Internalized Stigma of Mental Illness Scale (ISMI)

Abstract

The Internalized Stigma of Mental Illness (ISMI) Scale is a premier, internationally recognized psychometric instrument designed to evaluate the subjective experience of self-stigma among individuals diagnosed with psychiatric disorders. Developed by Jennifer Boyd Ritsher (later Jennifer E. Boyd), Karunya P. G. Otilingam, and Marlene Grajales in 2003 at the University of California, San Francisco and the San Francisco Department of Veterans Affairs Medical Center, the instrument operationalizes the psychological process through which individuals endorse, internalize, and apply societal prejudices to themselves. The full scale consists of 29 self-report items distributed across five theoretically driven subscales: Alienation, Stereotype Endorsement, Perceived Discrimination, Social Withdrawal, and Stigma Resistance. Responses are captured via a 4-point Likert scale ranging from 1 (Strongly Disagree) to 4 (Strongly Agree).

Psychometrically, the ISMI exhibits exceptional internal consistency (full-scale Cronbach’s alpha $\alpha = .90$, with subscale alphas typically spanning $.71$ to $.87$) and strong test-retest reliability ($r = .92$ over a six-week interval). Its construct validity is reinforced by substantial positive correlations with depressive symptomatology, demoralization, and psychiatric symptom severity, alongside robust negative correlations with self-esteem, self-efficacy, hope, empowerment, and social functioning. The instrument has been translated and psychometrically validated across more than 50 languages and clinical contexts worldwide. It remains the gold standard in epidemiological investigations, clinical trials, and recovery-oriented intervention research targeting the pervasive and deleterious sequelae of self-stigmatization.

Keywords

Internalized stigma, self-stigma, ISMI, mental illness stigma, psychometrics, alienation, stereotype endorsement, perceived discrimination, social withdrawal, stigma resistance

Authors

The Internalized Stigma of Mental Illness (ISMI) Scale was conceived, developed, and validated by a multidisciplinary team of psychiatric epidemiologists, clinical psychologists, and health services researchers:

• Jennifer Boyd Ritsher, Ph.D. (subsequently publishing as Jennifer E. Boyd): Associate Professor in Residence, Department of Psychiatry, University of California, San Francisco (UCSF), and Staff Psychologist/Health Science Specialist, San Francisco Veterans Affairs Medical Center (SFVAMC), San Francisco, California, United States.
• Karunya P. G. Otilingam, B.A.: Clinical Research Coordinator and Psychometric Analyst, SFVAMC and Department of Psychiatry, University of California, San Francisco, California, United States.
• Marlene Grajales, M.S.W.: Clinical Social Worker and Research Associate, Mental Health Service, San Francisco Veterans Affairs Medical Center, San Francisco, California, United States.

Purpose

Public stigma surrounding psychiatric disorders remains one of the most substantial structural and social barriers confronting individuals with mental illness. However, the intrapsychic manifestation of this phenomenon—termed internalized stigma or self-stigma—imposes an equally devastating psychological burden. The primary purpose of the Internalized Stigma of Mental Illness (ISMI) Scale is to systematically assess the extent to which individuals living with severe psychiatric conditions (including schizophrenia-spectrum disorders, bipolar disorder, and major depressive disorder) absorb negative societal representations, convert cultural stereotypes into enduring personal beliefs, and alter their behavior accordingly.

From a clinical and public health perspective, self-stigma is not a benign cognitive bias; it directly instigates the catastrophic cascade known as the “why try” effect. When individuals endorse prevailing stereotypes—such as beliefs that persons with psychiatric illnesses are incompetent, dangerous, or inherently untrustworthy—they often conclude that pursuing independent employment, sustained interpersonal relationships, educational advancement, or independent living is futile. This subjective resignation exacerbates demoralization, undermines self-esteem, diminishes perceived agency, and severely hampers therapeutic alliance and treatment compliance. Consequently, self-stigma directly fuels premature treatment discontinuation, secondary psychiatric morbidity, and elevated rates of suicidal ideation.

The ISMI was intentionally designed to bridge a prominent empirical gap: while public attitudes and enacted structural discrimination were frequently quantified, subjective, nuanced measurements of personal stigma internalization were largely underdeveloped or psychometrically inadequate. Beyond quantifying the psychosocial harms of internalized stigma, the ISMI uniquely captures a counter-mechanism: Stigma Resistance. This dimension operationalizes a respondent’s psychological resilience, critical consciousness, and capacity to actively challenge, neutralize, or deflect cultural biases. In clinical practice, the ISMI serves as an essential diagnostic baseline and treatment-monitoring metric for psychotherapeutic modalities such as Narrative Enhancement and Cognitive Therapy (NECT) and Ending Self-Stigma (ESS). In psychiatric research, it functions as a primary dependent, mediating, or moderating variable within recovery models.

Psychological Construct

Internalized stigma represents a multi-tiered, phenomenological construct situated at the intersection of psychiatric symptomatology, sociocognitive appraisal, and personal identity transformation. Rather than reflecting a unidimensional attitude, the construct assessed by the ISMI is multifaceted, comprising five distinct yet mutually reinforcing domains:

1. Alienation (6 items)

Alienation captures the visceral, subjective experience of being separated from mainstream humanity, fundamentally devalued, flawed, or rendered an outsider due to having a mental illness. This dimension taps feelings of existential isolation, profound self-disappointment, and shame (e.g., Item 1: “I feel out of place in the world because I have a mental illness”; Item 16: “I am embarrassed or ashamed that I have a mental illness”). Alienation directly reflects the existential fracture between the respondent’s pre-morbid or idealized self-concept and their post-diagnostic perceived identity.

2. Stereotype Endorsement (7 items)

Stereotype Endorsement measures cognitive agreement with widespread societal misconceptions and pejorative characterizations regarding psychiatric populations, along with the personal attribution of these deficits to oneself. It probes beliefs concerning inherent incompetence, violence, emotional childishness, and an inability to sustain conventional adult societal roles (e.g., Item 2: “Mentally ill people tend to be violent”; Item 6: “People with mental illness cannot live a good, rewarding life”; Item 27: “Because I have a mental illness, I am not capable of holding a job”). This subscale represents the cognitive conversion of cultural myth into autobiographical truth.

3. Perceived Discrimination (5 items)

Perceived Discrimination evaluates the respondent’s subjective perception of how they are actively treated by the social milieu—encompassing interpersonal hostility, structural bias, condescension, and social devaluation explicitly triggered by their psychiatric status (e.g., Item 3: “People discriminate against me because I have a mental illness”; Item 20: “People often patronize me, or treat me like a child, just because I have a mental illness”). Importantly, this scale measures not only external historical events, but the respondent’s psychological vigilance and subjective appraisal of systemic societal unfairness.

4. Social Withdrawal (6 items)

Social Withdrawal quantifies intentional behavioral seclusion, avoidance of interpersonal contact, and relational retreat undertaken proactively to prevent prospective rejection, preserve social status, or shield loved ones from perceived social shame (e.g., Item 9: “I don’t socialize as much as I used to because my mental illness might make me look or behave ‘weird'”; Item 19: “I stay away from social situations in order to protect my family or friends from embarrassment”). This dimension reflects self-protective behavioral constriction, which paradoxically cements social isolation and deepens objective disability.

5. Stigma Resistance (5 items)

Stigma Resistance assesses the adaptive capacity to remain psychologically intact, self-affirming, and uncooperative with societal stigmatization. It evaluates personal agency, resilience, self-efficacy, and the intellectual rejection of the stigmatizing label as an exhaustive definition of the self (e.g., Item 7: “I can have a good, fulfilling life, despite my mental illness”; Item 25: “I can make a significant contribution to society, despite my mental illness”). This subscale is framed positively; when computing overall internalized stigma scores, these items are reverse-coded, or alternatively, analyzed as an autonomous, orthogonal construct of psychological empowerment.

Theoretical Framework

The construction of the ISMI is rooted in several converging paradigms within medical sociology, social psychology, and clinical psychology, most notably Modified Labeling Theory, Social Identity Theory, and Cognitive-Behavioral Models of Self-Stigmatization.

Modified Labeling Theory

Formulated by Bruce G. Link and colleagues (1989), Modified Labeling Theory posits that during socialization, individuals develop generalized cognitive expectations regarding how society perceives and treats individuals labeled with mental illness. Most individuals internalize the belief that mental patients are socially devalued and rejected. As long as a person remains free of a psychiatric diagnosis, these expectations remain latent and benign. However, upon psychiatric diagnosis and formal clinical treatment, these culturally learned devaluation beliefs become self-referential.

The individual suddenly anticipates that societal rejection will now target them personally. To mitigate this threat, individuals often engage in defensive coping strategies—specifically social withdrawal, secrecy, and interpersonal constriction. Tragically, these self-protective behaviors restrict social support networks, impair vocational capacity, and precipitate objective socioeconomic marginalization, thereby validating the initial internalized devaluation. The ISMI was constructed directly to quantify the steps within Link’s model, particularly the cognitive internalization of devaluation and the corresponding behavioral withdrawal.

Goffman’s Theory of Spoiled Identity

The foundational sociological architecture traces back to Erving Goffman‘s seminal treatise Stigma: Notes on the Management of Spoiled Identity (1963). Goffman conceptualized stigma as an attribute that deeply discredits an individual, reducing them from a whole, usual person to a tainted, discounted one. Goffman differentiated between the discredited (those whose differences are immediately apparent) and the discreditable (those whose mental health status is concealable). The ISMI operationalizes the chronic, debilitating cognitive vigilance that discreditable individuals endure to prevent identity contamination, alongside the profound disruption of identity that ensues once a psychiatric label becomes indelible.

The Progressive Model of Self-Stigma

In contemporary clinical psychology, the ISMI aligns closely with Patrick Corrigan’s four-stage hierarchical paradigm of self-stigma development:

1. Awareness: The individual demonstrates cognitive awareness of ambient cultural stereotypes (e.g., “Society believes mental patients are unpredictable”).
2. Agreement: The individual cognitively concurs with these negative stereotypes (e.g., “Society is right; people with mental illness are unpredictable”).
3. Application: The individual personalizes these stereotypes to their self-concept (e.g., “Because I have schizophrenia, I am unpredictable and cannot hold a job”).
4. Harm: The individual suffers a decline in self-esteem, self-efficacy, and subjective dignity, withdrawing from normative societal participation.

The ISMI deliberately bypasses mere theoretical “awareness” to assess true agreement (Stereotype Endorsement), application (Alienation), and behavioral harm (Social Withdrawal), while simultaneously honoring the patient’s dialectical capacity to resist this trajectory (Stigma Resistance).

Validity

The psychometric validity of the ISMI has been rigorously documented across numerous empirical investigations involving varied clinical populations, ranging from outpatient community mental health clients to forensic and long-term inpatient cohorts.

Construct and Factorial Validity

In their original validation study involving 127 outpatients presenting with major depressive disorder, schizoaffective disorder, and schizophrenia, Ritsher et al. (2003) demonstrated robust construct validity. Principal components analysis and confirmatory models confirmed that the five subscales accounted for substantial variance and represented empirically distinct, yet theoretically related, domains of stigma experience. Subsequent large-scale investigations, such as those conducted by Boyd, Adler, Otilingam, and Peters (2014) in an international cohort of over 1,000 participants, reaffirmed the distinct multi-dimensional structural integrity of the instrument.

Convergent and Discriminant Validity

The convergent validity of the ISMI has been repeatedly demonstrated across clinical populations:

• Depression and Demoralization: ISMI total scores correlate positively and significantly with depressive symptom severity on the Beck Depression Inventory (BDI-II) and the Center for Epidemiologic Studies Depression Scale (CES-D), with correlation coefficients typically ranging between $r = .45$ and $r = .65$ ($p < .001$).
• Self-Esteem: Strong, negative correlations are routinely observed between the ISMI and the Rosenberg Self-Esteem Scale (RSES), with coefficients clustering between $r = -.55$ and $r = -.72$, demonstrating that elevated self-stigma corresponds with substantial decrements in global self-worth.
• Empowerment and Hope: Internalized stigma shows profound inverse associations with the Boston University Empowerment Scale ($r = -.50$ to $-.63$) and the State-Trait Hope Scale ($r = -.48$ to $-.60$).
• Symptom Burden and Insight: Moderate positive associations are observed with psychiatric symptom metrics (e.g., PANSS negative and general psychopathology scales). Crucially, clinical insight into mental illness often interacts with the ISMI; individuals with high clinical insight but low stigma resistance experience the highest rates of depressive despair, a phenomenon recognized in psychiatric literature as the “insight paradox.”

Discriminant validity is supported by weak to non-significant correlations with demographic characteristics such as age, biological sex, and absolute physical health status, confirming that the scale assesses a specific socio-psychological vulnerability rather than general medical distress or demographic variation.

Predictive and Criterion-Related Validity

Prospective longitudinal studies indicate that baseline ISMI scores systematically predict poor vocational trajectory, non-adherence to pharmacotherapy and psychotherapy, social network attrition, and reduced health-related quality of life (HRQoL) at 6-month and 12-month follow-ups. A meta-analysis by Livingston and Boyd (2010), synthesizing data from 45 empirical studies, confirmed that internalized stigma measured via the ISMI demonstrated a reliable, medium-to-large effect size in predicting negative psychiatric recovery outcomes.

Reliability

The ISMI displays robust reliability indices across psychometric evaluations, supporting its deployment in high-stakes research and individualized clinical diagnostics.

Internal Consistency

In the original normative sample by Ritsher et al. (2003), the overall 29-item scale demonstrated high internal consistency, yielding an overall Cronbach’s alpha of $\alpha = .90$. Subscale-level internal consistency coefficients demonstrated substantial reliability across the primary internalized domains:

• Alienation: $\alpha = .84$
• Social Withdrawal: $\alpha = .80$
• Stereotype Endorsement: $\alpha = .71$
• Perceived Discrimination: $\alpha = .82$
• Stigma Resistance: $\alpha = .58$

The modest internal consistency observed for the Stigma Resistance subscale ($\alpha = .58$ in the initial 2003 study, fluctuating between $.52$ and $.68$ in subsequent international cohorts) has been extensively discussed in psychometric literature. Factor analysts have identified that several Stigma Resistance items (specifically Item 22, concerning anger toward perpetrators of stigma) introduce distinct affective and attributional facets that deviate from pure cognitive resistance. Consequently, psychometricians frequently recommend analyzing Stigma Resistance as an independent subscale or evaluating its psychometric performance separately from the global internalized stigma composite score.

Test-Retest Reliability

Temporal stability was established by Ritsher and colleagues via a six-week re-administration interval to clinically stable outpatients. The test-retest reliability coefficient for the overall scale was $r = .92$ ($p < .0001$), signifying exceptional longitudinal stability in the absence of targeted psychosocial intervention. Subscale test-retest correlations were similarly elevated, spanning $r = .78$ (Stigma Resistance) to $r = .89$ (Alienation).

Factor Analysis

The latent structure of the ISMI has been subjected to rigorous exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) across dozens of independent cultural samples.

Original Exploratory Structural Findings

Ritsher, Otilingam, and Grajales (2003) initially subjected the preliminary item pool to principal components analysis (PCA) with oblimin oblique rotation, given the theoretical assumption that the dimensions of internalized stigma covary. The exploratory extraction confirmed that the items clustered cleanly across the five hypothesized conceptual domains, with primary factor loadings routinely exceeding $lambda = .45$ and cross-loadings remaining low to moderate.

Confirmatory Factor Analytic (CFA) Validations

Subsequent psychometric testing has systematically compared competitive factor models:

• Unidimensional Model: Posits that all 29 items load onto a single general internalized stigma factor. This model routinely demonstrates unacceptable fit indices (e.g., $\chi^2/df > 4.0$, $\text{RMSEA} > .09$, $\text{CFI} < .80$), confirming that internalized stigma cannot be psychometrically reduced to a single global attribute.
• Five-Factor Correlated Model: Represents the original five theoretical dimensions (Alienation, Stereotype Endorsement, Perceived Discrimination, Social Withdrawal, and Stigma Resistance) as correlated latent factors. This model exhibits good fit across diverse samples: $\chi^2/df \approx 1.85 – 2.20$, Root Mean Square Error of Approximation ($\text{RMSEA}$) $\approx .048 – .058$, Comparative Fit Index ($\text{CFI}$) $\approx .92 – .95$, and Standardized Root Mean Square Residual ($\text{SRMR}$) $\approx .045 – .060$.
• Hierarchical / Higher-Order Model: Posits that a higher-order latent construct (Internalized Stigma) accounts for the covariance among the first-order factors. When Stigma Resistance is excluded or modeled separately, the four-factor higher-order model (Alienation, Stereotype Endorsement, Perceived Discrimination, and Social Withdrawal) shows excellent psychometric fit indices across clinical cohorts (e.g., Boyd et al., 2014; Sibitz et al., 2010).

Because the Stigma Resistance subscale frequently displays low or negative factor loadings onto the overarching internalized stigma second-order factor, consensus in advanced structural equation modeling dictates treating Stigma Resistance either as an independent, moderating construct or utilizing the 24-item, four-factor variant when testing structural regression models of self-stigma pathology.

Instrument / Measurement Tool

• Instrument Name: Internalized Stigma of Mental Illness Scale (ISMI)
• Alternative Versions: ISMI-29 (Full Scale), ISMI-10 (Brief/Short Form), ISMI-9
• Instrument Type: Self-administered psychometric rating scale (examiner administration is permissible for respondents experiencing severe cognitive or psychiatric impairments)
• Target Population: Adolescents and adults (ages 16+) diagnosed with psychiatric disorders (schizophrenia-spectrum conditions, bipolar disorder, major depressive disorder, severe anxiety disorders, and personality disorders)
• Item Count: 29 items
• Subscale Allocation:
  • Alienation: 6 items (1, 5, 8, 16, 17, 21)
  • Stereotype Endorsement: 7 items (2, 6, 10, 13, 18, 23, 27)
  • Perceived Discrimination: 5 items (3, 14, 20, 24, 28)
  • Social Withdrawal: 6 items (4, 9, 11, 15, 19, 29)
  • Stigma Resistance: 5 items (7, 12, 22, 25, 26)
• Response Scale: 4-point Likert scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Agree
  • 4 = Strongly Agree
• Scoring and Transformation Rules:
  • Prior to calculating the full-scale composite score, the five Stigma Resistance items (7, 12, 22, 25, 26) must be reverse-scored: $1 to 4$, $2 to 3$, $3 to 2$, $4 to 1$.
  • Scores for individual subscales and the total composite scale are calculated as continuous mean item scores (sum of applicable item responses divided by the number of completed items in that scale), yielding an intuitive score range from 1.00 to 4.00.
  • Alternatively, researchers and clinicians often compute the total score using solely the 24 internalized stigma items (excluding Stigma Resistance entirely) and report the Stigma Resistance mean score as an autonomous index of empowerment and psychological resilience.
• Clinical Severity Cut-Off Categorization (Lysaker et al., 2007 method):
  • 1.00 – 2.00: Minimal to no internalized stigma
  • 2.01 – 2.50: Mild internalized stigma
  • 2.51 – 3.00: Moderate internalized stigma
  • 3.01 – 4.00: Severe internalized stigma

Permissions & Fee and Test Year

• Year of Formal Publication: 2003
• Intellectual Property & Licensing: The Internalized Stigma of Mental Illness (ISMI) Scale is published in the public domain for academic, clinical, and non-profit empirical research purposes. Dr. Jennifer E. Boyd and colleagues have explicitly made the instrument open-access to stimulate worldwide stigma research and intervention development.
• Usage Fee: Free ($0.00). No licensing royalties or authorization fees are assessed for educational, clinical, or non-commercial scientific applications.
• Commercial Applications: Commercial entities, pharmaceutical sponsors, or for-profit clinical trials wishing to incorporate the instrument into proprietary digital software or commercial trial batteries should seek permission from the lead author or corresponding institutions (UCSF / San Francisco VA Health Care System).
• Citation Requirement: Any publication, presentation, or derived implementation using the instrument must formally cite the primary psychometric validation article: Ritsher, J. B., Otilingam, P. G., & Grajales, M. (2003). Internalized stigma of mental illness: Psychometric properties of a new measure. Psychiatry Research, 121(1), 31–49.

References

• Boyd, J. E., Adler, E. P., Otilingam, P. G., & Peters, T. (2014). Internalized Stigma of Mental Illness (ISMI) scale: A review of its properties and additions to the literature. Stigma Research and Action, 4(1), 11–24. https://doi.org/10.5463/sra.v4i1.9
• Brohan, E., Gauci, D., Sartorius, N., Thornicroft, G., & GAMIAN-Europe Study Group. (2010). Self-stigma, empowerment and perceived discrimination among people with bipolar disorder or depression in 13 European countries: The GAMIAN-Europe study. Journal of Affective Disorders, 122(3), 232–238. https://doi.org/10.1016/j.jad.2009.07.007
• Corrigan, P. W., & Watson, A. C. (2002). The paradox of self-stigma and mental illness. Clinical Psychology: Science and Practice, 9(1), 35–53. https://doi.org/10.1093/clipsy.9.1.35
• Goffman, E. (1963). Stigma: Notes on the management of spoiled identity. Prentice-Hall.
• Link, B. G., Cullen, F. T., Struening, E., Shrout, P. E., & Dohrenwend, B. P. (1989). A modified labeling theory approach to mental disorders: An empirical assessment. American Sociological Review, 54(3), 400–423. https://doi.org/10.2307/2095613
• Livingston, J. D., & Boyd, J. E. (2010). Correlates and consequences of internalized stigma for people living with mental illness: A systematic review and meta-analysis. Social Science & Medicine, 71(12), 2150–2161. https://doi.org/10.1016/j.socscimed.2010.09.030
• Lysaker, P. H., Roe, D., & Yanos, P. T. (2007). Toward understanding the insight paradox: Internalized stigma moderates the association between insight and social functioning, hope, and self-esteem among people with schizophrenia spectrum disorders. Schizophrenia Bulletin, 33(1), 192–199. https://doi.org/10.1093/schbul/sbl016
• Ritsher, J. B., Otilingam, P. G., & Grajales, M. (2003). Internalized stigma of mental illness: Psychometric properties of a new measure. Psychiatry Research, 121(1), 31–49. https://doi.org/10.1016/j.psychres.2003.08.008
• Ritsher, J. B., & Phelan, J. C. (2004). Internalized stigma predicts erosion of morale among psychiatric outpatients. Psychiatry Research, 129(3), 257–265. https://doi.org/10.1016/j.psychres.2004.08.003
• Sibitz, I., Unger, A., Woppmann, A., Zidek, T., & Amering, M. (2010). Revised Internalized Stigma of Mental Illness (ISMI) scale: Validation of the German version. Schizophrenia Research, 122(1–3), 273–274. https://doi.org/10.1016/j.schres.2010.05.028

Items of the Scale