The Social Interaction Anxiety Scale (SIAS)<\/strong> is a premier self-report psychometric instrument designed to evaluate social interaction anxiety\u2014specifically the distress, apprehension, and behavioral inhibition experienced when initiating, engaging in, and sustaining interpersonal discourse with peers, authority figures, and strangers. Developed concurrently with its companion instrument, the Social Phobia Scale (SPS)<\/a>, by Australian clinical psychologists Richard P. Mattick<\/a> and J. Christopher Clarke (first circulated in 1989 and formally published in 1998), the SIAS was formulated to disentangle pervasive interactional fears from specific performance or scrutiny anxiety. The instrument comprises 20 items rated on a 5-point Likert-type response scale ranging from 0 (“Not at all characteristic or true of me”<\/em>) to 4 (“Extremely characteristic or true of me”<\/em>), yielding a total composite score between 0 and 80. Three items (Items 5, 9, and 11) are positively worded and reverse-scored to mitigate acquiescence bias. Across more than three decades of extensive psychometric investigation involving clinical cohorts diagnosed with Social Anxiety Disorder (SAD)<\/a>, subclinical groups, and non-clinical community samples, the SIAS has demonstrated exceptional internal consistency (typically ranging from α = .88 to α = .94), robust test-retest reliability across diverse test intervals<\/span> ($r = .86$<\/span> to $.92$<\/span>), and solid convergent and discriminant validity against parallel constructs of depression, generalized anxiety, and fear of negative evaluation. Confirmatory factor analytic investigations have continually affirmed either a dominant unidimensional construct or a bifactor structure wherein a substantive interaction anxiety factor explains the vast majority of common variance, with an ancillary method factor capturing variance attributable to the reverse-worded items. The SIAS remains an indispensable gold standard in clinical trials, psychopathology research, cognitive-behavioral assessment, and epidemiological screening worldwide.<\/p>\n Social Interaction Anxiety Scale, SIAS, Social Anxiety Disorder, Social Phobia, Psychometrics, Factor Analysis, Cognitive Behavioral Therapy, Assessment of Social Phobia, Scrutiny Fear, Interpersonal Anxiety<\/p>\n The Social Interaction Anxiety Scale was developed by Richard P. Mattick<\/strong> and J. Christopher Clarke<\/strong>. At the time of the scale’s foundational research and initial validation manuscripts, both authors were affiliated with the School of Psychology at the University of New South Wales (UNSW Sydney)<\/a> in Sydney, New South Wales, Australia. Richard P. Mattick subsequently served as Professor of Clinical Psychology and Executive Director of the National Drug and Alcohol Research Centre (NDARC)<\/a> at UNSW Sydney, publishing extensively on anxiety disorders, substance dependence, and psychotherapeutic efficacy. J. Christopher Clarke made foundational contributions to Australian behavior therapy, behavioral medicine, and the empirical conceptualization of human phobic conditioning.<\/p>\n The primary clinical and theoretical impetus for the development of the Social Interaction Anxiety Scale (SIAS) was the realization in the late 1980s that extant psychometric instruments measuring social anxiety<\/a> failed to cleanly distinguish between two clinically and topographically distinct symptom dimensions: performance scrutiny fears versus dynamic social interaction fears. Prior scales, such as Watson and Friend’s (1969) Social Avoidance and Distress (SAD) scale and Fear of Negative Evaluation (FNE) scale, provided valuable indices of distress and cognitive evaluation concerns, but blended disparate situational domains or emphasized cognitive constructs to the exclusion of behavioral and somatic anxiety evoked during live interpersonal exchanges.<\/p>\n Mattick and Clarke recognized that individuals diagnosed with social phobia under the Diagnostic and Statistical Manual of Mental Disorders<\/a> (spanning DSM-III, DSM-III-R, DSM-IV, and DSM-5) frequently exhibit divergent clinical presentations. Some individuals suffer primarily from discrete performance anxiety, which manifests when eating, drinking, writing, or speaking while being actively observed or scrutinized by others. Conversely, other patients present with generalized social phobia characterized by crippling anxiety in reciprocal social encounters\u2014such as maintaining small talk, attending parties, meeting authority figures, expressing personal opinions, initiating romantic conversations, or resolving interpersonal disagreements. The SIAS was designed to capture this latter dimension: the pervasive fear of interacting with others in one-on-one or group settings where verbal exchange, nonverbal communication, and dynamic mutual feedback occur.<\/p>\n In clinical practice, the SIAS serves three interrelated purposes: diagnostic screening, case conceptualization, and treatment outcome monitoring. Clinicians utilize the scale during intake assessments to gauge the severity of interpersonal impairment, establish baseline behavioral inhibition, and screen for Social Anxiety Disorder across community mental health, university counseling, and outpatient psychiatry contexts. Because the scale possesses elevated sensitivity to therapeutic change, it is ubiquitous in randomized controlled trials of Cognitive-Behavioral Therapy (CBT)<\/a>, exposure therapy, social skills training, and pharmacological agents (such as selective serotonin reuptake inhibitors; SSRIs). In research settings, the SIAS enables exact stratification of clinical versus subclinical participants, supports the testing of sophisticated cognitive models of psychopathology, and facilitates cross-cultural comparisons of social reticence and relational functioning.<\/p>\n The psychological construct assessed by the SIAS is social interaction anxiety<\/strong>. Social interaction anxiety represents a persistent, debilitating emotional state characterized by intense subjective distress, cognitive preoccupation with personal inadequacy, autonomic arousal, and behavioral inhibition or avoidance during situations that involve reciprocal interpersonal engagement. Unlike passive scrutiny situations where an individual is observed while performing a task, interaction situations demand continuous, unpredictable processing of incoming social cues, fluid conversational turn-taking, verbal articulation, and nonverbal signaling.<\/p>\n The construct encompasses several distinct yet highly interrelated behavioral, cognitive, and affective facets:<\/p>\n In total, the construct reflects a systemic failure to perceive the interpersonal realm as safe, leading individuals to interpret ambiguity as catastrophic social rejection, which drives vicious cycles of behavioral withdrawal, protective safety behaviors, and chronic loneliness.<\/p>\n The conceptual architecture of the SIAS is rooted in the cognitive-behavioral paradigms of social anxiety articulated by Clark and Wells (1995) and Rapee and Heimberg (1997). These models posit that individuals with social phobia harbor deep-seated, dysfunctional core beliefs regarding themselves (e.g., “I am inherently inept, boring, and socially flawed”) and others (e.g., “Others are critical, highly demanding, and looking for reasons to reject me”).<\/p>\n When an individual entering a social interaction is confronted by the requirement to converse, several interconnected mechanisms trigger acute interaction anxiety:<\/p>\n The construct, convergent, discriminant, and predictive validity of the SIAS have been comprehensively evaluated across clinical, college, and broad community populations spanning numerous cultural contexts.<\/p>\n In their seminal validation investigation, Mattick and Clarke (1998) examined the convergent validity of the SIAS in clinical samples of patients meeting DSM-III-R criteria for social phobia, agoraphobia, and panic disorder, alongside college students and community volunteers. The SIAS correlated substantially and positively with established indices of social anxiety, including the Fear of Negative Evaluation Scale (FNE; correlations ranging from $r = .55$<\/span> to $.71$<\/span>) and Watson and Friend’s Social Avoidance and Distress Scale (SAD; correlations from $r = .65$<\/span> to $.76$<\/span>). In contrast, its companion instrument, the SPS (which assesses scrutiny anxiety during public performance, eating, or writing), correlated with the SIAS at approximately $r = .65$<\/span> to $.72$<\/span>, demonstrating that while both tools capture overlapping variance within the overarching spectrum of social phobia, they retain distinct construct variance corresponding to performance versus interactive domains.<\/p>\n Subsequent psychometric evaluations across varied psychiatric cohorts have corroborated these relationships. Brown, Turovsky, Heimberg, et al. (1997) assessed the SIAS across diverse anxiety disorder populations, confirming that the scale correlates strongly with the Social Phobia and Anxiety Inventory (SPAI)<\/a> Social Phobia Subscale ($r = .78$<\/span>) and moderately with generalized indices of state-trait anxiety measured by the Spielberger State-Trait Anxiety Inventory (STAI-T; $r = .58$<\/span>) and the Beck Depression Inventory (BDI; $r = .43$<\/span> to $.54$<\/span>).<\/p>\n Discriminant validity is exceptionally robust. In comparative trials, patients diagnosed with Social Anxiety Disorder consistently attain dramatically higher mean scores on the SIAS ($M \\approx 34.6$<\/span> to $49.0$, depending on whether the sample has generalized or specific SAD) compared to healthy control participants ($M \\approx 14.0$<\/span> to $19.0$) and clinical cohorts diagnosed with Panic Disorder, Agoraphobia, Generalized Anxiety Disorder (GAD), or Major Depressive Disorder ($M \\approx 20.0$<\/span> to $28.0$). Brown et al. (1997) demonstrated that the SIAS selectively indexes social interaction fears rather than diffuse neuroticism or non-social somatic arousal. Peters (2000) evaluated the discriminant utility of the SIAS, SPS, and SPAI, demonstrating through receiver operating characteristic (ROC) curves that the SIAS effectively distinguishes between individuals with generalized social phobia and other anxiety disorders with high sensitivity and specificity.<\/p>\n The SIAS exhibits profound sensitivity to cognitive-behavioral and pharmacological treatment effects. Multiple clinical outcome studies (Mattick & Peters, 1988; Mattick, Peters, & Clarke, 1989; Heimberg et al., 1998) have recorded large pre-to-post treatment effect sizes (Cohen’s $d > 1.0$<\/span>) on the SIAS following structured CBT involving exposure and cognitive restructuring. Additionally, reductions in SIAS scores during therapy reliably predict long-term decreases in functional impairment and improvements in interpersonal relationship satisfaction and occupational functioning at 6- and 12-month follow-up evaluations.<\/p>\n The psychometric reliability of the SIAS has been replicated across thousands of clinical and non-clinical participants worldwide.<\/p>\n In their original validation report, Mattick and Clarke (1998) reported high internal consistency for the 20-item instrument, documenting a Cronbach’s alpha of $\\alpha = .94$<\/span> in a cohort of 243 patients diagnosed with DSM-defined social phobia, $\\alpha = .93$<\/span> in a sample of agoraphobic patients ($N = 65$<\/span>), and $\\alpha = .88$<\/span> among undergraduate students ($N = 225$<\/span>). Subsequent comprehensive studies have reported near-identical figures:<\/p>\n Temporal stability of the SIAS is high over diverse retest intervals, confirming that it measures a relatively stable clinical trait rather than transient state-dependent fluctuations. Mattick and Clarke (1998) reported a test-retest correlation coefficient of $r = .92$<\/span> across a 1-month interval in a sample of non-treatment-seeking participants, and $r = .92$<\/span> over a 3-week interval in a clinical waiting-list control sample. Osman et al. (1998) confirmed stability over a 4-week window in college students ($r = .86$<\/span>). In clinical intervention trials, untreated wait-list control participants demonstrate stable scores across intervals ranging from 8 to 12 weeks ($r = .79$<\/span> to $.89$<\/span>), verifying that natural remission without intervention is minimal and that instrument drift is negligible.<\/p>\n The latent dimensionality of the SIAS has been the subject of extensive psychometric investigation, sparking a lively and instructive debate in clinical psychological assessment.<\/p>\n Mattick and Clarke (1998) conducted exploratory factor analysis (EFA) on their clinical and student calibration samples. Utilizing principal axis factoring followed by orthogonal and oblique rotations, they determined that the scale was essentially unidimensional. A single massive interaction anxiety factor accounted for the primary share of total variance, with all 20 items loading substantially (typically above $.40$<\/span>) onto this dominant construct. Consequently, the authors recommended summing all 20 items into a single composite index of social interaction anxiety.<\/p>\n Subsequent confirmatory factor analytic (CFA) investigations, notably by Osman, Gutierrez, Barrios, Kopper, and Chiros (1998), questioned this single-factor solution. Osman et al. identified a 3-factor or 2-factor structure, wherein the three reverse-scored items (Item 5: “I find it easy to make friends my own age”<\/em>; Item 9: “I am at ease meeting people at parties, etc.”<\/em>; and Item 11: “I find it easy to think of things to talk about”<\/em>) consistently clustered together to form an isolated secondary factor, often labeled “Social Facilitation” or “Social Ease.”<\/p>\n This finding prompted rigorous methodological scrutiny by psychometric experts, most notably Rodebaugh, Woods, and Heimberg (2007) and Rodebaugh et al. (2004). Employing structural equation modeling (SEM) and item response theory (IRT), these authors demonstrated that the secondary factor formed by Items 5, 9, and 11 is not a distinct clinical construct; rather, it represents a well-documented methodological artifact<\/strong> resulting from reverse-worded item phrasing (method effect). When CFAs are specified as a bifactor model<\/em>\u2014featuring one general substantive factor (Social Interaction Anxiety) and one orthogonal method factor capturing the shared method variance of the reverse-scored items\u2014the model yields fit indices ($CFI > .95$<\/span>, $TLI > .95$<\/span>, $RMSEA < .06$). Furthermore, IRT analyses revealed that Items 5, 9, and 11 possess lower discrimination parameters and poorer measurement precision across lower and moderate levels of trait anxiety compared to the direct, straightforwardly phrased items.<\/p>\n Due to the method artifact introduced by the reverse-scored items and the clinical demand for brief screening instruments in high-volume settings, Fergus, Valentiner, McGrath, et al. (2012) and Peters, Sunderland, Andrews, et al. (2012) applied advanced IRT and CFA techniques to extract refined short forms. Peters et al. (2012) established the SIAS-6<\/strong>, a 6-item version that eliminates the reverse-scored items entirely and selects the most psychometrically discriminative straightforward items (Items 1, 2, 8, 12, 16, and 19). Studies by Le Blanc et al. (2014) confirmed that the SIAS-6 correlates at $r > .90$<\/span> with the full 20-item scale, replicates its diagnostic accuracy, and presents a pure unidimensional factor structure without method noise.<\/p>\n The Social Interaction Anxiety Scale was developed by Richard P. Mattick and J. Christopher Clarke. The scale items were initially circulated in an unpublished research manuscript in 1989<\/strong> and were published along with their definitive peer-reviewed validation data in the journal Behaviour Research and Therapy<\/em> in 1998<\/strong> (Mattick & Clarke, 1998).<\/p>\n In alignment with the scientific intent of its authors, the SIAS was placed into the public domain for research and clinical practice. It may be administered, scored, and reproduced without royalty fees or formal permission for non-commercial scientific research, clinical diagnostic screening, and personal assessment, provided that proper bibliographic credit is cited in publications and academic presentations. Commercial licensing, distribution in fee-for-service enterprise software platforms, or third-party digital health applications typically requires consultation of copyright terms held by the original publisher (Elsevier Ltd. \/ Behaviour Research and Therapy<\/em>) or direct contact with the primary author.<\/p>\n Instructions:<\/p>\n For each question, please indicate a score to indicate how much the following statements are characteristic or true of you:<\/p>\nKeywords<\/h2>\n
Authors<\/h2>\n
Purpose<\/h2>\n
Psychological Construct<\/h2>\n
\n
Theoretical Framework<\/h2>\n
\n
Validity<\/h2>\n
Construct and Convergent Validity<\/h3>\n
Discriminant Validity<\/h3>\n
Predictive and Treatment Validity<\/h3>\n
Reliability<\/h2>\n
Internal Consistency<\/h3>\n
\n
Test-Retest Stability<\/h3>\n
Factor Analysis<\/h2>\n
The Original Unidimensional Model<\/h3>\n
Method Effects of Reverse-Scored Items<\/h3>\n
Short-Form Derivatives: The SIAS-6<\/h3>\n
Instrument \/ Measurement Tool<\/h2>\n
\n
\n
\n
\n
\n
Permissions & Fee and Test Year<\/h2>\n
References<\/h2>\n
\n
Items of the Scale<\/h2>\n
\n 0<\/strong> = Not at all characteristic or true of me<\/span>
\n 1<\/strong> = Slightly characteristic or true of me<\/span>
\n 2<\/strong> = Moderately characteristic or true of me<\/span>
\n 3<\/strong> = Very characteristic or true of me<\/span>
\n 4<\/strong> = Extremely characteristic or true of me<\/span>\n <\/div>\n\n