Study of Religion, 33(4), 315–325. https://doi.org/10.2307/1386491
• Gorsuch, R. L., & McPherson, S. E. (1989). Intrinsic/extrinsic measurement: I/E-Revised and single-item scales. Journal for the Scientific Study of Religion, 28(3), 348–354. https://doi.org/10.2307/1386745
• Gorsuch, R. L., & Venable, G. D. (1983). Development of an “Age Universal” I-E Scale. Journal for the Scientific Study of Religion, 22(2), 181–187. https://doi.org/10.2307/1385677
• Jones, A. E. (2014). Moderating effect of religious orientation on the relationship between sexual self-discrepancies and guilt and anxiety (Master’s thesis). Indiana State University. Scholars’ BiS. http://tbr2.indstate.edu/bitstream/10484/5576/1/Ann%20E.%20Jones%20Aug%202014.pdf
• Kirkpatrick, L. A. (1989). A psychometric analysis of the Allport-Ross and Feagin measures of intrinsic-extrinsic religious orientation. Research in the Social Scientific Study of Religion, 1, 1–31.
• Maltby, J. (1999). The internal structure of a derived, revised, and amended measure of the Religious Orientation Scale: The ‘Age-Universal’ I-E Scale-12. Social Behavior and Personality: An International Journal, 27(4), 407–412. https://doi.org/10.2224/sbp.1999.27.4.407
• Maltby, J., & Day, L. (1998). Amending a measure of the Age-Universal I-E Scale. The Journal of Psychology, 132(2), 216–222. https://doi.org/10.1080/00223989809599163
• Trimble, D. E. (1997). The Religious Orientation Scale: Review and meta-analysis of social desirability effects. Educational and Psychological Measurement, 57(6), 970–986. https://doi.org/10.1177/0013164497057006007

13. Items of the Scale (Questionnaire)