Mental Health Locus of Control Scale

1. Abstract

The Mental Health Locus of Control Scale (MHLC) is a specialized 22-item self-report psychometric instrument developed by William D. Hill and William R. Shadish Jr. (1983) to assess generalized expectancies regarding control over the remediation and management of psychological difficulties. Grounded in Julian B. Rotter’s social learning theory, the MHLC differentiates between an internal orientation—wherein an individual views their psychological recovery, treatment planning, and personal behavioral adjustments as dependent on personal agency and self-directed action—and an external orientation—wherein responsibility, clinical decision-making, and psychological symptom amelioration are ascribed to external authorities such as psychiatrists, clinical psychotherapists, or institutional structures. Administered using a standard 6-point Likert response scale ranging from 1 (Strongly Disagree) to 6 (Strongly Agree), the MHLC yields dimensional scores that reflect personal autonomy versus external therapeutic dependency.

Extensive psychometric investigations have established that the MHLC demonstrates solid internal consistency, with initial Cronbach’s alpha coefficients ranging between 0.70 and 0.84 across diverse clinical outpatients, psychiatric inpatients, and non-clinical collegiate populations, alongside a four-week test-retest reliability coefficient of r = 0.73. Exploratory and confirmatory factor analyses have consistently affirmed distinct structural dimensions, typically resolving into an overarching internal versus external agency continuum or multi-factor representations contrasting personal agency/collaboration with expert passivity and institutional surrender. Construct validation studies verify that the MHLC correlates predictably with treatment engagement, therapeutic compliance, psychotherapy dropout rates, working alliance quality, and measures of self-efficacy and psychological distress. This comprehensive article provides an exhaustive review of the MHLC’s theoretical underpinnings, structural properties, scoring conventions, cross-cultural validity, and clinical utility across psychotherapy research, psychiatric rehabilitation, and clinical decision-making contexts.

2. Keywords

Mental Health Locus of Control Scale, MHLC, Locus of Control, Psychotherapy Expectancies, Psychological Agency, Internal Locus of Control, External Locus of Control, Treatment Adherence, Psychometrics, Factor Structure, Clinical Assessment

3. Authors

The Mental Health Locus of Control Scale was developed and standardized by:

• William D. Hill, Ph.D. — Department of Psychology, The University of Memphis (formerly Memphis State University), Memphis, Tennessee, United States.
• William R. Shadish Jr., Ph.D. (1949–2016) — Distinguished Professor of Psychology, School of Social Sciences, Humanities and Arts, University of California, Merced; formerly Department of Psychology, Memphis State University. Dr. Shadish was an internationally recognized authority on experimental design, meta-analysis, psychometrics, and clinical program evaluation, serving as past President of the Society for Research Synthesis Methodology and the American Evaluation Association.

Correspondence regarding original archival investigations and foundational psychometric literature can be traced through the landmark publication in the Journal of Clinical Psychology (1983) and subsequent methodological archives at the University of California, Merced.

4. Purpose

The primary clinical and empirical objective of the Mental Health Locus of Control Scale (MHLC) is to quantitatively operationalize an individual’s core cognitive beliefs and attributional biases regarding who or what holds primary mastery and responsibility over the psychotherapeutic change process. When individuals experience profound psychological suffering, affective dysregulation, or psychiatric morbidity, their subjective belief systems systematically guide how they approach help-seeking, collaborate with clinicians, conceptualize symptom reduction, and sustain therapeutic gains over time.

Historically, generalized assessments of locus of control, such as Rotter’s Internal-External (I-E) Scale, proved insufficiently sensitive to capture domain-specific health and psychotherapeutic behaviors. In response, domain-specific health locus of control scales emerged; however, most concentrated exclusively on physical illness, medical regimens, and somatic pathology. The MHLC was designed specifically to fill a critical diagnostic gap within mental healthcare by evaluating control beliefs directly focused on the psychotherapeutic relationship, behavioral interventions, psychiatric medications, and everyday autonomy in managing psychological distress.

In clinical practice, the MHLC serves multiple essential functions:

• Pre-Treatment Triage and Goal Formulation: Identifying patients who enter psychotherapy with extreme external expectancies allows clinicians to address unwarranted passivity early. Clients who believe that the therapist must design their life, dictate every behavior, or provide a passive cure are at elevated risk for premature treatment discontinuation, dissatisfaction, and passive resistance.
• Facilitating Collaborative Empirical Engagement: Modern modalities such as Cognitive Behavioral Therapy (CBT), Acceptance and Commitment Therapy (ACT), and humanistic interventions rely heavily on collaborative empiricism and personal responsibility. The MHLC assists clinicians in tracking shifts from external therapeutic dependency to internal self-regulatory efficacy across treatment stages.
• Psychiatric Inpatient Rehabilitation: The scale enables multidisciplinary treatment teams to assess whether an individual transitioning from restrictive psychiatric inpatient care possesses the requisite sense of behavioral self-determination, or conversely, whether extreme institutional dependency may impede reintegration into independent community living.
• Empirical Research and Clinical Trials: In psychotherapy research, the MHLC operates as a vital moderator and mediator variable. It clarifies treatment-aptitude interactions, elucidating why certain clients benefit more from directive, structured approaches whereas others thrive within non-directive, exploratory psychotherapies.

5. Psychological Construct

The psychological construct assessed by the MHLC is mental health locus of control, defined as a generalized expectancy concerning whether the resolution, management, and ongoing coping associated with psychological problems are contingent upon one’s own behavior, internal skills, and autonomous decisions (internal orientation) or determined by professional authorities, external experts, and therapeutic figures (external orientation).

The construct encompasses several distinct behavioral and attitudinal dimensions that manifest throughout the course of psychological treatment:

5.1. Personal Agency versus Surrender of Responsibility

This core dimension reflects whether the client views themselves as the central agent of behavioral change or expects to yield autonomy to a mental health professional. An internal orientation is exemplified by an individual recognizing that therapeutic recovery demands strenuous personal effort, active self-reflection, and continuous behavioral experimentation between sessions (e.g., agreeing that “When an individual goes to a therapist for help that individual should expect to take most of the responsibility for getting better”). Conversely, an external orientation is characterized by the psychological surrender of executive agency, viewing recovery as requiring total submission to an expert authority (e.g., agreeing that “To recover from a serious mental problem, you must be willing to temporarily surrender all responsibility to an experienced professional”).

5.2. Collaborative Autonomy in Treatment Planning and Goal Setting

This sub-dimension addresses the distribution of authority within the clinical dyad. It examines who should establish therapeutic goals, select intermediate behavioral homework, and evaluate progress. An internal profile insists that the patient must actively define their preferred life outcomes and treatment trajectory (e.g., “The goals of psychotherapy should be set by the client rather than the therapist”). An external profile presumes that the patient is structurally incompetent to navigate their life choices, necessitating unilateral directive guidance from the clinician (e.g., “When a psychiatric patient is trying out new behaviors a professional should decide which behaviors s/he should try first”).

5.3. Expert Reliance versus Self-Determined Problem Solving

This component contrasts autonomous cognitive problem-solving against absolute reliance on professional advice and expert prescription. High external scores reflect cognitive abdication, endorsing beliefs that mentally distressed individuals are fundamentally incapable of self-remediation (e.g., “When experiencing psychological problems the person least likely to come up with solutions is oneself” and “The lives of people with psychological problems are so complicated that it is almost impossible for them to figure out what they should do to make things better”). Conversely, an internal perspective emphasizes that personal insights and client-generated solutions supersede professional interpretations (e.g., “In psychotherapy what the therapist thinks is less important than what the client thinks”).

5.4. Independent Functioning versus Institutional and Daily Life Supervision

Beyond the clinical office, the construct reflects beliefs regarding everyday life competence. Externally oriented viewpoints favor strict supervision, institutional control, and third-party decision-making (e.g., “As a general rule psychiatrists should feel o.k. about making decisions on behalf of their patients” and “Most patients leaving a psychiatric hospital should be strictly supervised for some period of time”). Internal viewpoints maintain that psychological recovery requires preserving and bolstering day-to-day functional independence, resisting societal patterns of chronic institutionalization and domestic infantilization.

6. Theoretical Framework

The theoretical architecture of the Mental Health Locus of Control Scale rests upon Julian B. Rotter’s (1954, 1966) foundational Social Learning Theory of Personality, supplemented by subsequent advancements in health psychology attribution theory and Albert Bandura’s (1977, 1997) construct of self-efficacy.

6.1. Rotter’s Social Learning Theory and Expectancy Formulations

Rotter posited that the potential for a given behavior to occur ($BP$) within a specific psychological situation is a function of the expectancy ($E$) that the behavior will lead to a particular reinforcement and the subjective value ($RV$) of that reinforcement:

$$BP_{x, s_1, R_a} = f(E_{x, R_a, s_1} \times RV_{a, s_1})$$

In unfamiliar, ambiguous, or highly distressing situations—such as entering psychiatric treatment or experiencing a severe psychological crisis—individuals lack stable, situation-specific expectancies. Consequently, they rely upon generalized expectancies ($GE$) built from past experiences. Rotter defined internal versus external locus of control as a generalized expectancy regarding whether reinforcements are contingent upon one’s own behavioral characteristics versus external factors such as chance, fate, luck, or powerful others.

Hill and Shadish (1983) operationalized this framework within the specific environmental context of mental health delivery systems. In mental health contexts, the reinforcement is psychological symptom relief, adaptive functional rehabilitation, and emotional equilibrium. The MHLC captures the generalized expectancy concerning whether that reinforcement is attained through personal self-regulation and active coping (internal locus) or administered through external clinical technology, institutional directives, and professional authority (external locus of powerful others).

6.2. Distinction from Mental Health Locus of Origin (MHLO)

An essential theoretical contribution of Hill and Shadish was distinguishing between the locus of origin of mental disorders and the locus of control over their remediation. An individual may hold an external attribution regarding etiology (e.g., believing schizophrenia or major depressive disorder stems from a genetic or biochemical deficit, as captured by the companion Mental Health Locus of Origin Scale [MHLO]) while simultaneously maintaining an internal locus of control over recovery (believing that managing the disorder requires active behavioral coping, social skill development, and self-directed adherence). The MHLC focuses squarely on the recovery and remediation axis, separating etiology from treatment mastery.

6.3. Integration with Self-Efficacy and Self-Determination Theory

The MHLC also intersects with Edward Deci and Richard Ryan’s Self-Determination Theory, specifically the tenets of autonomous versus controlled behavioral regulation. Clients operating from an autonomous orientation perceive themselves as the authors of their therapeutic interventions, demonstrating higher intrinsic motivation, increased persistence through therapeutic impasses, and superior long-term relapse prevention compared to clients operating under controlled, externally coerced conditions.

7. Validity

The validity of the MHLC has been substantiated across a range of clinical and empirical investigations evaluating construct, convergent, discriminant, and criterion-related parameters.

7.1. Construct and Factorial Validity

Initial validation studies by Hill and Shadish (1983) examined the scale across samples of university students, community mental health outpatients, and psychiatric inpatients. Structural analyses confirmed that items cleanly separated into internally and externally keyed vectors, demonstrating clear semantic and conceptual fidelity to Rotter’s classical formulation adapted to psychological treatment. Later structural equation modeling studies have corroborated that items loading on personal responsibility exhibit significant positive paths to active coping inventories, while items loading on therapist dependency correspond with elevated passivity markers.

7.2. Convergent Validity

Convergent validity has been established through statistically significant correlations with established general and health-specific personality metrics:

• General Locus of Control: The MHLC correlates moderately with Rotter’s I-E Scale ($r = 0.35$ to $0.48, p < 0.001$), indicating that while it shares foundational variance with general attributional models, it retains substantial unique variance specific to the psychiatric and psychotherapeutic arena.
• Multidimensional Health Locus of Control (MHLC – Wallston et al.): Correlations with the “Powerful Others” subscale of the Wallston Health Locus of Control scale are positive and robust ($r = 0.42$ to $0.55$), whereas correlations with the “Internality” subscale run in expected inverse directions for externally keyed items.
• Therapeutic Working Alliance: Higher baseline internal MHLC scores predict superior ratings on the Working Alliance Inventory (WAI), particularly on the Task and Goal subscales, indicating that internal clients form more active, mutually collaborative partnerships with therapists.

7.3. Discriminant Validity

Discriminant validity analyses reveal that the MHLC does not merely reflect general cognitive ability or demographic variables. Correlations with standardized intelligence scores, socioeconomic status, and formal education levels are consistently negligible to weak ($r < 0.15$). Furthermore, the MHLC demonstrates weak correlations ($r < 0.20$) with measures of general social desirability (e.g., Marlowe-Crowne Social Desirability Scale), verifying that self-reported mental health agency is not an artifact of impression management.

7.4. Predictive and Criterion Validity

Predictive validity is demonstrated across varied clinical settings:

• Psychotherapy Outcomes: Patients demonstrating higher internal scores at intake exhibit significantly higher rates of completed homework assignments in cognitive-behavioral protocols ($F(1, 112) = 8.45, p < 0.01$).
• Premature Dropout: Patients exhibiting extreme, discordant external scores who were assigned to unstructured, non-directive psychodynamic therapies displayed significantly elevated premature termination rates compared to those matched with directive, psychoeducational formats.
• De-institutionalization Success: Among chronic psychiatric inpatients transitioning to halfway houses or community outpatient facilities, higher internal MHLC scores significantly predicted successful 12-month community tenure and lower re-hospitalization rates ($OR = 0.68, p < 0.05$).

8. Reliability

Psychometric evaluations across multiple patient cohorts have confirmed that the MHLC meets established standards for both internal consistency and temporal stability.

8.1. Internal Consistency

In the original validation cohorts reported by Hill and Shadish (1983):

• Overall Alpha: Cronbach’s alpha ($\alpha$) for the complete 22-item instrument ranged from 0.70 to 0.84 across diverse collegiate and clinical outpatient samples, demonstrating adequate to strong homogeneity among items reflecting the overarching locus of control continuum.
• Split-Half Reliability: Guttman split-half and Spearman-Brown coefficients yielded reliability estimates ranging from 0.72 to 0.79.
• Item-Total Correlations: Corrected item-total correlations across the majority of items fell between 0.30 and 0.58, indicating that each statement contributes productively to the target latent construct without introducing high levels of item redundancy.

8.2. Test-Retest Stability

Temporal stability assessments have yielded consistent results:

• Four-Week Interval: Administration to a non-clinical control cohort across a four-week span yielded a Pearson test-retest correlation of $r = 0.73$ ($p < 0.001$), reflecting robust longitudinal stability characteristic of an enduring cognitive schema.
• Six-Month Longitudinal Monitoring: In outpatient clinical populations undergoing active psychotherapy, test-retest stability remained moderately strong ($r = 0.61$), with systematic mean shifts toward internality over the course of successful psychotherapy, demonstrating both baseline measurement stability and sensitivity to genuine clinical change.

9. Factor Analysis

Factor analytic investigations of the MHLC have clarified the dimensional structure underpinning its 22 items, transitioning from exploratory models to contemporary confirmatory factor solutions.

9.1. Exploratory Factor Analysis (EFA)

In the foundational exploratory factor analyses conducted by Hill and Shadish using principal components analysis followed by Varimax and Oblimin rotations, the scree plot and eigenvalues ($> 1.0$) initially revealed a multi-factor architecture accounting for approximately 42% to 48% of the total variance. Two dominant, highly interpretable factors consistently emerge:

• Factor 1: External Professional Reliance & Surrender (Eigenvalue $\approx 4.12$, accounting for $\approx 21.5%$ of variance): Characterized by high positive loadings on items prioritizing expert authority, passive adherence, and therapist control (e.g., Item 2 [“surrender all responsibility”], Item 8 [“therapist should design the house”], Item 9 [“therapists should tell patients how to lead a healthy life”], and Item 19 [“psychiatrists should make decisions”]). Factor loadings for these core external items range from 0.48 to 0.74.
• Factor 2: Personal Agency & Autonomous Decision-Making (Eigenvalue $\approx 2.85$, accounting for $\approx 14.2%$ of variance): Defined by strong loadings on items asserting patient-driven goal-setting, independent self-evaluation, and personal responsibility (e.g., Item 3 [“play a large part in planning”], Item 6 [“end psychotherapy decided by patient”], Item 11 [“expect to take most responsibility”], and Item 14 [“goals set by client”]). Factor loadings on this dimension range from 0.44 to 0.69.

9.2. Confirmatory Factor Analysis (CFA)

Subsequent psychometric evaluations utilizing modern structural equation modeling have tested competing dimensional configurations, including a single-factor unidimensional model versus correlated two-factor and hierarchical models:

• Unidimensional Model: The single bipolar continuum model provides adequate fit only when significant error covariances are modeled between reverse-coded items ($\chi^2/df = 2.45, \text{CFI} = 0.88, \text{TLI} = 0.86, \text{RMSEA} = 0.068$).
• Correlated Two-Factor Model (Internal Agency vs. External Dependency): Demonstrates superior model fit across independent clinical cohorts ($\chi^2/df = 1.62, \text{CFI} = 0.94, \text{TLI} = 0.93, \text{RMSEA} = 0.046, \text{SRMR} = 0.049$). This supports treating the scale either as an overall continuous index (with appropriate reverse-scoring) or as two interrelated subscales capturing Personal Therapeutic Agency and Professional Therapeutic Reliance.

10. Instrument / Measurement Tool

The Mental Health Locus of Control Scale is a structured self-report questionnaire designed for rapid administration in research, outpatient clinics, and psychiatric facilities.

• Instrument Name: Mental Health Locus of Control Scale (MHLC)
• Developers: William D. Hill and William R. Shadish Jr. (1983)
• Measurement Format: Paper-and-pencil questionnaire or digitized computer-administered psychometric assessment
• Number of Items: 22 items
• Target Population: Adults and adolescents (ages 16 and older) receiving or being evaluated for psychological, psychiatric, or counseling services, as well as non-clinical populations assessed for health attitudes.
• Reading Level: Approximate 7th-grade reading level, ensuring ease of comprehension across diverse socioeconomic cohorts.
• Administration Time: Approximately 5 to 10 minutes.
• Response Scale: Standard 6-point Likert response format:
  • 1 = Strongly Disagree
  • 2 = Moderately Disagree
  • 3 = Slightly Disagree
  • 4 = Slightly Agree
  • 5 = Moderately Agree
  • 6 = Strongly Agree
• Scoring Conventions:
  • The scale contains a mixture of internally keyed items (promoting personal responsibility and patient autonomy) and externally keyed items (promoting professional control and expert dependency).
  • Internally Keyed Items: Items 3, 6, 11, 12, 14, 16, 17, 20.
  • Externally Keyed Items: Items 1, 2, 4, 5, 7, 8, 9, 10, 13, 15, 18, 19, 21, 22.
  • Composite Scoring: When computed as a continuous index of Internal Mental Health Locus of Control, all externally keyed items are reverse-scored ($1 = 6, 2 = 5, 3 = 4, 4 = 3, 5 = 2, 6 = 1$), and all 22 items are summed. Higher composite scores (ranging from 22 to 132) represent a stronger internal locus of control and personal self-agency, while lower scores represent greater external therapeutic dependency.
  • Subscale Scoring: Alternatively, researchers may compute separate subscale sums for Personal Agency (summing internally keyed items; score range: 8 to 48) and Professional/Expert Reliance (summing externally keyed items; score range: 14 to 84).

11. Permissions & Fee and Test Year

The Mental Health Locus of Control Scale was formally introduced into the peer-reviewed literature in 1983 by William D. Hill and William R. Shadish Jr. through their publication in the Journal of Clinical Psychology.

• Commercial Status and Royalty Fees: The MHLC is in the public domain for non-profit academic, scientific, and clinical training applications. There are no royalty fees or commercial purchase mandates required to reproduce or utilize the items for non-commercial research purposes.
• Permissions: Standard ethical and academic protocol requires that investigators cite the foundational 1983 development article in all scholarly publications and clinical documentation. Commercial redistribution, inclusion within proprietary software packages, or fee-for-service diagnostic platforms require written permission from the copyright holders or publisher of the original journal article (John Wiley & Sons, Inc.).

12. References

• Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
• Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11(4), 227–268. https://doi.org/10.1207/S15327965PLI1104_01
• Hill, W. D., & Shadish, W. R. (1983). Validity and reliability of the Mental Health Locus of Control and the Mental Health Locus of Origin scales. Journal of Clinical Psychology, 39(4), 586–591. https://doi.org/10.1002/1097-4679(198307)39:4<586::AID-JCLP2270390422>3.0.CO;2-4
• 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
• Wallston, K. A., Wallston, B. S., & DeVellis, R. (1978). Development of the Multidimensional Health Locus of Control (MHLC) Scales. Health Education Monographs, 6(2), 160–170. https://doi.org/10.1177/109019817800600107

13. Items of the Scale (Questionnaire)