Social Phobia Scale (SPS)

1. Abstract

The Social Phobia Scale (SPS) is a 20-item self-report psychometric instrument developed by Richard P. Mattick and J. Christopher Clarke to assess scrutiny anxiety, a hallmark dimension of social anxiety disorder (SAD) characterized by the pronounced fear of being observed, evaluated, or judged while carrying out routine, everyday tasks. Originally formulated alongside the Social Interaction Anxiety Scale (SIAS), the SPS isolates performance and scrutiny fears from broad interpersonal interaction distress, offering clinicians and researchers fine-grained sensitivity in diagnosing and treating specific performance-based phobic manifestations. Respondents evaluate items on a 5-point Likert-type scale ranging from 0 (“Not at all characteristic or true of me”) to 4 (“Extremely characteristic or true of me”), yielding a unidimensional global severity score between 0 and 80. Psychometric evaluations across clinical, undergraduate, and epidemiological cohorts consistently demonstrate exceptional internal consistency (Cronbach’s alpha typically exceeding .89 to .94) and robust test-retest reliability across multi-week intervals ($r > .90$). Extensive construct, convergent, and discriminant validity analyses confirm strong correlations with related indices of performance anxiety, fear of negative evaluation, and autonomic reactivity, while maintaining distinct conceptual separation from depressive symptoms, agoraphobic avoidance, and generalized trait anxiety. Exploratory and confirmatory factor analyses broadly support an overarching general factor of scrutiny anxiety, though empirical sub-dimensions reflecting fears of exhibiting visible somatic symptoms (trembling, blushing, sweating), ingestion behaviors, and public performance are frequently identified. The SPS represents a gold-standard diagnostic and treatment monitoring tool within empirical psychopathology, behavioral medicine, and cognitive behavioral therapy protocols worldwide.

2. Keywords

Social Phobia Scale, SPS, scrutiny anxiety, social anxiety disorder, social phobia, psychometrics, performance anxiety, fear of negative evaluation, cognitive behavioral therapy, behavioral assessment, observational anxiety, somatic symptoms

3. Authors

The Social Phobia Scale was created and psychometrically validated by Richard P. Mattick and J. Christopher Clarke. At the time of the instrument’s initial development and formal publication, both authors were affiliated with the School of Psychology at the University of New South Wales (UNSW Sydney) and the National Drug and Alcohol Research Centre (NDARC) in Sydney, New South Wales, Australia. Richard P. Mattick, PhD, has served as a distinguished Professor of Clinical Psychology and Drug and Alcohol Studies at UNSW, authoring seminal treatises on behavioral treatment paradigms, addictive behaviors, and anxiety disorders. J. Christopher Clarke, PhD, was an influential clinical psychologist and academic recognized for his theoretical and empirical work on behavioral conditioning, exposure therapy frameworks, and the conceptual modeling of human fears and phobias.

4. Purpose

The primary clinical and research objective of the Social Phobia Scale is to measure the severity of scrutiny fear—defined as the persistent, debilitating apprehension that one is being watched, monitored, or scrutinized by others during standard daily functioning. In the historical evolution of psychiatric nosology under the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R, DSM-IV, and DSM-5), social phobia was conceptualized as encompassing two distinct, albeit overlapping, symptom clusters: fears of dyadic social interaction (such as initiating conversations, approaching authority figures, or attending parties) and fears of being observed or evaluated during circumscribed performance activities (such as eating, drinking, writing, speaking, or moving in public spaces). Prior to the publication of the SPS and SIAS, standard assessment batteries—such as the Fear of Negative Evaluation Scale (FNE) and the Social Avoidance and Distress Scale (SAD)—primarily measured generalized social distress or cognitive concerns regarding disapproval, lacking the behavioral specificity required to isolate observational fears from conversational impairment.

To address this measurement gap, Mattick and Clarke constructed the SPS to capture situational scrutiny concerns across diverse real-world contexts. From a clinical perspective, the SPS serves multiple functions: as a baseline screening tool to identify patients whose social anxiety is primarily observation-focused; as a differential diagnostic aid to delineate social phobia from panic disorder, agoraphobia, and body dysmorphic disorder; and as a sensitive outcome measure capable of detecting therapeutic changes following cognitive restructuring, interoceptive exposure, or pharmacological interventions. Because patients presenting with scrutiny phobia often organize their lives around subtle safety behaviors—such as sitting against walls, concealing hands to hide tremors, or avoiding public restrooms—the SPS directly queries situations where scrutiny vulnerability is acute. In research contexts, the instrument permits structural modeling of social anxiety phenotypes, comparative treatment trials, experimental studies of attentional bias under social threat, and psychophysiological investigations examining autonomic hyperarousal in response to real or simulated evaluation.

5. Psychological Construct

The core psychological construct operationalized by the Social Phobia Scale is scrutiny anxiety (also referred to in the literature as observational fear or performance social anxiety). Scrutiny anxiety represents an excessive, irrational expectation that one’s behavior, physical coordination, or autonomic physiological responses will be monitored by onlookers, interpreted as evidence of personal inadequacy, mental instability, incompetence, or physical illness, and ultimately lead to severe humiliation, ridicule, or social rejection.

Within this overarching construct, three interconnected sub-domains or thematic symptom dimensions are systematically measured by the instrument:

• Visible Somatic Symptom Exposure: This dimension taps intense fears regarding the public disclosure of involuntary autonomic nervous system activation. Items query catastrophic worries that observers will notice shaking or trembling (e.g., Item 7, Item 12), facial blushing (Item 5), fainting, or visible illness (Item 9). Individuals scoring high on this dimension frequently develop hypervigilant interoceptive attention toward proprioceptive and cardiovascular cues, fearing that any physiological perturbation will betray their internal anxiety and invite disparagement.
• Scrutiny During Public Motor and Ingestive Behaviors: This domain evaluates the disruption of procedural motor activities when execution occurs in the visual field of others. Scenarios include eating in restaurants (Item 10), drinking in the presence of others (Item 13), writing or signing documents under observation (Item 1), and maintaining motor coordination while walking down the street (Item 4) or entering a room where people are already seated (Item 6). In these situations, the construct reflects a breakdown of automatic motor functioning driven by explicit internal monitoring and fear of perceived “clumsiness” or awkwardness (Item 11, Item 19).
• Formal Public Performance and Social Visibility: This facet captures anxiety triggered by structured, visible social positioning where the individual is the explicit focus of collective group attention. Situations include public speaking, giving an oral presentation to an audience (Item 16, Item 17), sitting directly opposite commuters on public transit (Item 8), utilizing public restroom facilities (paruresis; Item 2), or maintaining eye contact (Item 14). Here, the construct encompasses both fear of verbal stumbling or communicative incompetence and general distress related to unescapable public visibility.

6. Theoretical Framework

The theoretical architecture of the SPS is rooted in cognitive-behavioral and evolutionary models of psychopathology, most notably the cognitive models of social phobia advanced by Aaron T. Beck, David M. Clark, and Adrian Wells. Clark and Wells’ (1995) cognitive model posits that when individuals with social anxiety enter a feared social or performance situation, they experience a marked shift in attentional allocation from external social cues to an intense, self-focused monitoring process. The individual uses internal interoceptive information (such as somatic tremors, heart palpitations, or perceived facial heat) to construct a distorted, third-person “mental image” of how they appear to the audience. High scores on the SPS reflect this catastrophic self-evaluative process: the respondent presumes that an observer perceives an exaggerated, grotesque caricature of their visible anxiety.

Evolutionary psychology frameworks, such as those proposed by Paul Gilbert and Michael Trower, provide complementary theoretical grounding. These models suggest that social phobia stems from evolved bio-behavioral defense mechanisms designed to navigate dominance hierarchies and avoid social conflict. Subordinates within social hierarchies monitor their behavior to prevent signaling inappropriate competition or exhibiting unacceptable vulnerability that could trigger ostracism or aggressive displacement by higher-ranking conspecifics. Scrutiny fear can thus be conceptualized as an overactivated defense mechanism wherein an individual perceives all onlookers as judgmental evaluators possessing the power to marginalize, humiliate, or cast out the person who fails to conform to perceived behavioral standards.

Furthermore, behavioral learning theories emphasize the interplay of classical conditioning, vicarious learning, and operant reinforcement in maintaining scrutiny anxiety. A catastrophic conditioning event—such as spilling a beverage, experiencing a panic attack during a presentation, or suffering public ridicule during hand trembling—creates an enduring conditioned fear response to situational cues involving observation. To prevent re-exposure to this conditioned distress, individuals deploy safety behaviors (e.g., clasping glasses tightly, wearing heavy makeup, avoiding eye contact), which prevent the disconfirmation of unrealistic danger expectancies and maintain phobic avoidance over the lifespan.

7. Validity

The validity of the Social Phobia Scale has been extensively corroborated across diverse clinical cohorts (including treatment-seeking outpatients with DSM-defined social anxiety disorder) and non-clinical samples across North America, Europe, Asia, and Australasia.

Construct and Convergent Validity: In the original validation studies by Mattick and Clarke (1998), the SPS demonstrated substantial positive correlations with instruments tapping closely related dimensions of social evaluative anxiety. Strong convergent correlations were observed with the Social Interaction Anxiety Scale (SIAS; $r = .54$ to $.69$), the Fear of Negative Evaluation Scale (FNE; $r = .53$ to $.66$), and the Social Avoidance and Distress Scale (SAD; $r = .54$ to $.68$). When compared to physiological and performance-specific assessment tools, the SPS correlates robustly with the Personal Report of Confidence as a Speaker (PRCS; $r > .60$) and behavioral avoidance tasks involving impromptu public speaking or eating in front of confederates.

Discriminant Validity: The SPS exhibits clear discriminant divergence from constructs that do not center on social evaluation. In clinical trials, correlations between the SPS and general depression scales—such as the Beck Depression Inventory (BDI)—fall within a moderate range ($r = .35$ to $.48$), indicating that while dysphoria commonly co-occurs with severe social phobia, the SPS does not merely index general negative affectivity. Similarly, correlations with the agoraphobia and panic subscales of the Fear Questionnaire (FQ) remain modest ($r = .20$ to $.38$), demonstrating that the scale discriminates panic attacks driven by fears of catastrophic medical collapse (e.g., dying or losing control) from anxiety driven by public embarrassment.

Predictive and Known-Groups Validity: The SPS successfully differentiates individuals diagnosed with social anxiety disorder from both healthy controls and clinical outpatients diagnosed with other anxiety disorders, including panic disorder, generalized anxiety disorder, and obsessive-compulsive disorder. Clinical samples typically yield mean SPS scores between 28.0 and 38.0, whereas non-clinical undergraduate and community norms range from 9.0 to 15.0. Furthermore, the SPS demonstrates high predictive sensitivity to treatment effects: significant, large-magnitude decreases in SPS scores (often with effect sizes exceeding Cohen’s $d = 1.0$) are documented following courses of cognitive behavioral group therapy (CBGT) and selective serotonin reuptake inhibitor (SSRI) pharmacotherapy.

8. Reliability

The reliability of the Social Phobia Scale has been thoroughly documented, demonstrating high internal consistency and temporal stability across varying measurement intervals.

Internal Consistency: Across psychometric investigations, the SPS exhibits excellent internal consistency. In the foundational validation studies by Mattick and Clarke (1998), Cronbach’s alpha coefficients for the full 20-item scale were .89 within an undergraduate sample ($N = 243$) and .89 within an outpatient clinical social phobia cohort ($N = 127$). Subsequent large-scale investigations have replicated these metrics. For example, Peters (2000) documented an alpha of .94 in a large clinical sample, and Carleton et al. (2009) reported alphas ranging between .92 and .94 in independent undergraduate and patient groups. Corrected item-total correlations across the 20 items consistently surpass .45, with the majority falling between .55 and .75, indicating that each individual item contributes meaningfully to the measured construct without excessive redundancy.

Test-Retest Reliability: The temporal stability of the SPS over time is robust when administered to untreated samples. Mattick and Clarke (1998) evaluated test-retest reliability across intervals of one to three weeks in university cohorts and patient samples awaiting therapeutic intervention, observing test-retest correlation coefficients of $r = .91$ to $.93$. Longer test-retest evaluations over 4-week, 8-week, and 12-week spans within non-treatment control arms have yielded stability coefficients consistently between .78 and .86, demonstrating that scrutiny anxiety represents a stable, trait-like psychological vulnerability in the absence of targeted cognitive-behavioral or medical intervention.

9. Factor Analysis

The structural dimensionality of the SPS has been the focus of numerous exploratory factor analyses (EFA) and confirmatory factor analyses (CFA), yielding consistent psychometric evidence regarding its structural architecture.

Although Mattick and Clarke originally conceived the SPS as a predominantly unidimensional inventory representing broad scrutiny distress, subsequent EFAs using principal axis factoring and oblique rotation have demonstrated that the 20 items cluster into distinct, interpretable first-order factors that load onto a single higher-order scrutiny construct. Commonly resolved factor solutions identify three prominent sub-dimensions:

• Factor 1: Observation/Performance in Public (Motor/Ingestive Acts): Capturing items reflecting eating, drinking, writing, or sitting under public surveillance (e.g., Items 1, 6, 8, 10, 13). Item loadings for this factor typically range between .58 and .82.
• Factor 2: Fear of Involuntary Somatic Signs: Encompassing items specifically referencing the visibility of autonomic arousal, such as blushing, hand tremors, and physical illness (e.g., Items 5, 7, 9, 12). Item loadings typically range from .62 to .86.
• Factor 3: Formal Public Speaking and Center of Attention: Grouping items that describe speaking to groups, giving presentations, and making eye contact (e.g., Items 3, 14, 16, 17), with factor loadings generally spanning .60 to .85.

Confirmatory factor analytic studies (e.g., Carleton et al., 2009; Safren et al., 1999) have formally evaluated one-factor, orthogonal multi-factor, and hierarchical bifactor models. In large heterogeneous samples, while a strict unidimensional model often yields borderline acceptable fit indices ($\chi^2/df > 3.0$, Comparative Fit Index [CFI] $\approx .88-.91$, Root Mean Square Error of Approximation [RMSEA] $\approx .07-.08$), a hierarchical bifactor model—consisting of a dominant general scrutiny anxiety factor alongside specific orthogonal sub-factors for somatic manifestation and performance situations—exhibits superior fit metrics ($ ext{CFI} > .95$, Tucker-Lewis Index [TLI] > .94,$ ext{RMSEA} < .05$). Because the dominant general factor accounts for over 70% to 80% of the total common variance, psychometricians support the continued clinical use of the single, summed composite score as a parsimonious measure of scrutiny severity.

10. Instrument / Measurement Tool

• Full Instrument Name: Social Phobia Scale (SPS)
• Original Authors: Richard P. Mattick and J. Christopher Clarke (1998)
• Assessment Type: Self-report rating scale / paper-and-pencil questionnaire / computerized assessment
• Construct Measured: Scrutiny anxiety and observational performance fears within social anxiety disorder
• Number of Items: 20 items
• Response Format: 5-point Likert scale: 0 = Not at all characteristic or true of me, 1 = Slightly characteristic or true of me, 2 = Moderately characteristic or true of me, 3 = Very characteristic or true of me, 4 = Extremely characteristic or true of me
• Scoring Procedure: All items are scored from 0 to 4 and summed to yield a total score ranging from 0 to 80. There are no reverse-scored items.
• Clinical Interpretation Guidelines:
  • 0 – 19: Subclinical range / within normal non-clinical community limits.
  • 20 – 24: Mild scrutiny anxiety; potential subthreshold clinical concern.
  • 25 – 33: Moderate scrutiny anxiety; indicative of clinically significant social phobia requiring formal clinical evaluation.
  • 34 – 80: Severe to very severe scrutiny fear; typical of generalized social phobia with prominent performance and observational impairment.
• Administration Time: Approximately 5 to 10 minutes.

11. Permissions & Fee and Test Year

Publication Year: The Social Phobia Scale was officially published in 1998 in Behaviour Research and Therapy, following over a decade of extensive circulation, conference presentations, and preliminary psychometric evaluation dating back to the late 1980s (Mattick & Clarke, 1989 unpublished manuscript).

Permissions and Accessibility: The SPS is widely classified as an open-access psychometric instrument for clinical, educational, and non-commercial scientific research purposes. In their original work and subsequent communications, the authors placed the instrument into the academic public domain to facilitate standardized assessment across psychiatric research settings. No licensing fees or royalty payments are required to administer, score, or reproduce the questionnaire for non-profit academic research, university teaching, or clinical practice. Commercial enterprises, electronic health record vendors, or proprietary test developers intending to incorporate the scale into fee-for-service digital platforms should review copyright guidelines held by the original publisher (Elsevier) regarding the journal article content.

12. References

Carleton, R. N., Collimore, K. C., Asmundson, G. J., McCabe, R. E., Rowa, K., & Antony, M. M. (2009). Refining and validating the Social Phobia Scale and the Social Interaction Anxiety Scale. Depression and Anxiety, 26(2), E71–E81. https://doi.org/10.1002/da.20480

Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In R. G. Heimberg, M. R. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: Diagnosis, assessment, and treatment (pp. 69–93). Guilford Press.

Heimberg, R. G., Mueller, G. P., Holt, C. S., Hope, D. A., & Liebowitz, M. R. (1992). Assessment of anxiety in social situations and examination of the Social Phobia Scale and the Social Interaction Anxiety Scale. Anxiety Research, 5(2), 163–173. https://doi.org/10.1080/08917779208248790

Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36(4), 455–470. https://doi.org/10.1016/S0005-7967(97)10031-6

Peters, L. (2000). Discriminant validity of the Social Phobia and Social Interaction Anxiety Scales. Behaviour Research and Therapy, 38(9), 943–950. https://doi.org/10.1016/S0005-7967(99)00130-1

Safren, S. A., Turk, C. L., & Heimberg, R. G. (1999). Factor structure of the Social Phobia Scale and the Social Interaction Anxiety Scale. Behaviour Research and Therapy, 37(12), 1179–1189. https://doi.org/10.1016/S0005-7967(98)00201-9

13. Items of the Scale