Internal-External Locus of Control Scale

1. Abstract

The Internal-External Locus of Control Scale (commonly referred to as the I-E Scale) was developed by American psychologist Julian B. Rotter in 1966 as an empirical operationalization of generalized expectancies regarding reinforcement contingencies within his broader social learning theory. The instrument measures an individual’s generalized belief system regarding whether life outcomes, reinforcements, rewards, and punishments are primarily contingent upon one’s own actions, personal agency, effort, and enduring capacities (internal locus of control) or are predominantly dictated by external forces beyond personal control, such as chance, luck, fate, social structures, or powerful others (external locus of control). The standard I-E Scale comprises 29 forced-choice dyadic items, consisting of 23 scored item pairs and 6 non-scored filler items designed to attenuate transparent demand characteristics and diminish response bias related to political ideology and social desirability.

Each scored item forces respondents to adjudicate between an internal statement and an external statement. The aggregate score represents the sum total of externally keyed choices, yielding an integer score ranging from 0 (maximum internal orientation) to 23 (maximum external orientation). Across diverse populations—including university undergraduates, secondary school students, clinical cohorts, and cross-cultural samples—the I-E Scale exhibits acceptable psychometric properties, with Kuder-Richardson Formula 20 (KR-20) internal consistency coefficients typically clustering around .70 and one-month test-retest reliability indices evaluated at .72. Although originally conceptualized and statistically validated as an essentially unidimensional measure of generalized expectancy, subsequent structural equation modeling and exploratory factor analytic investigations have illuminated meaningful multidimensional facets, including personal efficacy, systemic control, and interpersonal agency. This article provides a comprehensive academic review of the scale’s theoretical origins, construct taxonomy, psychometric validation history, factor analytic debates, scoring mechanics, and enduring application across contemporary behavioral, educational, and clinical research.

2. Keywords

Internal-External Locus of Control, Julian B. Rotter, I-E Scale, generalized expectancies, social learning theory, perceived control, attribution theory, psychological agency, psychometrics, forced-choice format

3. Authors

The Internal-External Locus of Control Scale was authored by Julian B. Rotter (October 22, 1916 – January 6, 2014), one of the seminal architects of twentieth-century clinical psychology and cognitive-behavioral perspectives. At the time of the scale’s foundational validation and publication in Psychological Monographs: General and Applied (1966), Rotter served as Professor of Psychology and Director of the Clinical Psychology Training Program at The Ohio State University, prior to his tenure as Professor Emeritus at the University of Connecticut. Rotter collaborated closely with doctoral students and colleagues at Ohio State, including E. Jerry Phares and William H. James, whose preliminary investigations into instrumental conditioning, expectancy shifts under skill versus chance orientations, and generalized reinforcement expectations directly formed the empirical springboard for the final 29-item instrument.

4. Purpose

The primary clinical, diagnostic, and empirical purpose of the Internal-External Locus of Control Scale is to measure individual differences in generalized expectancy regarding the causal determinants of reinforcement. In Rotter’s theoretical framework, human behavior across novel, ambiguous, or complex social situations is driven not merely by the objective reinforcement value of potential outcomes, but fundamentally by the subjective expectancy that a specific behavior will successfully elicit that desired outcome. The I-E Scale was specifically constructed to quantify where an individual resides along a continuous spectrum ranging from internal attribution (the belief that life events are causally tethered to personal decisions, industriousness, and ability) to external attribution (the fatalistic or structural belief that outcomes occur irrespective of personal actions, governed instead by capricious circumstance, destiny, or dominant societal gatekeepers).

In academic and psychological research, the I-E Scale serves as a foundational predictor of goal-directed persistence, academic achievement, self-regulation, occupational burnout, and proactive coping styles. Decades of empirical literature demonstrate that individuals with a predominantly internal locus of control exhibit higher degrees of information seeking, superior problem-focused coping mechanisms, higher cognitive engagement in educational settings, and greater resistance to subtle forms of social influence and coercive persuasion. Conversely, an external orientation correlates reliably with psychological phenomena such as learned helplessness, passive avoidance, depressive symptom burden, anxiety disorders, and elevated occupational strain, particularly in high-demand, low-latitude workplace environments.

Clinically, the instrument provides practitioners with an invaluable framework for cognitive assessment and intervention design. Patients who attribute psychological distress entirely to unalterable external factors often present severe resistance to cognitive-behavioral interventions that mandate personal behavioral activation and cognitive restructuring. By identifying a client’s baseline locus of control, clinicians can tailor interventions—such as attributional retraining, graded mastery tasks, and personal agency experiments—to gradually shift the individual from paralyzing external fatalism toward adaptive internal accountability, thereby fostering self-efficacy and therapeutic alliance.

5. Psychological Construct

The psychological construct assessed by the I-E Scale is operationalized as generalized expectancy for internal versus external control of reinforcement. Locus of control is neither a typological binary nor a rigid intrapsychic personality trait in the classical psychodynamic sense. Rather, it represents an enduring cognitive-perceptual orientation that reflects an individual’s accumulated history of reinforcement learning across heterogeneous situations. Within Rotter’s taxonomy, the construct is conceptualized along a single continuous dimension, though its theoretical manifestations encompass several distinct life domains:

• Personal Agency and Effort Contingency (Internal Dimension): This dimension encapsulates the enduring conviction that personal mastery, diligence, forethought, and skill directly dictate success or failure. An individual demonstrating high internal expectancies posits that academic grades, vocational advancement, interpersonal affinity, and life satisfaction stem directly from personal competence and perseverance. When confronted with negative outcomes, such individuals formulate internal, controllable attributions (e.g., “I failed the examination because my study strategy was inadequate”) rather than projecting blame outward.
• Chance, Luck, and Fatalism (External Dimension): This sub-facet of externality reflects the generalized belief that life is inherently unpredictable, chaotic, and ruled by random circumstance or metaphysical destiny. Individuals scoring high in fatalistic externality believe that future planning is largely futile because unpredictable twists of fate inevitably nullify systematic effort. This orientation often engenders behavioral passivity, as the subjective probability of effort producing a stable reinforcement approaches zero.
• Systemic and Powerful Others Control (External Dimension): Distinct from pure randomness, this external facet reflects the belief that social reality is dominated by opaque, immovable hierarchies, political elites, or unyielding authority figures. In this domain, the individual perceives that the “little guy” has no meaningful voice or leverage over societal, governmental, or institutional outcomes, creating an attitude of political alienation, civic disengagement, and perceived powerlessness.

Importantly, Rotter emphasized that generalized expectancies operate most prominently when an individual enters novel, ambiguous, or unstructured settings where past specific experience offers little behavioral guidance. In contrast, in highly familiar, scripted scenarios (such as driving a car or executing a routine job task), specific expectancies (expectancies grounded in direct, localized history with that exact task) supersede generalized locus of control beliefs. Consequently, the I-E Scale measures the macro-level cognitive schema that an individual imports into novel environments, determining their preliminary orientation toward action versus resignation.

6. Theoretical Framework

The I-E Scale is directly derived from Julian B. Rotter’s Social Learning Theory of Personality (Rotter, 1954), which integrated the behavioral rigor of stimulus-response learning paradigms with the cognitive mediating variables championed by cognitive psychology and Gestalt theories. In contrast to strict radical behaviorism, which dismissed internal cognitive states as epiphenomena, Rotter argued that human learning occurs primarily in complex social contexts and is mediated by cognitive interpretations of meaning and probability.

The foundational mathematical formulation governing social learning theory is expressed as:

BP_x,s1,Ra = f(E_x,Ra,s1 & RV_a,s1)

This formula states that the Behavior Potential (BP) for behavior x to occur in situation 1 in relation to reinforcement a is a function of the Expectancy (E) that behavior x will lead to reinforcement a in situation 1, and the subjective Reinforcement Value (RV) of that reinforcement in that specific situation. As an individual encounters numerous disparate situations across developmental time, specific expectancies generalize into higher-order cognitive constructs:

E* = f(E’ + GE / N)

Here, total generalized expectancy (E*) is composed of the specific expectancy derived from identical prior encounters (E’) plus a generalized expectancy (GE) divided by the number of novel situational inputs (N). Rotter’s I-E Scale operationalizes precisely this generalized expectancy component (GE). In Rotter’s experimental paradigms involving skill versus chance games (e.g., card guessing versus motor dexterity tasks), participants exhibited radically different expectancy curve trajectories. Under skill conditions, successful reinforcement led to pronounced increases in expectancy for future trials, and failure led to systematic decreases. Under chance conditions, expectancy shifts were erratic, non-linear, and prone to the gambler’s fallacy. Rotter realized that individuals habitually view their everyday lives through one of these two operational prisms: as a skill task (internal) or as a chance task (external).

7. Validity

The construct, convergent, discriminant, and predictive validity of the I-E Scale has been extensively documented in psychometric literature:

• Discriminant Validity: Rotter demonstrated robust discriminant validity during scale construction by deliberately pruning items that correlated substantially with social desirability. Correlations between the finalized 23-item I-E Scale and the Marlowe-Crowne Social Desirability Scale range between -.07 and -.35 (typically averaging approximately -.20), indicating that the scale does not merely capture an individual’s need for social approval. Furthermore, the scale demonstrates near-zero correlations with general intelligence (IQ) tests (r ranging from .03 to .11) and minimal correlation with political affiliation or political liberalness, ensuring that the construct of perceived control is independent of cognitive capability and overt party loyalty.
• Convergent Validity: The I-E Scale correlates significantly with conceptually related personality and cognitive measures. External locus of control correlates positively with the Taylor Manifest Anxiety Scale (r = .30 to .45), Beck Depression Inventory scores, and neuroticism indices on the Eysenck Personality Questionnaire. Conversely, internal scores correlate positively with measures of self-efficacy (e.g., Bandura’s General Self-Efficacy Scale), academic self-concept, autonomous motivation, and conscientiousness.
• Predictive and Behavioral Validity: In behavioral laboratory and field paradigms, the I-E Scale displays remarkable predictive power. In early validation experiments cited by Rotter (1966), internal individuals engaged in significantly more active information gathering regarding health, occupational, and political issues. For instance, hospitalized tuberculosis patients scoring high on internal locus of control demonstrated significantly greater knowledge about their medical condition and rehabilitation requirements than external patients of comparable intelligence. Similarly, during the Civil Rights movement, African American college students who actively engaged in civil rights demonstrations and political activism exhibited markedly more internal scores on personal agency than non-participating peers.

8. Reliability

The psychometric reliability of the I-E Scale has been examined across a vast array of normative, student, and clinical cohorts since its original 1966 standardization:

• Internal Consistency: In Rotter’s initial validation sample comprising 200 male and 200 female Ohio State University undergraduates, the Kuder-Richardson Formula 20 (KR-20) coefficient was calculated at .70 for both sexes. Subsequent investigations across diverse undergraduate, adolescent, and adult samples have reported internal consistency coefficients ranging between .65 and .79. Given that the scale uses a forced-choice dichotomy across a wide variety of behavioral arenas (academic, personal, political, interpersonal), a reliability coefficient of .70 is considered strong for a broad generalized measure.
• Test-Retest Stability: Temporal stability coefficients reflect adequate reliability over moderate durations. Rotter (1966) reported a one-month test-retest reliability coefficient of r = .72 in a sample of psychology undergraduates (N = 60). Over a two-month interval, test-retest reliability decreased to r = .55, which Rotter attributed to intervening reinforcement experiences and developmental transitions common among collegiate cohorts entering new institutional environments. Subsequent studies with stable adult populations over 4- to 6-week periods report test-retest coefficients consistently spanning .60 to .83.
• Biserial Item Correlations: Rotter reported item-total biserial correlations for each of the 23 scored items, with values generally ranging between .15 and .48. Items exhibiting marginal or negative item-total correlations during the scale’s initial 100-item and 60-item iterations were systematically eliminated to optimize psychometric homogeneity.

9. Factor Analysis

Although Julian Rotter originally maintained that the 23 scored items formed an essentially unidimensional construct—demonstrated by early centroid factor analyses showing that all items loaded to a moderate degree on a single general factor accounting for the majority of shared variance—subsequent psychometricians utilizing sophisticated exploratory and confirmatory factor analysis (EFA and CFA) have repeatedly identified underlying multidimensionality.

The most influential structural critique was conducted by Gurin, Gurin, Lao, and Beattie (1969), who demonstrated that the I-E Scale bifurcates cleanly into at least two distinct sub-constructs:

• Personal Control (Internal Agency): Items worded in the first person (“I”, “me”) assessing beliefs regarding whether an individual can influence their personal outcomes (e.g., Item 13: “When I make plans, I am almost certain that I can make them work”). Factor loadings on this personal control factor typically range from .45 to .68.
• Control Ideology (Systemic Expectancy): Items worded in the third person (“people”, “citizens”, “students”) assessing generalized cultural beliefs regarding how society operates at large (e.g., Item 4: “In the long run people get the respect they deserve in this world”). Factor loadings for control ideology cluster between .40 and .65.

In 1971, Mirels performed a principal component factor analysis with varimax rotation on the 23 items using 316 university students. Mirels extracted two clear, orthogonal factors:

• Factor I: Belief in Personal Mastery / Fate vs. Hard Work (items reflecting personal efficacy, academic achievement, and career success, such as items 11, 15, and 28), accounting for the largest percentage of variance.
• Factor II: Belief in Political / Systemic Control (items reflecting the degree to which an individual citizen can influence political processes, world affairs, and governance, such as items 3, 12, 17, and 22).

Modern Confirmatory Factor Analyses (CFA) have confirmed that while a higher-order bi-factor model or a correlated two-factor model provides a significantly superior statistical fit (e.g., Comparative Fit Index [CFI] > .91, Root Mean Square Error of Approximation [RMSEA] < .06) compared to a strict unidimensional model, the total external score remains a robust, empirically valid proxy for overall generalized expectancy across multidisciplinary research settings.

10. Instrument / Measurement Tool

The technical parameters, administration details, and scoring mechanisms of the instrument are summarized below:

• Formal Instrument Name: Internal-External Locus of Control Scale (commonly cited as the I-E Scale).
• Original Author: Julian B. Rotter (1966).
• Instrument Type: Self-report, forced-choice psychometric questionnaire.
• Total Item Count: 29 paired statements.
  • Scored Items: 23 item pairs measuring the internal-external reinforcement expectancy spectrum.
  • Filler Items: 6 non-scored item pairs (Items 1, 8, 14, 19, 24, and 27) included to mask the test’s intentionality and neutralize social desirability and political bias.
• Response Format: Forced-choice dyad (Option a vs. Option b). For each number, the respondent is required to select the single statement that they more strongly believe to be the case.
• Administration Time: Approximately 10 to 15 minutes.
• Target Population: Adolescents (grades 10–12) and adults across educational, occupational, and community populations.
• Scoring Formula: The scale is keyed in the external direction. The total score is computed by summing the number of external alternatives selected by the respondent.
  • Externally Scored Options (1 point each):
    • Item 2: Choice a
    • Item 3: Choice b
    • Item 4: Choice b
    • Item 5: Choice b
    • Item 6: Choice a
    • Item 7: Choice a
    • Item 9: Choice a
    • Item 10: Choice b
    • Item 11: Choice b
    • Item 12: Choice b
    • Item 13: Choice b
    • Item 15: Choice b
    • Item 16: Choice a
    • Item 17: Choice a
    • Item 18: Choice a
    • Item 20: Choice a
    • Item 21: Choice a
    • Item 22: Choice b
    • Item 23: Choice a
    • Item 25: Choice a
    • Item 26: Choice b
    • Item 28: Choice b
    • Item 29: Choice a
  • Unscored Filler Items: Items 1, 8, 14, 19, 24, and 27 are ignored entirely during tabulation.
  • Score Range: 0 to 23 points.
    • Low Scores (0 – 8): Indicate a predominantly Internal Locus of Control (strong belief in personal efficacy, agency, and responsibility for outcomes).
    • Intermediate Scores (9 – 12): Indicate an intermediate, balanced orientation reflecting realistic appraisal of both personal agency and situational constraints.
    • High Scores (13 – 23): Indicate a predominantly External Locus of Control (strong generalized belief that outcomes are determined by luck, chance, fate, or powerful external systems).

11. Permissions & Fee and Test Year

The Internal-External Locus of Control Scale was published in 1966 by Julian B. Rotter in the American Psychological Association’s archival journal Psychological Monographs: General and Applied (Vol. 80, No. 1, Whole No. 609). Because the complete 29-item instrument was published in an academic public domain format within an APA journal prior to contemporary proprietary psychological instrumentation standards, the scale is broadly considered an open-access academic instrument.

No royalty or commercial license fee is required for non-commercial scientific research, academic dissertations, educational instruction, or non-profit clinical inquiry, provided that appropriate scholarly attribution is formally accorded to Julian B. Rotter (1966) and the American Psychological Association. Researchers planning commercial integration, enterprise assessment deployment, or digital software distribution should consult APA copyright policies regarding archival monograph reproduction.

12. References

The following seminal works and contemporary validation studies form the primary scholarly foundation of the I-E Scale:

• Bevis, K. A. (2008). Teacher burnout: Locus of control and its correlation to teacher burnout and job satisfaction (Ed.S. dissertation). Marshall University. https://mds.marshall.edu/etd/637/
• DeAndrade, Y. (2009). The relationship of intentional learning, locus of control, and self-efficacy among online and on-ground college students (D.Mgt. dissertation). Webster University. ProQuest Dissertations & Theses Global.
• Gurin, P., Gurin, G., Lao, R. C., & Beattie, M. (1969). Internal-external control in the motivational dynamics of Negro youth. Journal of Social Issues, 25(3), 29–53. https://doi.org/10.1111/j.1540-4560.1969.tb00605.x
• Mirels, H. L. (1970). Dimensions of internal versus external control. Journal of Consulting and Clinical Psychology, 34(2), 226–228. https://doi.org/10.1037/h0029005
• Nufer, S. (2012). The effects of locus of control and leader-member exchange as predictors of stress and burnout in the workplace (Psy.D. dissertation). Chicago School of Professional Psychology. ProQuest Dissertations & Theses Global.
• Phares, E. J. (1976). Locus of control in personality. General Learning Press.
• 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, Whole No. 609), 1–28. https://doi.org/10.1037/h0092976
• Rotter, J. B. (1975). Some problems and misconceptions related to the construct of internal versus external control of reinforcement. Journal of Consulting and Clinical Psychology, 43(1), 56–67. https://doi.org/10.1037/h0076301
• Traub, E. (2010). The effect of internet usage on social alienation and locus of control (Psy.D. dissertation). University of the Rockies. ProQuest Dissertations & Theses Global.
• Zimmerman, M. T. (2010). Examination of locus of control, health locus of control and their key predictors in urban vs. rural populations (Ed.D. dissertation). University of Memphis. ProQuest Dissertations & Theses Global.

13. Items of the Scale (Questionnaire)