Anxiety (ANX)

1. Abstract

The Anxiety scale (ANX) represents one of the three core subscales of the Depression Anxiety Stress Scales (DASS), developed by Peter F. Lovibond and Sydney H. Lovibond in 1995 at the University of New South Wales. Designed to address the long-standing psychometric challenge of diagnostic overlap and symptom conflation between affective and anxiety disorders, the ANX subscale specifically isolates the unique dimension of autonomic arousal, skeletal muscle reactivity, situational apprehension, and the subjective experience of anxious affect. The full version comprises 14 items (with a psychometrically equivalent 7-item short form in the DASS-21), scored on a 4-point Likert-type scale ranging from 0 (“Did not apply to me at all”) to 3 (“Applied to me very much, or most of the time”), evaluating symptom frequency and severity over the past week.

Extensive psychometric investigations confirm that the ANX scale demonstrates outstanding internal consistency, with Cronbach’s alpha typically ranging between .84 and .92 in non-clinical cohorts and exceeding .90 in clinical populations. Confirmatory factor analyses repeatedly validate its empirical distinctiveness from the Depression (DEP) and Stress (STR) subscales, demonstrating superior discriminant validity when compared to classic legacy instruments such as the Beck Anxiety Inventory (BAI) and the State-Trait Anxiety Inventory (STAI). The scale strongly aligns with the physiological hyperarousal component of Clark and Watson’s Tripartite Model of Anxiety and Depression. Widely translated into over 50 languages and normalized across diverse international demographics, the DASS-ANX subscale serves as a foundational instrument for psychiatric epidemiology, clinical psychology assessment, neuropsychological research, and longitudinal treatment-monitoring programs worldwide.

2. Keywords

Anxiety Scale, DASS-ANX, Depression Anxiety Stress Scales, Autonomic Arousal, Psychometrics, Physiological Hyperarousal, Tripartite Model, Construct Validity, Confirmatory Factor Analysis, Internal Consistency, Clinical Assessment, Neuroticism

3. Authors

The Anxiety scale (ANX) of the Depression Anxiety Stress Scales was developed by:

• Peter F. Lovibond, Ph.D. — Professor of Psychology, School of Psychology, University of New South Wales (UNSW), Sydney, Australia. Specialist in human associative learning, conditioning, anxiety disorders, and cognitive psychopathology.
• Sydney H. Lovibond, Ph.D. (1924–2010) — Late Emeritus Professor of Psychology, School of Psychology, University of New South Wales, Sydney, Australia. Renowned clinical psychologist, psychometrician, and behavioral scientist who pioneered behavioral medicine and empirical assessment paradigms across Australasia.

Official correspondence regarding the instrument, normative tables, and permissions has historically been administered via the School of Psychology at the University of New South Wales and the Psychology Foundation of Australia.

4. Purpose

The primary purpose of the Anxiety (ANX) scale is to provide a pure, dimensionally focused psychometric index of anxiety that is empirically separated from depressive mood and generalized tension or stress. Historically, self-report inventories measuring emotional distress—such as early iterations of the Beck Depression Inventory (BDI), the Minnesota Multiphasic Personality Inventory (MMPI) mood scales, and the Taylor Manifest Anxiety Scale—suffered from extensive construct contamination. Most self-report anxiety scales correlated so heavily with depression inventories (often r = .60 to .80) that clinicians and clinical researchers were unable to determine whether an intervention specifically alleviated anxiety, ameliorated dysphoric mood, or merely reduced an undifferentiated general factor of psychological distress or neuroticism.

The Lovibonds initiated a long-term research program aimed at mapping the precise empirical boundaries of non-specific negative affectivity versus the specific core symptoms unique to anxiety and depression. Through iterative empirical item analysis with non-clinical university students and clinical patient cohorts, they discovered that anxiety, in its purest differentiating manifestation, is defined by physiological hyperarousal, autonomic hyperactivity (e.g., tachycardia, diaphoresis, dyspnea, dry mouth), skeletal musculature trembling, and immediate situational panic-like terror. The purpose of the ANX scale is therefore twofold:

• Clinical Diagnostic Screening and Symptom Tracking: The scale enables mental health clinicians, psychiatrists, and allied health professionals to rapidly screen for the severity of panic-like and somatic anxiety symptomatology. Because the scale isolates physiological arousal from pervasive cognitive worry (which Lovibond and Lovibond identified as loading onto the distinct Stress scale), it provides unique predictive insight into clinical conditions characterized by sympathetic nervous system hyperreactivity, such as panic disorder, agoraphobia, specific phobias, and somatic symptom disorders. Furthermore, administered across sequential therapy sessions, it acts as a sensitive metric of symptom reduction following pharmacological interventions (e.g., SSRIs, SNRIs, beta-blockers) or cognitive-behavioral therapies (e.g., interoceptive exposure, applied relaxation, biofeedback).
• Empirical Psychopathology and Epidemiological Research: In research contexts, the ANX subscale allows clinical scientists to measure the unique variance of anxiety within structural equation models, experimental paradigms, and epidemiological surveys without introducing confounding overlap with dysphoric anhedonia or cognitive tension. It enables structural psychometric investigation into the shared versus unique genetic, neurological, and behavioral pathways governing internalizing disorders.

5. Psychological Construct

The construct assessed by the Anxiety (ANX) scale is multi-faceted yet strictly delimited, operationalizing anxiety as an acute emotional response marked by autonomic activation, subjective apprehension, and physical somatic sensations. Rather than measuring a heterogeneous composite of diffuse worries, everyday life hassles, or existential demoralization, the ANX scale measures four explicit sub-dimensions of the anxiety construct:

Autonomic Arousal and Sympathetic Activation

The predominant core of the ANX subscale reflects the peripheral physiological manifestations of the sympathetic nervous system’s “fight-or-flight” cascade. This includes items assessing cardiovascular hyperreactivity (e.g., conscious awareness of heart pounding, palpitations, or tachycardia in the absence of physical exertion), respiratory distress (e.g., shortness of breath, hyperventilation, difficulty catching one’s breath without physical cause), and secretory or thermal dysregulation (e.g., dryness of the mouth, sudden perspiration, cold or clammy extremities). For instance, a patient experiencing autonomic surge during an uncued panic episode rapidly endorses severe manifestations on these somatic markers.

Skeletal Musculature and Motor Tension Effects

A second central dimension focuses on overt musculoskeletal reactions induced by neuroendocrine discharge (e.g., epinephrine and norepinephrine signaling). This includes involuntary tremors, shaking hands, postural instability, and physical shakiness. Unlike generalized muscle tightness (which characterizes the DASS Stress subscale and generalized anxiety disorder criteria), the skeletal items of the ANX subscale capture visible, destabilizing somatic tremor that interferes with motor stability and signals immediate neurovegetative arousal.

Situational and Phobic Apprehension

The ANX scale captures the anticipatory dread and situational panic that emerges when an individual encounters or anticipates situations that could precipitate acute fear or social embarrassment. It measures the anticipatory sense that panic or incapacitation is imminent (e.g., fear of panicking and making a fool of oneself in public or interpersonal contexts). This symptom connects the subjective cognitive fear of loss of physical or behavioral control directly to interoceptive cues.

Subjective Felt Experience of Anxious Affect

Finally, the scale measures subjective terror, panic-like apprehension, faintness, and dread that occur independently of objective external threats. This includes descriptions of experiencing acute episodes of unprovoked fear, feeling “scared for no good reason,” feeling close to panic, or sensing bodily disequilibrium (such as dizziness, faintness, or feeling on the verge of physical collapse). Together, these facets capture the physiological, subjective, and interoceptive constellation that clinical psychopathology terms acute somatic anxiety.

6. Theoretical Framework

The theoretical architecture underpinning the Anxiety (ANX) scale is grounded in the intersection of cognitive-behavioral theory, neurobiological paradigms of fear conditioning, and quantitative dimensional models of psychopathology. Most notably, the development of the scale was directly informed by, and serves as one of the primary empirical pillars for, the Tripartite Model of Anxiety and Depression formulated by Lee Anna Clark and David Watson (1991).

The Tripartite Model Integration

Clark and Watson proposed that the pervasive diagnostic comorbidity and high psychometric correlation between depressive and anxious syndromes could be mathematically explained by decomposing internalizing psychopathology into three distinct dimensions:

1. General Distress / Negative Affectivity (NA): A shared, non-specific temperamental vulnerability factor involving general demoralization, irritability, emotional instability, and distress, which is elevated in both depression and anxiety.
2. Anhedonia / Low Positive Affect (PA): A specific factor uniquely characteristic of depressive disorders, operationalized by the loss of pleasure, lack of positive emotional engagement, fatigue, and psychomotor blunting.
3. Physiological Hyperarousal (PH): A specific factor uniquely characteristic of anxiety disorders, manifested as autonomic reactivity, somatic trembling, panic sensations, and sympathetic nervous system surge.

Lovibond and Lovibond’s empirical findings (1995) verified that most established anxiety instruments—such as the Beck Anxiety Inventory (BAI) and the State-Trait Anxiety Inventory (STAI)—were saturated either with general negative affect or with heterogeneous mixtures of worry and dysphoria. The DASS ANX subscale was engineered to be the purest self-report operationalization of the Physiological Hyperarousal (PH) dimension. While the DASS Depression (DEP) subscale maps onto Low Positive Affect / Anhedonia, and the DASS Stress (STR) subscale maps onto chronic tension and high Negative Affectivity, the ANX subscale specifically targets autonomic nervous system dysregulation.

Neurobiological and Cognitive Dimensions

From a neurobiological standpoint, the ANX scale reflects threshold sensitivities within the hypothalamic-pituitary-adrenal (HPA) axis and the central amygdalar fear circuitry. When an individual exhibits elevated ANX scores, it indexes a hyper-sensitized neurovisceral loop: amygdala efferents stimulate the locus coeruleus and paraventricular nucleus, generating rapid somatic feedback (tachycardia, tachypnea, peripheral vasoconstriction) that is catastrophically misinterpreted via interoceptive cognitive appraisal mechanisms (as conceptualized in David M. Clark’s cognitive model of panic). Thus, the Lovibond ANX scale encapsulates both the primitive somatic output of fear neurocircuitry and the subjective dread that accompanies acute interoceptive surges.

7. Validity

The construct, convergent, discriminant, and criterion-related validity of the Anxiety (ANX) scale has been substantiated through hundreds of independent empirical investigations worldwide across non-clinical, student, psychiatric outpatient, and community populations.

Convergent Validity

The ANX subscale demonstrates high, statistically significant convergent correlations with established legacy measures of anxiety:

• Beck Anxiety Inventory (BAI): In the seminal Lovibond & Lovibond (1995) validation study (N = 717), the ANX scale correlated strongly with the BAI (r = .81), demonstrating that both instruments tap into somatic and physiological arousal symptoms. Subsequent clinical validations (e.g., Antony et al., 1998; Brown et al., 1997) reported convergent correlations ranging from r = .78 to .85 in clinical populations diagnosed with panic disorder, generalized anxiety disorder, and social anxiety disorder.
• State-Trait Anxiety Inventory (STAI-T): Moderate to high correlations are consistently observed with the STAI Trait scale (typically r = .65 to .75). The correlation is slightly attenuated compared to the BAI because the STAI contains substantial variance attributable to general negative affectivity rather than pure somatic hyperarousal.
• Hamilton Anxiety Rating Scale (HAM-A): In psychiatric outpatient samples, clinician-rated HAM-A scores exhibit strong concordance with the self-report ANX scale (r = .68 to .76), confirming its utility as a patient-reported outcome measure reflecting clinician-observable anxiety pathology.

Discriminant Validity

The primary psychometric triumph of the DASS ANX scale lies in its exceptional discriminant validity:

• Beck Depression Inventory (BDI / BDI-II): While the BAI typically exhibits high cross-correlations with the BDI (often r = .60 or higher), the DASS ANX scale demonstrates significantly lower correlations with the BDI (r = .42 to .54), confirming that it does not unnecessarily tap into depressive cognitive content or anhedonia.
• DASS Subscale Divergence: Within the instrument itself, confirmatory factor analyses show that the ANX scale correlates moderately with the DASS Depression scale (r ≈ .45 to .58) and moderately-to-highly with the DASS Stress scale (r ≈ .65 to .72), yet remains statistically distinct in non-clinical and psychiatric cohorts. The latent correlations between factors confirm that while all three share higher-order negative affect, their unique variances do not collapse into a single unidimensional construct.

Criterion and Predictive Validity

Criterion validity is evidenced by the scale’s capacity to differentiate between distinct psychiatric diagnostic categories established by structured clinical interviews (such as the SCID). Patients meeting DSM-IV and DSM-5 diagnostic criteria for Panic Disorder, Agoraphobia, and Specific Phobias score significantly higher on the DASS-ANX subscale than patients diagnosed with Major Depressive Disorder (MDD) or Generalized Anxiety Disorder (GAD), whose scores are instead characterized by elevations on the DEP and STR scales, respectively (Brown et al., 1997). In longitudinal clinical trials, reductions in ANX scores systematically predict reductions in autonomic panic frequency during cognitive-behavioral exposure therapies.

8. Reliability

The reliability of the Anxiety (ANX) scale has been documented across dozens of psychometric investigations, proving exceptionally robust across varied demographic, cultural, and linguistic cohorts.

Internal Consistency

Internal consistency estimates using Cronbach’s alpha ($lpha$) and McDonald’s omega ($\omega$) consistently fall well above the recommended academic threshold of .80 for basic research and .90 for clinical decision-making:

• Original Normative Sample (Lovibond & Lovibond, 1995): In a normative university cohort of N = 2,914 non-clinical adults, the 14-item ANX scale achieved a Cronbach’s alpha of $lpha = .84$.
• British Community Cohort (Crawford & Henry, 2003): In a large, normative adult sample representative of the general population (N = 1,794), Crawford and Henry reported an alpha of $lpha = .89$ for the 14-item ANX subscale, and $lpha = .82$ for the abbreviated 7-item DASS-21 version.
• Clinical Outpatient Cohorts (Antony et al., 1998; Brown et al., 1997): Across psychiatric cohorts presenting with anxiety and mood disorders, the 14-item scale yielded alpha values between $lpha = .89$ and $lpha = .92$, illustrating superior reliability even in populations with extreme distress ranges.
• Composite Reliability and McDonald’s Omega: Contemporary structural equation modeling studies evaluating McDonald’s categorical omega total ($\omega_t$) report coefficients typically spanning .88 to .93, confirming that variance is largely attributable to the target latent construct rather than random measurement error.

Test-Retest Stability

Because the DASS ANX scale evaluates symptom state severity over the preceding week (“over the past week”), test-retest reliability reflects both temporal stability of the latent tendency and legitimate responsiveness to situational change:

• Over a two-week interval in untreated non-clinical cohorts, test-retest reliability estimates range between $r_{tt} = .75$ and $r_{tt} = .83$, indicating adequate short-term stability without ceiling rigidity.
• Over a four-to-eight-week interval, correlation coefficients typically sit between $r_{tt} = .65$ and $r_{tt} = .72$, capturing natural fluctuations in life stressors and autonomic reactivity.

Standard Error of Measurement

The Standard Error of Measurement (SEM) for the 14-item ANX scale is generally small (SEM ≈ 1.8 to 2.3 score points on the 0–42 raw metric), indicating high precision around individual observed scores. This low measurement error provides clinicians with tight confidence intervals (typically $\pm 4$ points at the 95% level) when evaluating whether an individual’s score crosses clinical cutoffs.

9. Factor Analysis

The structural dimensionality of the DASS, and specifically the Anxiety (ANX) scale, has been one of the most rigorously debated and replicated areas in modern psychometric literature.

Exploratory Factor Analysis (EFA)

In the original scale construction phases, Lovibond and Lovibond (1995) subjected large item pools representing affective distress to principal components analysis and maximum-likelihood exploratory factor analyses with oblique (oblimin) rotation. The items specifically written to capture autonomic arousal, trembling, situational panic, and interoceptive dread cleanly loaded onto a discrete factor (eigenvalues substantially greater than 1.0), with primary item factor loadings typically exceeding .50, while cross-loadings on the Depression and Stress factors were virtually all below .25. This clean separation formed the 14 items comprising the final ANX scale.

Confirmatory Factor Analysis (CFA)

Extensive CFA studies have consistently tested competitive models to determine whether negative affect is better represented as a unidimensional construct, a two-factor construct (anxiety vs. depression), a three-factor correlated model, or a quadripartite bifactor model:

• Crawford & Henry (2003): Conducted rigorous CFA on a normative community sample (N = 1,794) comparing multiple models. The three-factor correlated model (Depression, Anxiety, Stress) demonstrated vastly superior goodness-of-fit compared to one-factor or two-factor alternatives. Fit indices for the three-factor model met strict psychometric standards: Comparative Fit Index ($ ext{CFI} = .93$), Tucker-Lewis Index ($ ext{TLI} = .92$), and Root Mean Square Error of Approximation ($ ext{RMSEA} = .046$; 90% CI [.044, .048]).
• Antony et al. (1998): Validated the three-factor model in a clinical sample of N = 437 outpatients. The CFA confirmed that ANX items loaded robustly onto their designated latent factor, with standardized factor loadings ($lambda$) ranging from .52 to .78 ($p < .001$).
• Bifactor Modeling: In recent years, structural psychometricians (e.g., Henry & Crawford, 2005; Osman et al., 2012) have applied bifactor CFA models to examine whether the ANX subscale retains sufficient unique specific variance after extracting a general ‘p-factor’ (general distress or negative affectivity). The bifactor models show that while the general factor accounts for roughly 50–60% of common variance, the ANX subscale retains substantial and psychometrically interpretable specific variance (omega subscale $\omega_s$ between .30 and .45), primarily driven by the somatic and autonomic arousal items. This solidifies the ANX scale as a psychometrically sound, unique dimensional marker of physiological hyperarousal.

10. Instrument / Measurement Tool

The Anxiety (ANX) scale can be administered as an independent stand-alone scale or as part of the complete Depression Anxiety Stress Scales (either the full 42-item instrument or the 21-item short form). Below is the comprehensive technical profile of the measurement tool:

• Test Type: Standardized psychological self-report symptom rating scale.
• Administration Format: Paper-and-pencil, computer-administered, or web-based psychological assessment.
• Target Population: Adolescents (aged 14+) and adults. (A specialized DASS-Y version exists for younger children/youth).
• Completion Time:
  • Full 14-item ANX scale: Approximately 3 to 5 minutes.
  • 7-item short-form ANX scale (from DASS-21): Approximately 1.5 to 2 minutes.
• Item Count:
  • DASS-42 Version: 14 specific items (Items 2, 4, 7, 9, 15, 19, 20, 23, 25, 28, 30, 36, 40, and 41 of the full DASS).
  • DASS-21 Version: 7 specific items (Items 2, 4, 7, 9, 15, 19, and 20 of the short form).
• Response Format: 4-point Likert-type severity/frequency rating system evaluating the participant’s state over the previous 7 days:
  • 0 = Did not apply to me at all (Never)
  • 1 = Applied to me to some degree, or some of the time (Sometimes)
  • 2 = Applied to me to a considerable degree, or a good part of time (Often)
  • 3 = Applied to me very much, or most of the time (Almost Always)
• Scoring and Transformation Rules:
  • DASS-42 Scoring: Sum the raw numerical scores (0 to 3) for all 14 anxiety items. The total raw score ranges from 0 to 42.
  • DASS-21 Scoring: Sum the raw numerical scores (0 to 3) for the 7 anxiety items (range 0 to 21), then multiply by 2 to convert the short-form score into the standard 0–42 DASS-42 equivalent score.
• Severity Categorization and Clinical Cutoff Norms (Full Scale Equivalent Metric 0–42):
  • Normal: Raw score 0 – 7 (Percentile: 0 – 78%)
  • Mild: Raw score 8 – 9 (Percentile: 79 – 86%)
  • Moderate: Raw score 10 – 14 (Percentile: 87 – 94%)
  • Severe: Raw score 15 – 19 (Percentile: 95 – 97%)
  • Extremely Severe: Raw score 20+ (Percentile: 98 – 100%)

11. Permissions & Fee and Test Year

The Depression Anxiety Stress Scales, including the Anxiety (ANX) subscale, were officially published in 1995 by Peter F. Lovibond and Sydney H. Lovibond following nearly a decade of psychometric piloting and scale refinement.

• Copyright and Ownership: The copyright for the DASS instrument is held by the Psychology Foundation of Australia.
• Permissions and Open Access Status: Unlike proprietary commercial psychological tests published by corporate test houses (such as the BDI-II, BAI, or MMPI), the DASS was intentionally placed in the public domain for clinical and research purposes by its authors to advance open science and psychological assessment accessibility.
• User Fees: The scale is completely free of charge for non-commercial academic research, epidemiological studies, university teaching, and clinical practice. No formal individual user licensing fees or per-administration royalty payments are required.
• Commercial and Digital Use Terms: Any commercial deployment, inclusion in proprietary health software applications, digital electronic medical record (EMR) integrations, or re-publishing for commercial sale requires prior written permission from the Psychology Foundation of Australia / School of Psychology at UNSW. Any authorized reproduction must retain the full academic citation and copyright notice acknowledging P. F. Lovibond and S. H. Lovibond.

12. References

Below is a curated selection of foundational literature and psychometric validation studies on the Anxiety (ANX) scale formatted in APA 7th edition:

• Antony, M. M., Bieling, P. J., Cox, B. J., Enns, M. W., & Swinson, R. P. (1998). Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychological Assessment, 10(2), 176–181. https://doi.org/10.1037/1040-3590.10.2.176
• Brown, T. A., Chorpita, B. F., Korotitsch, W., & Barlow, D. H. (1997). Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behaviour Research and Therapy, 35(1), 79–89. https://doi.org/10.1016/S0005-7967(96)00068-X
• Clark, L. A., & Watson, D. (1991). Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology, 100(3), 316–336. https://doi.org/10.1037/0021-843X.100.3.316
• Crawford, J. R., & Henry, J. D. (2003). The Depression Anxiety Stress Scales (DASS): Normative data and latent structure in a large non-clinical sample. British Journal of Clinical Psychology, 42(2), 111–131. https://doi.org/10.1348/014466503321903544
• Henry, J. D., & Crawford, J. R. (2005). The short-form version of the Depression Anxiety Stress Scales (DASS-21): Construct validity and normative data in a large non-clinical sample. British Journal of Clinical Psychology, 44(2), 227–239. https://doi.org/10.1348/014466505X29657
• Lovibond, P. F., & Lovibond, S. H. (1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33(3), 335–343. https://doi.org/10.1016/0005-7967(94)00075-U
• Lovibond, S. H., & Lovibond, P. F. (1995). Manual for the Depression Anxiety Stress Scales (2nd ed.). Psychology Foundation of Australia.
• Osman, A., Wong, J. L., Bagge, C. L., Feldner, M., Yancey, J. R., & Campbell, A. (2012). The Depression Anxiety Stress Scales-21 (DASS-21): Further examination of dimensions, scale reliability, and correlates. Journal of Clinical Psychology, 68(12), 1322–1338. https://doi.org/10.1002/jclp.21908
• Page, A. C., Hooke, G. R., & Morrison, D. L. (2007). Psychometric properties of the Depression Anxiety Stress Scales (DASS) in depressed clinical samples. British Journal of Clinical Psychology, 46(3), 283–297. https://doi.org/10.1348/014466506X158996

13. Items of the Scale