Spence Preschool Anxiety Scale

1. Abstract

The Spence Preschool Anxiety Scale (PAS) is an established, parent-report psychometric instrument designed to assess a comprehensive range of anxiety symptoms in young children aged approximately 2.5 to 6.5 years. Developed by psychometrician Susan H. Spence and colleagues, the scale addresses the critical developmental need for early identification and screening of pediatric internalizing spectrum disorders. Grounded conceptually in the diagnostic nosology of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV and DSM-5), the core PAS comprises 28 core items distributed across five correlated first-order factors: Generalized Anxiety (5 items), Social Anxiety (6 items), Obsessive-Compulsive Disorder symptoms (5 items), Physical Injury Fears (7 items), and Separation Anxiety (5 items). In addition, an optional supplemental traumatic events module consists of a screening item (Item 29) and five subsequent items (Items 30–34) measuring symptoms congruent with Post-Traumatic Stress Disorder (PTSD).

Caregivers evaluate their child’s behavioral and physiological manifestations on a 5-point Likert scale ranging from 0 (Not True at All) to 4 (Very Often True). Across extensive international standardization and validation cohorts, the PAS exhibits robust internal consistency, with total scale Cronbach’s alpha coefficients routinely exceeding α = .88 to .92, and subscale coefficients ranging from α = .66 to .86. Confirmatory factor analyses (CFA) have repeatedly demonstrated superior construct validity for a five-factor correlated model or a hierarchical model featuring five specific factors loading onto a single higher-order anxiety construct. Demonstrating sound convergent validity with concurrent behavioral instruments such as the Child Behavior Checklist (CBCL) and the Behavior Assessment System for Children (BASC), robust discriminant validity against externalizing conditions, and strong test-retest reliability (r ≈ .70–.85), the PAS represents a gold-standard psychometric tool for clinical diagnostics, epidemiological screening, and developmental psychopathology research.

2. Keywords

Spence Preschool Anxiety Scale, preschool anxiety, early childhood assessment, separation anxiety, social anxiety disorder, generalized anxiety disorder, psychometrics, parent-report rating scale, childhood emotional disorders, pediatric internalizing symptoms

3. Authors

The primary authors and investigators responsible for the development, conceptualization, and initial psychometric standardization of the Spence Preschool Anxiety Scale include:

• Susan H. Spence, Ph.D. — Emeritus Professor, School of Applied Psychology and Australian Institute for Suicide Research and Prevention (AISRAP), Griffith University, Mount Gravatt, Queensland, Australia. Dr. Spence is an internationally recognized expert in pediatric clinical psychology and cognitive-behavioral assessment. Contact: [email protected].
• Ronald M. Rapee, Ph.D. — Distinguished Professor, Department of Psychology and Centre for Emotional Health, Macquarie University, Sydney, New South Wales, Australia. Renowned for seminal work on etiology, risk factors, and interventions in early childhood anxiety.
• Carolyn McDonald, B.Psych. (Hons) — Researcher and Clinical Psychologist, School of Psychology, University of Queensland, St. Lucia, Brisbane, Australia.
• Marissa Ingram, B.Psych. (Hons) — Research Associate, School of Psychology, University of Queensland, Brisbane, Australia.

4. Purpose

Anxiety disorders are among the most prevalent psychiatric conditions encountered across the lifespan, frequently initiating during the preschool and early childhood years. Historically, clinical psychometrics neglected children under the age of seven due to persistent psychodynamic and early developmental assumptions suggesting that the cognitive structures of preschoolers were insufficiently developed to harbor complex anxiety pathologies. Empirical research in developmental psychopathology has comprehensively disproven this premise, revealing that clinically significant manifestations of separation distress, phobic avoidance, social inhibition, and pervasive worry emerge robustly prior to primary school entry. Left untreated, early childhood anxiety exhibits notable homotypic and heterotypic continuity, escalating the longitudinal risk for academic impairment, depression, peer victimization, and chronic adult psychopathology.

The primary purpose of the Spence Preschool Anxiety Scale is to provide a developmentally calibrated, psychometrically rigorous, and DSM-congruent parent-report instrument capable of mapping anxiety symptomatology in toddlers and preschool-aged children (ages 2.5 to 6.5 years). Clinically, the PAS fulfills multiple roles:

• Early Detection and Screening: It enables pediatricians, clinical psychologists, and early childhood educators to identify children demonstrating elevated anxiety manifestations exceeding normative developmental milestones.
• Differential Profiling: Through its multi-factor structure, the PAS delineates between specific symptomatic domains—such as distinguishing normative transient fears (e.g., darkness, common animals) from pervasive separation anxiety, generalized worry, or clinically impairing social withdrawal.
• Treatment Formulation and Monitoring: Clinicians utilize the scale to benchmark symptom severity prior to early interventions, such as parent-focused cognitive-behavioral programs (e.g., the Cool Little Kids program), and to evaluate longitudinal treatment response across therapeutic intervals.
• Epidemiological and Longitudinal Research: The instrument facilitates phenotypic characterization within large-scale prospective cohort studies investigating behavioral inhibition, neurobiological vulnerabilities, and parent-child transactional dynamics.

5. Psychological Construct

The Spence Preschool Anxiety Scale evaluates internalizing symptomatology organized across five core psychological dimensions plus a supplementary post-traumatic stress screen. Each construct reflects clinically observable behavioral, cognitive, and somatic markers adapted specifically for the developmental capabilities of young children.

Generalized Anxiety Disorder (GAD)

In preschool children, Generalized Anxiety manifests not through intricate existential rumination, but via pervasive, uncontrollable apprehensive expectations regarding everyday family routines, developmental tasks, and novel events. The child frequently displays autonomic tension, motor restlessness, irritability, and sleep onset difficulties attributable to worry. Items capture difficulty halting anxious thoughts (Item 1), somatic and mood manifestations such as tenseness and irritability (Item 4), sleep disturbances rooted in worry (Item 8), high-frequency daily worrying (Item 14), and an excessive, repetitive need for parental reassurance regarding benign matters (Item 28).

Social Anxiety

The construct of Social Anxiety at this developmental epoch involves pronounced fear, distress, or behavioral avoidance in social evaluative contexts and interactions with both peer groups and unfamiliar adults. Unlike transient shyness, social anxiety in preschoolers causes functional impairment in early learning centers and community settings. Symptoms include cognitive apprehension about social embarrassment or appearing foolish (Items 2 and 19), reluctance or dread of asking authoritative adults for assistance (Item 5), inhibition when encountering unfamiliar individuals (Item 11), expressive freezing or terror during structured public performance tasks like “show and tell” (Item 15), and marked reticence when prompted to initiate peer play (Item 23).

Obsessive-Compulsive Disorder (OCD)

Preschool-aged children commonly demonstrate benign behavioral rituals, such as bedtime routines or lining up toys. However, the OCD construct within the PAS isolates pathological manifestations characterized by compulsive checking, contamination rituals, cognitive rigidity, and magical thinking intended to neutralize perceived danger. The scale measures repetitive verification routines, such as continuously checking taps or closed doors (Item 3), compulsive hand-washing rituals (Item 9), symmetry or exact ordering behaviors designed to avert catastrophic outcomes (Item 18), recurrent intrusive distressing images or thoughts (Item 21), and mental neutralizing rituals (Item 27).

Physical Injury Fears

This subscale operationalizes specific phobic constructs common to early childhood, evaluating whether fear intensities exceed typical developmental parameters and produce functional interference. The construct encompasses evolutionary and environmental threat cues: heights (Item 7), claustrophobic or crowded environments (Item 10), meteorological phenomena such as thunderstorms (Item 13), aquatic immersion/swimming (Item 17), entomological phobias including spiders and insects (Item 20), cynophobia (fear of dogs; Item 24), and nyctophobia (fear of darkness; Item 26).

Separation Anxiety

Constructed around the distress elicited by actual or anticipated physical separation from primary attachment figures, Separation Anxiety in the PAS captures cognitive, emotional, and behavioral protests. Core symptoms include extreme resistance to co-sleeping or sleeping away from the parent (Item 6), catastrophic worries regarding the parent’s health or survival (Item 12), pervasive fears of abduction or getting lost (Item 16), acute behavioral distress at daycare or babysitter drop-offs (Item 22), and separation-themed nightmare parasomnias (Item 25).

Traumatic Stress Symptoms (Post-Traumatic Stress Disorder Module)

Reflecting the unique manifestations of traumatic intrusion and reactivity in early childhood, this conditional module activates only if the caregiver endorses Item 29. It measures intrusive trauma-related nightmares (Item 30), distress provoked by spontaneous recall (Item 31), cue-induced distress (Item 32), traumatic behavioral reenactment or dissociation (Item 33), and physiological hyperarousal upon exposure to reminders (Item 34).

6. Theoretical Framework

The construction of the Spence Preschool Anxiety Scale is situated within modern developmental psychopathology, cognitive-behavioral theories of emotional vulnerability, and attachment frameworks.

Developmental Nosology and DSM Congruence

Historically, developmental theorists debated whether early childhood anxiety reflects a diffuse, undifferentiated emotional distress syndrome or whether distinct diagnostic syndromes exist in early childhood. Spence and colleagues (2001) posited that while anxiety manifestations are influenced by the child’s developmental maturation, the underlying latent phenotypic architecture mirrors the categorical entities described in adult and adolescent nosology (e.g., DSM-IV, DSM-5). By operationalizing behaviors aligned with specific disorders (GAD, SAD, Social Phobia, Specific Phobia, OCD), the PAS tests and validates the construct validity of differentiated anxiety disorders in children under six years of age.

Behavioral Inhibition and Cognitive Diathesis

The scale integrates Jerome Kagan’s construct of Behavioral Inhibition to the Unfamiliar, a temperamental vulnerability characterized by hypervigilance, withdrawal, and autonomic reactivity in response to novel people, objects, and environments. Cognitive-behavioral models (e.g., Rapee & Spence, 2004) further stipulate that preschool children at high risk for anxiety demonstrate early attentional biases toward threatening environmental cues and exaggerated threat-appraisal mechanisms. These vulnerabilities are explicitly tapped by the scale’s items measuring apprehension about future harm, social evaluation, and somatic distress.

Attachment and Transactional Family Systems

Grounded in John Bowlby’s attachment theory, the Separation Anxiety dimension assesses the functional security of the child-caregiver dyad. Insecure, anxious-ambivalent attachment manifests through hyperactivation of the attachment system, where the child experiences intense separation protest and catastrophic cognitions concerning caregiver safety. Transactional models emphasize that parental overprotection or accommodation of anxious avoidance can reinforce and maintain internalizing symptoms, making maternal and paternal report a clinically rich method for capturing the home ecology of anxiety.

7. Validity

The Spence Preschool Anxiety Scale has undergone comprehensive psychometric evaluation across diverse international cohorts, consistently demonstrating exemplary structural, convergent, and discriminant validity.

Construct and Structural Validity

In the seminal psychometric validation study by Spence, Rapee, McDonald, and Ingram (2001), involving a normative community sample of 411 preschool children and their mothers, confirmatory factor analysis affirmed that a correlated five-factor model matched the hypothesized nosological structure significantly better than a unidimensional model. Model fit indices substantiated the multidimensionality: comparative fit index (CFI) values routinely exceed .92 to .95, and root mean square error of approximation (RMSEA) values consistently fall below .05 to .06 across replication studies in Australia, Europe, and Asia.

Convergent and Concurrent Validity

Convergent validity has been established by correlating PAS scores with established psychometric indices of child emotionality:

• Strong positive correlations (r = .55 to .72, p < .001) have been documented between the PAS Total Score and the Internalizing Scale of the Child Behavior Checklist (CBCL 1.5–5), specifically the Anxious/Depressed and Emotionally Reactive subscales.
• The Social Anxiety subscale demonstrates robust correlations with observational measures of behavioral inhibition (r = .48 to .61) and parent-rated shyness scales.
• Separation Anxiety scores show significant positive associations with interview-based clinical diagnoses derived from the Anxiety Disorders Interview Schedule for DSM-IV, Child and Parent Versions (ADIS-P).

Discriminant and Known-Groups Validity

Discriminant validity is supported by modest correlations observed between PAS subscales and the Externalizing Problems scale of the CBCL (r = .18 to .32), demonstrating that the instrument cleanly differentiates internalizing symptoms from oppositional defiant and conduct problems. Furthermore, known-groups validity has been repeatedly demonstrated: children with formally diagnosed anxiety disorders score significantly higher across all PAS subscales than matched non-clinical community controls (F-tests yielding p < .001, partial η² > .30).

8. Reliability

The PAS exhibits exceptional psychometric stability and internal consistency across both non-clinical community populations and referred clinical samples.

Internal Consistency

Extensive normative studies report high internal consistency for the overall instrument and acceptable to strong reliability for individual subscales:

• Total Anxiety Score (Items 1–28): Cronbach’s α coefficients consistently range between .88 and .92.
• Generalized Anxiety: α ranges from .75 to .83.
• Social Anxiety: α ranges from .78 to .85.
• Obsessive-Compulsive Disorder: α ranges from .70 to .80.
• Physical Injury Fears: α ranges from .70 to .79.
• Separation Anxiety: α ranges from .76 to .82.
• PTSD Module (Items 30–34): α ranges from .80 to .88 among trauma-exposed subsets.

Test-Retest Stability

Temporal stability evaluated across non-clinical preschool cohorts demonstrates robust test-retest reliability. Spence et al. (2001) demonstrated an intraclass correlation coefficient (ICC) of r = .79 for the total score across a 12-month retest window, indicating that while minor normative fluctuations occur, the rank-order stability of child anxiety relative to peers remains remarkably stable throughout early childhood. Shorter test-retest intervals (2 to 4 weeks) commonly yield Pearson correlations between r = .82 and .90.

9. Factor Analysis

The latent structure of the PAS has been examined across various cultures using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Confirmatory Factor Analytic Findings

In the original investigation by Spence et al. (2001), CFA was applied to compare competing structural paradigms:

• Model 1 (Single-Factor Model): Posited that all 28 items loaded on a single general anxiety dimension. Fit indices indicated poor convergence (χ²/df > 3.5, CFI = .78, RMSEA = .082).
• Model 2 (Correlated Five-Factor Model): Posited five distinct yet interrelated factors (GAD, Social Anxiety, OCD, Physical Injury Fears, Separation Anxiety). Fit indices revealed marked improvement (χ²/df = 1.62, CFI = .93, RMSEA = .044), verifying the empirical differentiation of symptoms.
• Model 3 (Higher-Order Hierarchical Model): Evaluated the five primary factors loading onto a single second-order “Overall Anxiety” factor. This hierarchical structure demonstrated equivalent statistical fit, confirming that clinicians can interpret both specific subscale scores and a meaningful omnibus total anxiety score.

Factor Loadings

Standardized factor loadings across international CFA replications consistently demonstrate robust item-to-factor alignment, with the vast majority of items loading above .50 on their designated construct:

• Separation Anxiety: Standardized loadings range from .58 to .78, with Item 22 (“Becomes distressed about your leaving him/her…”) typically exhibiting the highest path loading (.78).
• Generalized Anxiety: Loadings range from .54 to .74; Item 14 (“Spends a large part of each day worrying…”) and Item 1 load most strongly.
• Social Anxiety: Loadings range from .55 to .79; Item 15 (“Is afraid of talking in front of the class…”) and Item 19 consistently display dominant loadings.
• Obsessive Compulsive: Loadings range from .48 to .72, with compulsive hand-washing (Item 9) and orderliness (Item 18) emerging as core indicators.
• Physical Injury Fears: Loadings range from .42 to .68; fear of heights (Item 7) and darkness (Item 26) show stable loadings across age cohorts.

10. Instrument / Measurement Tool

• Instrument Type: Caregiver-report behavioral rating scale.
• Target Population: Children aged approximately 2.5 to 6.5 years (preschool and kindergarten).
• Administration Format: Paper-and-pencil questionnaire, clinician-administered interview, or secure web-based electronic survey.
• Completion Time: Approximately 8 to 12 minutes.
• Item Count:
  • 28 core anxiety items (evaluating five diagnostic constructs).
  • 1 trauma filter question (Item 29; dichotomous YES/NO).
  • 5 conditional post-traumatic stress items (Items 30–34, completed only if Item 29 is endorsed YES).
• Response Scale (Core Items 1–28 & 30–34): 5-point Likert-type scale scored as:
  • 0 = Not True at All
  • 1 = Seldom True
  • 2 = Sometimes True
  • 3 = Quite Often True
  • 4 = Very Often True
• Subscale Composition:
  • Generalized Anxiety (5 items): Sum of Items 1, 4, 8, 14, 28 (Theoretical range: 0–20).
  • Social Anxiety (6 items): Sum of Items 2, 5, 11, 15, 19, 23 (Theoretical range: 0–24).
  • Obsessive Compulsive Disorder (5 items): Sum of Items 3, 9, 18, 21, 27 (Theoretical range: 0–20).
  • Physical Injury Fears (7 items): Sum of Items 7, 10, 13, 17, 20, 24, 26 (Theoretical range: 0–28).
  • Separation Anxiety (5 items): Sum of Items 6, 12, 16, 22, 25 (Theoretical range: 0–20).
• Scoring Rules:
  • Total Score: Calculated by summing the scores of Items 1 through 28 (Theoretical range: 0 to 112). Higher scores reflect greater generalized and specific anxiety severity.
  • Trauma Screen / PTSD: Item 29 establishes the presence of life trauma. If affirmed, Items 30 to 34 are summed (Theoretical range: 0–20) to derive an index of acute/post-traumatic stress reactivity.
  • Missing data rule: If more than 2 items are missing from the core 28 items, calculation of the total score is invalid. For single missing items within a subscale, mean imputation based on remaining subscale items is permissible in research contexts.

11. Permissions & Fee and Test Year

The Spence Preschool Anxiety Scale was initially published in 2001 by Susan H. Spence, Ronald M. Rapee, Carolyn McDonald, and Marissa Ingram following developmental piloting beginning in the late 1990s.

Copyright and Usage Policy: In line with Professor Susan H. Spence’s commitment to advancing pediatric mental health screening, the PAS is made freely accessible for non-commercial clinical, research, and educational purposes. Healthcare professionals, academic researchers, and educators may utilize, print, and administer the scale without licensing fees or royalties. The official questionnaires, international translations, automated scoring algorithms, and normative tables are publicly accessible via the dedicated Spence Children’s Anxiety Scale portal (www.scaswebsite.com). Reproduction or commercial redistribution within proprietary for-profit testing batteries requires written permission from the copyright holder.

12. References

Edwards, S. L., Rapee, R. M., & Kennedy, S. (2010). Prediction of anxiety symptoms in preschool-aged children: Examination of maternal and paternal influences. Journal of Child Psychology and Psychiatry, 51(3), 299–307. https://doi.org/10.1111/j.1469-7610.2009.02160.x

Rapee, R. M., & Spence, S. H. (2004). The etiology of social phobia: Empirical evidence and an emotional processing model. Clinical Psychology Review, 24(7), 737–767. https://doi.org/10.1016/j.cpr.2004.06.004

Spence, S. H. (1997). Structure of anxiety symptoms in children: A confirmatory factor-analytic study. Journal of Abnormal Psychology, 106(2), 280–297. https://doi.org/10.1037/0021-843X.106.2.280

Spence, S. H., Barrett, P. M., & Turner, C. M. (2003). Psychometric properties of the Spence Children’s Anxiety Scale with young adolescents. Journal of Anxiety Disorders, 17(6), 605–625. https://doi.org/10.1016/S0887-6185(02)00236-0

Spence, S. H., Rapee, R., McDonald, C., & Ingram, M. (2001). The structure of anxiety symptoms among preschoolers. Behaviour Research and Therapy, 39(11), 1293–1316. https://doi.org/10.1016/S0005-7967(00)00098-X

13. Items of the Scale (Questionnaire)