Penn State Worry Questionnaire (PSWQ)

1. Abstract

The Penn State Worry Questionnaire (PSWQ) is widely recognized as the gold standard self-report instrument designed to quantify the generality, excessive nature, and uncontrollability of pathological worry. Developed by Thomas J. Borkovec and colleagues in 1990, the scale isolates worry as a distinctive cognitive phenomenon, differentiating it from generalized forms of anxiety, panic, and somatic tension. Comprising 16 items evaluated on a 5-point Likert scale ranging from 1 (“Not at all typical of me”) to 5 (“Very typical of me”), the PSWQ yields a total score ranging from 16 to 80. Psychometrically, the instrument exhibits exceptional internal consistency, with Cronbach’s alpha typically ranging between .86 and .95 across clinical, undergraduate, and community samples. It also demonstrates high temporal stability over intervals extending from several weeks to one month (test-retest correlations between .74 and .93). Construct validation confirms robust convergent validity with diagnoses of Generalized Anxiety Disorder (GAD) as outlined in the DSM-5, while displaying solid discriminant validity against somatic complaints and major depressive episodes. Factor analytic research predominantly identifies a unidimensional substantive factor representing pathological worry, alongside a secondary method effect factor driven by five reverse-scored items. The questionnaire is integral to epidemiological research, psychiatric screening, and empirical evaluations of cognitive behavioral therapy interventions targeting cognitive perseveration and worry severity.

2. Keywords

Penn State Worry Questionnaire, PSWQ, pathological worry, Generalized Anxiety Disorder, cognitive perseveration, psychometrics, Borkovec, anxiety assessment, factor analysis, self-report scale

3. Authors

The Penn State Worry Questionnaire was designed and empirically evaluated by a team of clinical psychological researchers at the Pennsylvania State University and affiliated academic research centers:

• Thomas J. Borkovec, Ph.D.: Professor Emeritus of Psychology, Department of Psychology, The Pennsylvania State University, University Park, PA, USA. Dr. Borkovec is a pioneering authority on the nature, function, and cognitive-behavioral treatment of generalized anxiety and pathological worry.
• Terry J. Meyer, Ph.D.: Department of Psychology, The Pennsylvania State University, University Park, PA, USA. Primary investigator involved in initial psychometric validation and item pool development.
• Michael L. Miller, Ph.D.: Department of Psychology, The Pennsylvania State University, University Park, PA, USA. Co-investigator on cognitive avoidance models and scale construction.
• Richard L. Metzger, Ph.D.: Department of Psychology, Stevenson University (formerly Villa Julie College) / The Pennsylvania State University. Psychometrician and cognitive psychology researcher.

4. Purpose

The primary clinical and research purpose of the Penn State Worry Questionnaire is to assess the trait dimension of worry as an independent cognitive symptom cluster, separating cognitive perseverative thought from autonomic hyperactivity, somatic distress, and episodic panic. Pathological worry constitutes the cardinal diagnostic marker of Generalized Anxiety Disorder (GAD) under psychiatric diagnostic classification systems, including the Diagnostic and Statistical Manual of Mental Disorders (DSM-5-TR) and the International Classification of Diseases (ICD-11). Before the publication of the PSWQ in 1990, diagnostic protocols struggled to separate generalized anxiety from general distress, major depression, or panic-related arousal because existing psychometric tools heavily emphasized autonomic symptoms such as diaphoresis, tremors, tachycardia, and muscle tension.

The PSWQ addresses this diagnostic limitation by measuring worry independent of specific situational contexts or trigger contents. Instead of asking what individuals worry about (such as finances, social evaluation, or physical health), the instrument focuses on the structural parameters of the worrying process: its excessiveness, frequency, chronicity, and the perceived inability to terminate it once initiated. By doing so, the scale offers an index of cognitive vulnerability that remains robust across diverse clinical populations, cultural backgrounds, and age brackets.

In clinical practice, the PSWQ is utilized during diagnostic intake as an initial screening instrument for GAD and related anxiety disorders, as a continuous index of symptom severity, and as a treatment-monitoring metric. It is widely used to track symptom changes during pharmacological interventions, mindfulness-based cognitive therapies, and standardized Cognitive Behavioral Therapy (CBT) protocols. A significant reduction in PSWQ scores corresponds closely with clinical remission, reduction in meta-cognitive distress, and decreased relapse rates. In psychopathology research, the tool serves as a reliable benchmark to investigate cognitive avoidance theories, information processing biases, attentional threat monitoring, and emotional regulation deficits.

5. Psychological Construct

The psychological construct evaluated by the Penn State Worry Questionnaire is trait worry, conceptualized as an uncontrollable, excessive, and pervasive sequence of negative cognitive activities characterized by catastrophic verbal-linguistic thoughts. Unlike transient state worry, which occurs as a proportional response to real-world acute stressors and dissipates upon problem resolution, pathological trait worry persists across circumstances, resists cognitive control, and occurs in the absence of imminent danger.

The scale captures three primary clinical features of this construct:

• Excessiveness and Generality: Pathological worry is disproportionate to the actual probability or impact of anticipated negative events. This is assessed via items addressing worry across varying situations (e.g., Item 4: “Many situations make me worry”; Item 9: “As soon as I finish one task, I start to worry about everything else I have to do”; Item 16: “I worry about projects until they are all done”). These items capture the pervasive, diffuse nature of the construct across multiple life domains.
• Uncontrollability and Intrusion: Patients with GAD perceive worry as involuntary and resistant to dismissal. This cognitive intrusion is reflected directly in Item 2 (“My worries overwhelm me”), Item 5 (“I know I should not worry about things, but I just cannot help it”), and Item 14 (“Once I start worrying, I cannot stop”). These items measure the individual’s perceived loss of executive and meta-cognitive control over mental activities.
• Chronicity and Pervasiveness: Pathological worry reflects a continuous, lifelong cognitive habit rather than temporary mood fluctuations. This chronic quality is highlighted by Item 7 (“I am always worrying about something”), Item 12 (“I have been a worrier all my life”), and Item 15 (“I worry all the time”).

The scale also captures the absence of worry resilience through five reverse-scored items that depict adaptive cognitive flexibility and worry termination (e.g., Item 1: “If I do not have enough time to do everything, I do not worry about it”; Item 8: “I find it easy to dismiss worrisome thoughts”; Item 11: “When there is nothing more I can do about a concern, I do not worry about it any more”). A deficit in these adaptive cognitive processes reflects the perseverative nature characteristic of GAD.

6. Theoretical Framework

The PSWQ is grounded in the Cognitive Avoidance Theory of Worry formulated by Thomas D. Borkovec. According to this model, worry is a predominantly verbal-linguistic cognitive activity that functions as an internal avoidance strategy. Borkovec proposed that individuals prone to pathological worry generate verbal-symbolic thoughts about potential future threats to suppress more distressing mental imagery and the uncomfortable somatic, autonomic responses linked to emotional processing.

By engaging in abstract verbal processing rather than concrete perceptual imagery, the individual attenuates intense cardiovascular and visceral fear reactions. However, this dampening comes at a significant cost: it prevents complete emotional exposure and cognitive habituation, thereby reinforcing threat appraisals and perpetuating the worry cycle. The PSWQ operationalizes the core components of this cognitive avoidance architecture by measuring the chronic, continuous maintenance of worry states that shield the individual from primary emotional processing.

Furthermore, the PSWQ aligns with Adrian Wells’s Metacognitive Model of Generalized Anxiety Disorder. Wells posited that severe worry develops through two distinct layers: Type 1 worry (worrying about external events and non-cognitive somatic states) and Type 2 worry (meta-worry, or “worrying about worry”). When individuals hold negative metacognitive beliefs concerning the uncontrollability and dangerousness of their thoughts (e.g., “My worrying could make me lose my mind” or “I cannot stop these thoughts”), worry escalates into a clinical disorder. Items on the PSWQ such as “I know I should not worry about things, but I just cannot help it” directly reflect this perceived loss of control and the metacognitive distress highlighted by Wells.

7. Validity

The validity of the Penn State Worry Questionnaire has been extensively confirmed across diverse samples, including clinical outpatients, primary care populations, and college cohorts.

Construct and Convergent Validity

The PSWQ demonstrates strong convergent validity with related measures of anxiety and negative affectivity. In the original validation studies by Meyer et al. (1990), PSWQ scores correlated substantially with other established psychometric instruments: the State-Trait Anxiety Inventory-Trait version (STAI-T; $r = .64$ to $.79$), the Fear Questionnaire ($r = .36$ to $.45$), and the Beck Depression Inventory (BDI; $r = .36$ to $.62$). Subsequent investigations confirmed correlations between the PSWQ and the Generalized Anxiety Disorder 7-item scale (GAD-7; $r = .70$ to $.82$), supporting its ability to quantify the core cognitive dimensions of generalized anxiety.

Discriminant Validity

A critical achievement of the PSWQ is its ability to differentiate GAD from other psychological conditions. While general trait anxiety scales often overlap with major depressive disorder and panic disorder, the PSWQ demonstrates clean discriminant validity. Clinical studies show that individuals diagnosed with GAD score significantly higher on the PSWQ ($M = 67.2, SD = 9.1$) compared to individuals diagnosed with Panic Disorder ($M = 44.8$), Social Anxiety Disorder ($M = 48.2$), Major Depressive Disorder ($M = 53.1$), and non-clinical controls ($M = 39.4$). These findings confirm that while worry occurs across internalizing disorders, the excessive, chronic, and uncontrollable worry measured by the PSWQ is distinctive to GAD.

Criterion and Predictive Validity

Receiver Operating Characteristic (ROC) analyses demonstrate that the PSWQ shows high sensitivity and specificity in detecting GAD. A clinical cut-off score of 60 to 62 typically yields an optimal balance, providing sensitivity rates between 80% and 85% and specificity rates above 80% when differentiating individuals with GAD from non-anxious and non-clinical populations. In intervention contexts, decreases in PSWQ scores closely track reductions in worry-driven cognitive avoidance and reliably predict long-term clinical recovery following cognitive behavioral therapy.

8. Reliability

The Penn State Worry Questionnaire displays robust reliability metrics across diverse demographic and clinical populations:

Internal Consistency

The scale consistently exhibits excellent internal consistency. In the foundational validation work by Meyer et al. (1990), Cronbach’s alpha values reached .91 in a large sample of 347 college undergraduates and .93 in an independent sample of 344 students. Subsequent clinical investigations with GAD outpatient cohorts reported alpha coefficients between .88 and .95. Systemic psychometric evaluations in community samples, adolescent populations, and geriatric cohorts confirm that internal consistency regularly exceeds the .85 threshold, indicating strong item inter-relatedness.

Test-Retest Stability

The PSWQ demonstrates stable temporal reliability, consistent with its theoretical design as a trait assessment. Meyer et al. (1990) observed a test-retest correlation coefficient of $r = .92$ over an interval of 2 to 4 weeks. Over extended observational periods of 8 to 10 weeks without targeted therapeutic intervention, test-retest reliability estimates remain high, ranging from $r = .74$ to $r = .88$. This stability indicates that the instrument measures a durable cognitive trait rather than temporary emotional fluctuations.

9. Factor Analysis

The internal latent structure of the PSWQ has generated considerable discussion and empirical research within psychometrics. While originally developed as a unidimensional instrument, exploratory and confirmatory factor analyses often reveal a more complex picture.

Unidimensional vs. Two-Factor Models

Exploratory factor analyses (EFA) frequently identify two related factors: a primary factor comprising the 11 positively worded items (reflecting worry excessiveness and uncontrollability) and a secondary factor composed of the 5 negatively worded, reverse-scored items (Items 1, 3, 8, 10, and 11, reflecting the absence of worry or worry resilience). In earlier studies, these two factors were occasionally interpreted as substantive constructs: “Worry Engagement” and “Absence of Worry”.

Method Effects and Bifactor Solutions

Subsequent psychometric investigations using confirmatory factor analysis (CFA) demonstrate that this two-factor split largely reflects a method effect associated with item reverse-scoring rather than substantive theoretical dimensions. Studies utilizing structural equation modeling (SEM) indicate that a bifactor model (or a single substantive factor model with correlated uniquenesses among the reverse-scored items) provides superior model fit over an uncorrelated or correlated two-factor model:

• Standard Confirmatory Factor Model: $\chi^2/df > 3.5$, Root Mean Square Error of Approximation (RMSEA) $\approx .075-.090$, Comparative Fit Index (CFI) $\approx .88-.91$.
• Bifactor / Correlated Uniqueness Model: $\chi^2/df < 2.2$, $\text{RMSEA} \approx .035-.048$, $\text{CFI} \approx .97-.99$, Tucker-Lewis Index (TLI) $\approx .96-.98$.

In these analyses, the general trait factor accounts for the vast majority of the common variance (often exceeding 80–85%), supporting the clinical practice of summing all 16 items into a single composite score.

10. Instrument / Measurement Tool

• Instrument Name: Penn State Worry Questionnaire (PSWQ)
• Target Population: Adults, college students, adolescents, and older adults (an abbreviated PSWQ-A exists for older adults, though the full 16-item version is widely validated across age groups).
• Construct Measured: Trait worry (generality, excessiveness, uncontrollability).
• Item Count: 16 self-report items.
• Response Format: 5-point Likert scale: 1 = Not at all typical of me, 2 = Rarely typical of me, 3 = Somewhat typical of me, 4 = Often typical of me, 5 = Very typical of me.
• Scoring Protocol:
  • Direct-scored items (11 items): Items 2, 4, 5, 6, 7, 9, 12, 13, 14, 15, and 16 are scored from 1 to 5.
  • Reverse-scored items (5 items): Items 1, 3, 8, 10, and 11 are reverse scored ($1 \rightarrow 5$, $2 \rightarrow 4$, $3 \rightarrow 3$, $4 \rightarrow 2$, $5 \rightarrow 1$).
  • Total Score: Sum of all 16 items (possible range: 16 to 80).
• Interpretive Score Categories:
  • 16 – 39: Low Worry (typical of non-anxious community baselines).
  • 40 – 59: Moderate Worry (subclinical elevations, mild anxious apprehension).
  • 60 – 80: High Worry (indicative of pathological worry; typical mean scores for clinical Generalized Anxiety Disorder range between 60 and 68).

11. Permissions & Fee and Test Year

The Penn State Worry Questionnaire was developed in 1990 by Thomas J. Borkovec and his colleagues at the Pennsylvania State University. The original psychometric development was published in Behaviour Research and Therapy.

The instrument is widely available in the public domain for clinical and academic research purposes, requiring no licensing fees or commercial acquisition costs. While standard academic attribution to Meyer, Miller, Metzger, and Borkovec (1990) is required in scientific reports and clinical presentations, the questionnaire can be freely reproduced, digitized, and integrated into clinical workflows. Commercial diagnostic developers wishing to package the instrument within proprietary testing batteries should contact the primary copyright holders and Pennsylvania State University’s technology licensing department.

12. References

• Borkovec, T. D., Ray, S. D., & Stöber, J. (1998). Worry: A cognitive phenomenon intimately linked to affective, physiological, and interpersonal processes. Cognitive Therapy and Research, 22(6), 561–576. https://doi.org/10.1023/A:1018790003416
• Brown, T. A., Antony, M. M., & Barlow, D. H. (1992). Psychometric properties of the Penn State Worry Questionnaire in a clinical anxiety disorders sample. Behaviour Research and Therapy, 30(1), 33–37. https://doi.org/10.1016/0005-7967(92)90093-V
• Fresco, D. M., Mennin, D. S., Heimberg, R. G., & Turk, C. L. (2003). Using the Penn State Worry Questionnaire to identify individuals with generalized anxiety disorder: A receiver operating characteristics analysis. Journal of Behavior Therapy and Experimental Psychiatry, 34(3-4), 283–291. https://doi.org/10.1016/j.jbtep.2003.09.001
• Hazlett-Stevens, H., Ullman, J. B., & Craske, M. G. (2004). Factor analysis of the Penn State Worry Questionnaire: Examination of a method effect. Assessment, 11(4), 361–370. https://doi.org/10.1177/1073191104269872
• Meyer, T. J., Miller, M. L., Metzger, R. L., & Borkovec, T. D. (1990). Development and validation of the Penn State Worry Questionnaire. Behaviour Research and Therapy, 28(6), 487–495. https://doi.org/10.1016/0005-7967(90)90135-6
• Molina, S., & Borkovec, T. D. (1994). The Penn State Worry Questionnaire: Psychometric properties and associated characteristics. In G. C. L. Davey & F. Tallis (Eds.), Worrying: Perspectives on theory, assessment and treatment (pp. 265–283). John Wiley & Sons.
• Wells, A. (2005). The metacognitive model of GAD: Assessment of meta-worry and relationship with DSM-IV generalized anxiety disorder. Cognitive Therapy and Research, 29(1), 107–121. https://doi.org/10.1007/s10608-005-1652-0

13. Items of the Scale