Opinions About Mental Illness Scale

1. Abstract

The Opinions About Mental Illness Scale (OMI), developed by Jacob Cohen and Elmer L. Struening in 1962, stands as a foundational psychometric instrument designed to measure social, institutional, and personal attitudes toward psychiatric conditions and individuals diagnosed with severe psychiatric disorders. Originating in response to the growing recognition that the therapeutic atmosphere of psychiatric hospitals is profoundly shaped by the socio-emotional climate and ideological orientations of clinical and non-clinical staff, the OMI was constructed to replace unidimensional conceptions of stigma with a sophisticated, multidimensional operational model. Comprising 51 items evaluated on a six-point Likert-type scale ranging from “Strongly Agree” to “Strongly Disagree,” the instrument captures five distinct, empirically derived attitudinal dimensions: Authoritarianism (Scale A), Benevolence (Scale B), Mental Hygiene Ideology (Scale C), Social Restrictiveness (Scale D), and Interpersonal Etiology (Scale E). Psychometric evaluations across diverse institutional settings—initially benchmarked within large Veterans Administration (VA) neuropsychiatric hospitals—revealed internal consistency coefficients ranging widely from moderate-to-low levels for specific subscales (e.g., coefficient alpha of .21 for Social Restrictiveness in certain custodial cohorts) to high reliability (alpha of .82 for Authoritarianism). Factor validity coefficients have demonstrated strong construct coherence (.43 to .89 across subscales), establishing the instrument’s utility in differentiating occupational strata, tracking the ideological impact of professional psychiatric training, and informing the sociology of total institutions. Over six decades, the OMI has catalyzed global cross-cultural research into psychiatric stigma, provided the structural blueprint for successor instruments such as the Community Attitudes toward the Mentally Ill (CAMI) inventory, and remains a seminal benchmark in behavioral health psychometrics.

2. Keywords

Opinions About Mental Illness Scale, OMI, Jacob Cohen, Elmer L. Struening, psychiatric stigma, mental health personnel attitudes, Authoritarianism, Benevolence, Mental Hygiene Ideology, Social Restrictiveness, Interpersonal Etiology, psychometrics, custodialism, psychiatric deinstitutionalization

3. Authors

The Opinions About Mental Illness Scale was conceptualized, formulated, and validated through the collaborative empirical work of two prominent mid-twentieth-century American behavioral scientists: Jacob Cohen, Ph.D., and Elmer L. Struening, Ph.D.

Jacob Cohen (1923–1998) was an eminent American statistician and psychologist, widely celebrated for his pioneering contributions to quantitative psychology, statistical power analysis, effect size formulation (e.g., Cohen’s d, Cohen’s kappa), and applied multivariate behavioral research. Throughout his tenure as a Professor of Psychology at New York University (NYU), Cohen dedicated substantial analytical focus to psychometric precision, scale standardization, and psychiatric epidemiology. At the time of the OMI’s development, Cohen served as a statistical and research consultant for the Veterans Administration, addressing critical measurement dilemmas in institutional neuropsychiatric populations.

Elmer L. Struening (1927–2003) was a renowned social scientist, psychiatric epidemiologist, and methodological authority who held leading research appointments at the Columbia University Department of Psychiatry and the New York State Psychiatric Institute. Struening served as the Director of the Epidemiology of Mental Disorders Research Department at the New York State Psychiatric Institute and played an instrumental role in exploring community mental health delivery models, institutional environments, and the psychosocial determinants of chronic mental distress. His career was marked by continuous investigations into the organizational dynamics of public mental health facilities, community attitudes, and psychiatric evaluation methodologies.

Initial development of the instrument was sponsored through the Veterans Administration Psychiatric Evaluation Project, centered across multiple Veterans Administration Neuropsychiatric Hospitals located in the United States.

4. Purpose

The primary purpose of the Opinions About Mental Illness Scale is to evaluate, categorize, and quantify the underlying ideological, cognitive, and affective dimensions of opinions held by individuals toward severe mental disorders, their etiology, management, prognosis, and institutional confinement. Developed primarily for deployment among psychiatric hospital personnel—encompassing psychiatrists, clinical psychologists, psychiatric social workers, registered psychiatric nurses, nursing assistants, and non-clinical ancillary staff—the scale was designed to interrogate how institutional social systems conceptualize psychiatric deviance and to what extent custodial or therapeutic philosophies dominate institutional functioning.

From a clinical and organizational perspective, the therapeutic milieu of any psychiatric inpatient facility is dramatically constrained or facilitated by the prevailing socio-cultural belief systems of its workforce. Prior to the development of the OMI, institutional studies often treated attitudes toward mental illness as a global, unidimensional dichotomy: an individual was either “enlightened” or “unenlightened,” “humanistic” or “custodial.” Cohen and Struening recognized this simplistic framing as scientifically deficient and clinically misleading. Staff members could, for example, express genuine moral benevolence, warm charity, and a protective orientation toward patients while simultaneously harboring strong beliefs in authoritarian control, compulsory institutional confinement, and the social disenfranchisement of the mentally afflicted.

The clinical and operational rationale behind the OMI involves several foundational objectives:

• Diagnostic Milieu Profiling: Assessing the institutional and ward-level culture to detect punitive or counter-therapeutic orientations that negatively affect patient rehabilitation, ward autonomy, and discharge preparation.
• Occupational Hierarchy Stratification: Examining systematic ideological variations among different tiers of hospital staff, revealing how socioeconomic status, academic training, and professional socialization influence clinical assumptions regarding the controllability and dangerousness of psychiatric patients.
• Evaluation of Clinical Education and Training: Measuring longitudinal modifications in attitudes among nursing students, medical residents, and healthcare paraprofessionals following didactic instruction, psychiatric rotations, or anti-stigma interventions.
• Community Integration and Deinstitutionalization Forecasting: Determining the extent to which staff and surrounding community members endorse social integration versus perpetual segregation, an objective that became increasingly critical as the deinstitutionalization movement accelerated in the late 1960s and 1970s.
• Cross-Cultural and Comparative Research: Providing an empirically stable framework to compare cultural belief systems concerning mental health etiology and treatment across diverse global populations, linguistic translations, and sociological settings.

5. Psychological Construct

The core psychological construct measured by the OMI is the multidimensional structure of social attitudes toward mental illness. Rejecting the assumption that stigma is a monolithic construct, Cohen and Struening operationalized opinions along five distinct, empirically derived psychological dimensions. Each dimension represents a coherent cognitive schema that informs how a person perceives, explains, and interacts with individuals experiencing severe psychiatric distress.

Factor A: Authoritarianism

Authoritarianism reflects an attitude toward mental patients based on an underlying philosophy of authoritarian personality theory, conceptualizing patients as fundamentally inferior, dangerous, unpredictable, and distinct from the “normal” population. Individuals scoring high on Factor A believe that psychiatric patients belong to a specialized, degraded social category requiring strict discipline, rigid surveillance, coercion, and immediate custodial containment. This orientation stresses obedience, moral weakness as a precursor to mental breakdown, and the absolute necessity of external behavioral control. Illustrative thematic statements captured within this construct include beliefs that mental patients are essentially like children requiring strict supervision, that locked wards and physical restraints are the most indispensable features of a psychiatric hospital, and that patients lack the capacity for rational self-determination.

Factor B: Benevolence

Benevolence embodies a compassionate, paternalistic, and morally sustained orientation toward mental patients. Anchored in Christian charity, religious humanism, and personal kindliness, this construct frames the psychiatric patient not with suspicion or punitive animus, but with warm sympathy and protective nurturance. While distinctly more therapeutic than Authoritarianism, Benevolence maintains a subtle paternalistic distance: the patient is viewed as an unfortunate, child-like victim of circumstance who must be protected, cared for, and shielded from harsh reality. Rather than advocating for full sociopolitical empowerment and self-advocacy, high scorers on Benevolence champion kind custodialism and gentle shelter. Core expressions of this dimension center on the obligation of society to provide comfortable, sheltered asylums where the mentally broken can be sheltered without experiencing the competitive stresses of modern civilization.

Factor C: Mental Hygiene Ideology

Mental Hygiene Ideology mirrors the conceptual perspective of modern psychiatric medicine, public health education, and the mental hygiene movement popularized in the mid-twentieth century. High scorers view mental illness through an explicit biomedical and psychological health paradigm: mental illness is an “illness like any other.” Consequently, psychiatric conditions are perceived as natural, diagnosable medical disorders amenable to rational scientific investigation, clinical psychotherapy, pharmacological intervention, and professional rehabilitation. This construct rejects moralistic or punitive attributions, asserting that patients can recover full functional competence and that early intervention, professional expertise, and scientific methodology are the decisive elements of clinical management. The construct also tends to emphasize that “normal” people and “mentally ill” people exist on a continuous spectrum rather than occupying entirely distinct biological categories.

Factor D: Social Restrictiveness

Social Restrictiveness encapsulates the defensive desire to constrain the civil liberties, societal participation, and community integration of individuals diagnosed with severe mental illness. High scores on this subscale reflect enduring anxieties regarding societal safety, community contamination, and perceived danger. Respondents endorsing this ideology believe that individuals with a history of mental hospitalization must be barred from exercising full citizenship rights, including voting, marrying, parenting children, holding responsible employment, or living freely within ordinary residential neighborhoods. The underlying affective tone is one of social protectionism: even following symptom stabilization, the formerly hospitalized individual is permanently marked by an indelible stain of risk, necessitating ongoing surveillance and continuous civil limitation to protect the non-afflicted public.

Factor E: Interpersonal Etiology

Interpersonal Etiology represents a specialized, psychodynamically informed cognitive schema attributing the root causes of mental illness to maladaptive interpersonal dynamics, specifically negative child-rearing practices, defective parental socialization, lack of parental affection, and early domestic trauma. Highly popular during the ascendancy of psychoanalytic and family systems paradigms in the 1950s and 1960s, this construct downplays organic, genetic, or somatic etiologies in favor of psychosocial causation. High scorers adhere to the thesis that mental disorders arise primarily because parents failed to love, nurture, or communicate constructively with their offspring during formative developmental stages. It externalizes the historical origin of the disorder onto the family system while attributing clinical pathology to interpersonal conflict rather than inherent constitutional vulnerabilities.

6. Theoretical Framework

The Opinions About Mental Illness Scale was built at the intersection of several influential theoretical paradigms: Authoritarian Personality theory, the sociological analysis of custodial versus humanistic institutional climates, and mid-twentieth-century social-psychological attitude theory.

Adorno’s Theory of the Authoritarian Personality

The primary psychological foundation of the OMI traces directly to the seminal work of Theodor Adorno, Else Frenkel-Brunswik, Daniel Levinson, and Nevitt Sanford in The Authoritarian Personality (1950). Adorno and colleagues established that prejudice, social intolerance, and ethnocentrism are manifestations of a generalized personality syndrome characterized by rigid conventionalism, submissive respect for in-group authority, generalized hostility toward out-groups, and projectivity. Cohen and Struening posited that the mentally ill represent a distinct, highly visible, and deeply devalued out-group within modern society. Consequently, the social prejudice directed toward psychiatric patients mirrors the psychological dynamics of racial and ethnic prejudice. The OMI’s Authoritarianism subscale was explicitly modeled to mirror facets of the California F-Scale, measuring whether an individual applies hierarchical dominance, punitive intolerance, and punitive surveillance to psychiatric patients.

Gilbert and Levinson’s Custodialism-Humanism Continuum

A second crucial theoretical pillar is Doris C. Gilbert and Daniel J. Levinson’s pioneering work (1956, 1957) on the “Custodial Mental Illness Ideology” (CMI) scale. Gilbert and Levinson analyzed how staff ideologies in public mental hospitals operate along a spectrum anchored at one pole by “Custodialism” and at the other by “Humanism”:

• Custodial Orientation: Conceives of the mental hospital primarily as a detention facility dedicated to protecting society from dangerous inmates. The custodial model views patients as irrational, uncommunicative, and fundamentally distinct from staff, requiring rigid discipline, physical security, and bureaucratic depersonalization.
• Humanistic Orientation: Conceives of the hospital as a therapeutic community. Humanism conceptualizes mental illness as an understandable emotional crisis or reaction to socio-environmental stress, views patients as capable of meaningful interpersonal engagement, and advocates for egalitarian ward dynamics, therapeutic freedom, and personal dignity.

While acknowledging Gilbert and Levinson’s monumental theoretical contributions, Cohen and Struening contended that forcing institutional attitudes into a single bipolar continuum (Custodialism vs. Humanism) obscures essential nuances. An institutional worker might repudiate authoritarian coercion while simultaneously advocating for strict community exclusion (Social Restrictiveness), or harbor deeply compassionate, warm feelings toward patients (Benevolence) while fundamentally rejecting the scientific premise of recovery and autonomy (Mental Hygiene Ideology). By utilizing multidimensional factor-analytic methods, the OMI was constructed to transcend the limits of unidimensional custodial-humanistic scales.

Sociology of Total Institutions

The OMI also reflects early conceptual developments in the sociology of psychiatric asylums, notably articulated by Erving Goffman in his landmark 1961 work, Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Goffman documented the pervasive social cleavages within “total institutions” that enforce a sharp caste division between the supervisory staff and the patient group. In this context, institutional opinions function not merely as idiosyncratic individual attitudes, but as institutional defense mechanisms that legitimize staff control, preserve emotional detachment, and rationalize the systematic stripping of patient autonomy. The OMI’s subscales directly capture these institutional sociological phenomena.

7. Validity

The construct, factorial, convergent, and discriminant validity of the Opinions About Mental Illness Scale has been extensively evaluated across clinical settings, general population samples, and diverse cultural milieus.

Factorial and Construct Validity

Factorial validity was established in Cohen and Struening’s seminal 1962 investigation conducted across two large Veterans Administration neuropsychiatric hospitals (Hospital I with 1,194 respondents; Hospital II with 548 respondents). Factor analyses demonstrated remarkable structural stability across these independent institutional settings, confirming that the five identified dimensions were neither localized administrative artifacts nor idiosyncratic sample fluctuations. Factor validity coefficients (correlations between the raw subscales and their underlying standardized factor dimensions) demonstrated strong construct representation across settings, with values spanning from .43 to .89 across the two hospitals.

Convergent and Criterion Validity

Convergent validity has been repeatedly corroborated by evaluating the OMI against established measures of social conservatism, dogmatism, and institutional custody:

• California F-Scale Correlations: The Authoritarianism scale (Scale A) demonstrated substantial positive correlations with Adorno et al.’s California F-Scale (typically r = .55 to .70), affirming that hospital personnel who express authoritarian attitudes toward mental patients also score high on generalized socio-political authoritarianism and ethnocentrism.
• Gilbert and Levinson’s CMI Scale: The OMI’s Authoritarianism and Social Restrictiveness subscales exhibited strong positive correlations with the Custodial Mental Illness Ideology scale (r > .60), whereas the Mental Hygiene Ideology and Benevolence subscales correlated positively with measures of humanistic treatment orientation.

Known-Groups Discriminant Validity

One of the most robust demonstrations of the OMI’s construct validity lies in its capacity to systematically discriminate between distinct professional, educational, and occupational strata within psychiatric facilities (known-groups validation):

• Mental Health Professionals vs. Paraprofessionals: Psychiatrists, clinical psychologists, and psychiatric social workers consistently achieve the lowest scores on Authoritarianism (Factor A) and Social Restrictiveness (Factor D), alongside the highest scores on Mental Hygiene Ideology (Factor C).
• Custodial Nursing Aides and Non-Clinical Staff: Hospital nursing aides, maintenance employees, and administrative clerks routinely score significantly higher on Authoritarianism and Social Restrictiveness, while displaying lower endorsement of Mental Hygiene Ideology.
• Educational Stratification: When educational attainment is partialled out, occupational group differences attenuate but remain statistically significant, demonstrating that both formal higher education and specialized clinical training independently reduce authoritarian attitudes toward psychiatric populations.

Cross-Cultural and Successor Validity Studies

Cross-cultural research applying translated versions of the OMI (in nations such as the United Kingdom, Italy, Canada, Japan, and Israel) has generally verified the structural emergence of the five core factors, though localized variations in factor loadings emerge in response to differing national healthcare systems and socio-cultural values. Furthermore, Taylor and Dear (1981) drew heavily upon the validated factorial structure of the OMI to formulate the Community Attitudes toward the Mentally Ill (CAMI) scale, reaffirming the ecological validity of the OMI’s dimensions in modern non-institutional settings.

8. Reliability

The internal consistency and test-retest reliability of the Opinions About Mental Illness Scale have been widely examined, revealing notable psychometric strengths alongside specific dimension-level vulnerabilities.

Internal Consistency Metrics

In Cohen and Struening’s initial 1962 multi-hospital standardization, internal consistency reliability coefficients (Cronbach’s alpha and split-half reliability) varied substantially across the five subscales and across the two hospital cohorts:

• Scale A (Authoritarianism): Demonstrates the highest and most robust internal consistency within the instrument, yielding alpha coefficients ranging from .77 to .82 across Hospital I and Hospital II. This scale possesses high internal homogeneity and robust item-total correlations.
• Scale B (Benevolence): Displays moderate-to-high internal consistency, with alpha coefficients fluctuating between .65 and .74 across clinical samples.
• Scale C (Mental Hygiene Ideology): Exhibits moderate internal consistency, characterized by alpha coefficients typically between .50 and .61. The lower homogeneity of this subscale reflects its broad composition, spanning scientific etiology, medical management, and optimism regarding therapeutic prognosis.
• Scale D (Social Restrictiveness): Yielded lower and more variable internal consistency in the original psychiatric hospital validation, reporting an alpha coefficient as low as .21 in Hospital I, while improving to .52 to .68 in later community and revised institutional samples. Cohen and Struening attributed the initial low alpha to severe range restriction among hospital staff, who overwhelmingly rejected extreme civil restriction items.
• Scale E (Interpersonal Etiology): Demonstrates moderate internal consistency, with alpha coefficients settling between .58 and .69 in standard adult populations.

Test-Retest Stability

Temporal stability assessments conducted across varying intervals (ranging from two weeks to six months) have confirmed that the OMI measures enduring ideological belief systems rather than transient situational moods. Test-retest reliability coefficients for Scale A and Scale D consistently fall within the .70 to .85 range across stable occupational cohorts. Scales B, C, and E routinely exhibit test-retest coefficients between .60 and .75 over comparable temporal windows, demonstrating sufficient longitudinal stability for repeated-measures evaluative designs and educational intervention trials.

9. Factor Analysis

The factorial architecture of the Opinions About Mental Illness Scale represents one of the earliest large-scale applications of multivariate exploratory factor analysis within psychiatric sociology and clinical psychometrics.

Original Extraction and Derivation

Cohen and Struening developed an initial experimental battery of 70 items, culled from extensive transcripts of hospital case conferences, clinical team meetings, literature on social prejudice, and modified statements from previous custodial ideology scales. This pool was administered to 1,194 personnel of a massive Veterans Administration neuropsychiatric facility (Hospital I). A principal axis factor extraction was executed on the inter-item correlation matrix, followed by both oblique and orthogonal (Varimax) rotations to interpret the latent dimensional landscape.

Through this analytical procedure, 51 items demonstrated salient, distinct factor loadings on five primary factors, while items exhibiting excessive cross-loadings, factorial ambiguity, or negligible variance were excised. The five extracted factors accounted for the substantial majority of the common variance across the battery.

Cross-Sample Replication

To establish the empirical stability of the factor solution, Cohen and Struening performed an independent replication by administering the identical 51 items to 548 employees of a geographically separate VA hospital (Hospital II). The factor extraction yielded an almost identical structural configuration:

• Factor Alignment: The congruence coefficients between the factor matrices of Hospital I and Hospital II approached unity for the primary factors, confirming that Authoritarianism, Benevolence, Mental Hygiene Ideology, Social Restrictiveness, and Interpersonal Etiology operated as invariant cognitive dimensions across distinct administrative populations.
• Item-Factor Loadings: Salient item loadings on targeted factors routinely exceeded .40, with central defining items on the Authoritarianism and Benevolence dimensions attaining loadings above .65. Cross-loadings were modest, although a minor theoretical overlap emerged between Authoritarianism and Social Restrictiveness, which frequently correlate moderately with one another (r ≈ .45 to .55).

Methodological Evolution and Confirmatory Factor Analysis (CFA)

In subsequent decades, independent psychometricians re-evaluated the OMI’s factor model using contemporary Confirmatory Factor Analysis (CFA) techniques. While these contemporary re-analyses have corroborated the overarching validity of the five-dimensional framework, they have frequently noted that several items within Scale C (Mental Hygiene Ideology) and Scale D (Social Restrictiveness) exhibit multidimensionality and lower standardized path coefficients (loadings < .35). Modern structural equation modeling has supported the utility of modified, shortened variants of the OMI or hybrid models that refine item composition to optimize goodness-of-fit indices (e.g., Comparative Fit Index [CFI] > .90, Root Mean Square Error of Approximation [RMSEA] < .06).

10. Instrument / Measurement Tool

The Opinions About Mental Illness Scale is a standardized, self-administered, paper-and-pencil (or modern computerized) rating inventory comprising 51 declarative statements.

Structural Specifications

• Test Type: Multidimensional attitudinal rating scale / Self-report questionnaire.
• Item Count: 51 standardized items distributed across five primary subscales.
• Subscale Allocation:
  • Scale A (Authoritarianism): 11 items assessing custodial coercion, discipline, and patient inferiority.
  • Scale B (Benevolence): 14 items reflecting paternalistic charity, humanistic kindness, and protective shelter.
  • Scale C (Mental Hygiene Ideology): 9 items measuring adherence to the modern medical/scientific model of mental illness.
  • Scale D (Social Restrictiveness): 10 items assessing the civil disenfranchisement, containment, and social exclusion of patients.
  • Scale E (Interpersonal Etiology): 7 items attributing mental disorder causation to maladaptive childhood rearing and family dynamics.
• Response Format: 6-point Likert-type response scale. The instrument explicitly forces a directional choice by omitting a neutral midpoint:

Administration Guidelines

• Target Population: Psychiatric and medical hospital staff (physicians, psychologists, social workers, nurses, paraprofessionals), allied health students, and adult community samples (ages 18 and older).
• Administration Time: Approximately 15 to 25 minutes for complete self-administration.
• Instructions to Respondents: Respondents are directed to read each statement carefully and express their personal point of view by selecting the rating that most accurately reflects their level of agreement or disagreement. Standard instructions emphasize that there are no objectively “right” or “wrong” answers and assure respondents of institutional anonymity to minimize social desirability bias.

Scoring Protocol and Directionality

• Raw Item Scoring: Numerical values from 1 to 6 are assigned to each response according to whether the item is keyed positively or negatively toward the target dimension.
• Subscale Score Derivation: Individual subscale scores are computed by summing the numerical weights of the designated items for each factor. Alternatively, mean subscale scores (ranging from 1.0 to 6.0) are calculated by dividing the sum by the number of items on that specific subscale, preserving the intuitive 6-point response metric.
• Normative Interpretation: High scores on Scale A (Authoritarianism) and Scale D (Social Restrictiveness) indicate higher levels of custodial, restrictive, and stigmatizing beliefs. High scores on Scale B (Benevolence) indicate strong paternalistic empathy. High scores on Scale C (Mental Hygiene Ideology) indicate an endorsement of medical-scientific and psychosocial rehabilitation frameworks. High scores on Scale E (Interpersonal Etiology) indicate the attribution of mental breakdown to early family and parental socialization failures.

11. Permissions & Fee and Test Year

Test Publication Year: 1962.

Historical Context & Copyright: The Opinions About Mental Illness Scale was initially formulated under the auspices of the Veterans Administration (VA) Psychiatric Evaluation Project, a publicly funded United States government research initiative, and published in the peer-reviewed scholarly literature through the American Psychological Association (APA) in the Journal of Abnormal and Social Psychology (Cohen & Struening, 1962). Under United States legal precedents concerning archival scientific research from that era, the theoretical dimensions, methodological framework, and scientific findings reside within the academic domain.

Licensing and Research Availability: For independent academic, non-commercial, educational, and clinical research purposes, the OMI is generally accessed via scholarly libraries and scientific archives under standard academic fair use provisions. Researchers are not required to pay licensing fees for non-profit academic research. However, formal commercial reproduction, clinical software integration, or inclusion in commercial test batteries remains subject to the copyright and reproduction permissions of the original publisher (American Psychological Association) and the legal estates of the original authors.

Contact and Sourcing: Investigators wishing to adapt, translate, or digitally deploy the complete formal test inventory are advised to consult the original 1962 publication through the APA archives or obtain permissions through the APA Rights and Permissions office.

12. References

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950). The authoritarian personality. Harper & Row.

Cohen, J., & Struening, E. L. (1962). Opinions about mental illness in the personnel of two large mental hospitals. The Journal of Abnormal and Social Psychology, 64(5), 349–360. https://doi.org/10.1037/h0045526

Cohen, J., & Struening, E. L. (1963). Opinions about mental illness: Mental hospital occupational profiles and profiles of issue. Psychological Reports, 12(1), 111–124. https://doi.org/10.2466/pr0.1963.12.1.111

Cohen, J., & Struening, E. L. (1964). Opinions about mental illness: Hospital social atmosphere profiles and their relevance to effectiveness. The Journal of Consulting Psychology, 28(4), 291–298. https://doi.org/10.1037/h0041285

Gilbert, D. C., & Levinson, D. J. (1956). Ideology, personality, and institutional policy in the mental hospital. The Journal of Abnormal and Social Psychology, 53(3), 263–271. https://doi.org/10.1037/h0045863

Gilbert, D. C., & Levinson, D. J. (1957). “Custodialism” and “humanism” in mental hospital ideology and in staff personality. In M. Greenblatt, D. J. Levinson, & R. H. Williams (Eds.), The patient and the mental hospital (pp. 20–35). Free Press.

Goffman, E. (1961). Asylums: Essays on the social situation of mental patients and other inmates. Doubleday Anchor Books.

Rabkin, J. G. (1972). Opinions about mental illness: A review of the literature. Psychological Bulletin, 77(3), 153–171. https://doi.org/10.1037/h0032338

Struening, E. L., & Cohen, J. (1963). Factorial invariance and validity of the Opinions About Mental Illness scale. The Journal of Abnormal and Social Psychology, 66(2), 134–141. https://doi.org/10.1037/h0046648

Taylor, S. M., & Dear, M. J. (1981). Scaling community attitudes toward the mentally ill. Schizophrenia Bulletin, 7(2), 225–240. https://doi.org/10.1093/schbul/7.2.225

13. Items of the Scale