The Spence Preschool Anxiety Scale<\/strong> (PAS; often designated in literature as the Spence Preschool Anxiety Scale) is a widely recognized, parent-report psychometric instrument designed to evaluate anxiety symptoms and related clinical phenotypes in early childhood, typically targeting children aged 2.5 to 6.5 years. Developed by Susan H. Spence, Ronald M. Rapee, Catherine McDonald, and Megan Ingram (2001), the instrument adapts diagnostic criteria from the Diagnostic and Statistical Manual of Mental Disorders<\/a> (DSM) to capture the developmentally nuanced presentation of anxiety in early developmental stages. The scale comprises 34 items: 28 core items assessing anxiety across five distinct, empirically derived dimensions, accompanied by an additional traumatic event screener (Item 29) and five subsequent post-traumatic stress symptom items (Items 30 to 34). The core dimensions encompass Generalized Anxiety (5 items), Social Anxiety (6 items), Obsessive-Compulsive Disorder (5 items), Physical Injury Fears (7 items), and Separation Anxiety (5 items). Items are rated on a 5-point Likert scale ranging from 0 (“Not True at All”) to 4 (“Very Often True”), yielding a continuous core score spanning 0 to 112. Extensive psychometric evaluations demonstrate robust internal consistency across community and clinical cohorts (total scale Cronbach\u2019s α typically ranging from .86 to .92; subscale alphas spanning .65 to .85), alongside adequate test-retest reliability and solid convergent and discriminant validity with established broadband instruments such as the Child Behavior Checklist (CBCL)<\/a>. Confirmatory factor analyses support a hierarchical model composed of five correlated first-order factors loading onto a single higher-order anxiety construct, mirroring the clinical architecture observed in older youth. The scale serves as an essential tool in clinical diagnostics, pediatric screening, early intervention assessment, and longitudinal developmental psychopathology research.<\/p>\n Spence Preschool Anxiety Scale, Preschool Anxiety Scale, Early Childhood Anxiety, Pediatric Anxiety Assessment, Parent-Report Measure, Generalized Anxiety, Separation Anxiety, Social Phobia, Psychometrics, Confirmatory Factor Analysis, Developmental Psychopathology<\/p>\n The Spence Preschool Anxiety Scale was developed through a collaborative research initiative led by prominent clinical psychologists specializing in pediatric anxiety and child psychopathology in Australia:<\/p>\n The primary purpose of the Spence Preschool Anxiety Scale is to provide a developmentally calibrated, psychometrically robust, parent-report screening and assessment instrument capable of identifying anxiety symptoms in young children aged approximately 2.5 to 6.5 years. Historically, the assessment of psychiatric symptomatology in preschool-aged children was constrained by theoretical assumptions suggesting that early fears were strictly normative, transitory, and diagnostically undifferentiated. However, emerging developmental psychopathology literature in the late 1990s and early 2000s revealed that severe anxiety during early childhood frequently runs a chronic course, predicts adverse trajectories into middle childhood and adolescence, and results in pronounced psychosocial and functional impairment.<\/p>\n Because children under seven years of age lack the advanced metacognitive, introspective, and communicative capacities necessary to complete structured self-report psychometric inventories reliably, clinical assessment relies heavily on standardized caregiver observation. The PAS was engineered to operationalize anxiety symptoms via observable, concrete behavioral manifestations that parents can reliably detect in routine daily environments\u2014such as home routines, bedtime transitions, peer interactions, and daycare or preschool environments.<\/p>\n From an applied perspective, the PAS serves multiple critical functions across clinical and empirical contexts:<\/p>\n The Spence Preschool Anxiety Scale operationalizes childhood anxiety as a multifaceted construct reflecting cognitive, behavioral, affective, and somatic expressions of maladaptive fear and apprehension. Although developmental psychologists acknowledge that transient fears (e.g., fear of the dark, brief separation distress) are universal during the preschool period, clinical anxiety emerges when such reactions are disproportionate in magnitude, inflexible across contexts, developmentally inappropriate, and disruptive to normative daily functioning. The PAS captures this construct through five primary dimensions totaling 28 items, plus a supplemental trauma-related screener and symptom domain:<\/p>\n The Generalized Anxiety dimension captures diffuse, pervasive, and difficult-to-control worry and autonomic apprehension that is not confined to a single environmental trigger or discrete situational stimulus. Young children with elevated scores on this subscale manifest continuous cognitive vigilance, anticipating catastrophic outcomes across disparate daily domains (e.g., minor changes in routines, future events). Observable behavioral markers include somatic restlessness, motor tension, irritability driven by internal distress, severe sleep onset difficulties due to rumination, and an incessant, repetitive need for parental reassurance even in benign situations.<\/p>\n This subscale assesses marked, persistent fear and avoidance of social evaluative situations, peer groups, and unfamiliar adults. In preschoolers, social anxiety manifests not only as shyness but as debilitating distress when the child is expected to interact with peers, participate in structured classroom routines (e.g., speaking during “show and tell”), or approach unfamiliar individuals. The items evaluate the child’s underlying concern about looking foolish or doing something embarrassing, fear of approaching novel peer groups, and intense reluctance or outright mutism when needing to communicate with authority figures such as preschool teachers.<\/p>\n The Obsessive-Compulsive Disorder (OCD) dimension evaluates persistent, intrusive, egodystonic mental events or thoughts alongside repetitive, rigid, and ritualistic behavioral routines. Because preschool children naturally engage in developmental rituals and game-like repetition, the PAS specifically targets behaviors performed to alleviate internal distress or prevent subjectively catastrophic events. Items assess compulsive checking behaviors (e.g., verifying repeatedly that taps are turned off or doors are shut), compulsive cleaning or excessive handwashing, rigid adherence to symmetry or ordering of personal items, the intrusion of distressing internal thoughts\/images, and the use of covert mental rituals (e.g., counting or repeating specific words to avert harm).<\/p>\n Reflecting the domain of specific phobias, this dimension captures severe, circumscribed, and irrational fears of specific environmental objects, situations, or natural phenomena that threaten bodily integrity or invoke acute terror. While evolutionary preparedness makes low-level wariness of heights, darkness, or animals common in early childhood, the PAS identifies clinical-level avoidance and physiological panic. The items measure severe terror of heights, enclosed or crowded spaces, thunder and lightning storms, water\/swimming, specific animals (dogs), insects\/spiders, and pervasive dread of the dark.<\/p>\n Separation Anxiety represents excessive, developmentally non-normative distress concerning real or anticipated separation from primary attachment figures (most commonly parents). The construct reflects core evolutionary attachment mechanisms gone awry. Symptoms include intense panic, crying, or somatic complaints during drop-offs at daycare, preschool, or with babysitters; persistent refusal or reluctance to sleep alone or sleep away from home; recurring nightmares centered on themes of parental loss or abandonment; and catastrophic ruminations that the parent will be injured, fall ill, be killed, or that the child will be lost or kidnapped.<\/p>\n Although not incorporated into the core 28-item total anxiety score, the scale provides a structured trauma module. Item 29 serves as a gateway screener identifying whether the child has experienced a severe, objectively traumatic event (e.g., catastrophic accident, physical assault, natural disaster, sudden loss). If positive, items 30 to 34 evaluate classic post-traumatic stress disorder (PTSD) reactivity, including intrusive re-experiencing, trauma-related nightmares, behavioral reenactment (reliving the event during play), acute distress upon exposure to reminders, and pronounced autonomic hyperarousal.<\/p>\n The conceptual architecture of the Spence Preschool Anxiety Scale is anchored in modern developmental psychopathology<\/a> and integrates multiple foundational models of emotional dysregulation, including Barlow\u2019s Triple Vulnerability Model, Bowlby\u2019s Attachment Theory, and Rachman\u2019s conditioning and cognitive theories of fear acquisition.<\/p>\n Developmental psychopathology posits that psychiatric disorders in youth are best understood not as discrete, static adult illnesses transplanted onto children, but as deviations from normative ontogenic developmental pathways. Normative development in early childhood is characterized by a series of transient, evolutionary fears: stranger distress emerges in late infancy; separation anxiety peaks between 12 and 18 months; fears of monsters, darkness, and large animals dominate the toddler years; and performance or evaluative fears typically manifest as children enter social schooling contexts. The theoretical framework of the PAS asserts that clinical anxiety disorders reflect quantitative exaggerations and qualitative dysfunctions of these normative developmental systems. By translating formal DSM diagnostic categories into behavioral indicators tailored to early childhood, the PAS bridges nosological psychiatry and developmental psychology.<\/p>\n The internal organization of the scale closely mirrors David H. Barlow’s<\/a> Triple Vulnerability Model, which outlines three interactive risk dimensions in the etiology of anxiety:<\/p>\n The conceptualization of the Separation Anxiety subscale is deeply rooted in John Bowlby’s<\/a> Attachment Theory and Mary Ainsworth’s<\/a> internal working models. Under optimal conditions, the caregiver serves as a secure base enabling exploratory behavior and an emotional haven offering solace during threat. In separation-anxious young children, the secure base phenomenon collapses: the child perceives external exploration as perilous, constantly monitors the physical proximity of the primary caregiver, and experiences catastrophic distress when physical contact is severed, reflecting an insecure-anxious or ambivalent attachment organization.<\/p>\n Extensive psychometric investigations across international clinical and community samples provide comprehensive evidence supporting the construct, convergent, discriminant, and criterion-related validity of the Spence Preschool Anxiety Scale.<\/p>\n The structural validity of the PAS was rigorously evaluated in the foundational study by Spence, Rapee, McDonald, and Ingram (2001) using a community sample of 270 Australian parents of preschoolers aged 3 to 5 years, along with a validation sample of 1,008 children. Confirmatory factor analysis (CFA) demonstrated that a hierarchical model\u2014featuring five distinct first-order factors corresponding to Generalized Anxiety, Social Anxiety, Obsessive-Compulsive Disorder, Physical Injury Fears, and Separation Anxiety, all loading onto a single higher-order global anxiety factor\u2014provided a superior fit to the data compared to alternative one-factor or uncorrelated models. This factor structure confirmed that early childhood anxiety manifests in differentiated symptom clusters analogous to those observed in older children and adults, while maintaining strong shared variance reflecting broad internalizing distress.<\/p>\n Convergent validity has been established by cross-referencing PAS scores with established broadband and narrowband socio-emotional parent-rating scales. Spence et al. (2001) demonstrated robust, statistically significant positive correlations between the PAS core total score and the Internalizing Problems scale of the Child Behavior Checklist (CBCL 1.5\u20135)<\/em> ($r = .68$<\/span> to $.74, p < .001$<\/span>). At the subscale level, the PAS Separation Anxiety score correlated strongly with the CBCL Withdrawn\/Depressed and Anxious\/Depressed scales. Subsequent studies evaluating convergent validity against the Behavior Assessment System for Children (BASC-2)<\/em> and the Infant-Toddler Social and Emotional Assessment (ITSEA)<\/em> confirmed high concordance ($r > .60$<\/span>) between corresponding anxiety and emotional reactivity indices.<\/p>\n The scale effectively distinguishes internalizing anxiety symptoms from externalizing disruptive behaviors. Spence et al. (2001) reported that while the PAS total score exhibited modest correlations with the CBCL Externalizing scale ($r = .29$<\/span> to $.36$<\/span>), these values were significantly lower than its correlations with the Internalizing scale, demonstrating clear divergence between emotional distress and behavioral conduct problems\/hyperactivity. Furthermore, the PAS subscales demonstrate divergent validity amongst themselves; for instance, the Physical Injury Fears subscale shares only low-to-moderate variance with the Social Anxiety subscale, confirming that parents distinguish situational\/phobia reactivity from interpersonal evaluative distress.<\/p>\n The clinical utility of the PAS is reinforced by its capacity to discriminate sharply between clinically diagnosed and non-clinical populations. In a clinical validation study conducted by Rapee, Kennedy, Ingram, Edwards, and Sweeney (2005), preschool children presenting to outpatient anxiety specialty clinics exhibited significantly higher mean scores across all five PAS subscales and the total scale compared to age-matched community controls ($d > 1.20, p < .001$<\/span>). Receiver Operating Characteristic (ROC) analyses demonstrated excellent diagnostic accuracy, with Area Under the Curve (AUC) statistics exceeding .85 for the total score in differentiating clinical anxiety cases from non-anxious children. Longitudinal follow-up studies (e.g., Edwards, Rapee, & Kennedy, 2010) demonstrated that elevated PAS scores in preschool significantly predicted formal DSM anxiety disorders at age 9, establishing longitudinal predictive validity.<\/p>\n The Spence Preschool Anxiety Scale exhibits exemplary psychometric reliability across community, educational, and clinical populations, as evidenced by evaluations of internal consistency, test-retest stability, and inter-rater agreement.<\/p>\n In the original normative sample of 1,008 Australian children (Spence et al., 2001), the internal consistency of the core 28-item total scale was excellent, yielding a Cronbach\u2019s alpha ($\u0007lpha$<\/span>) of .89. Subscale internal consistencies demonstrated solid reliability, particularly given the developmental heterogeneity of the age bracket and the brief nature of the subscales:<\/p>\n Independent cross-cultural validation studies have systematically replicated these figures. For example, in a Dutch validation study involving 399 preschool children, Broeren and Muris (2008) reported a total scale alpha of .89, with subscale alphas ranging from .65 (Physical Injury Fears) to .81 (Social Anxiety). Edwards et al. (2010) reported McDonald’s omega ($\\omega$<\/span>) coefficients exceeding .88 for the total score in community samples, demonstrating that the scale’s internal structure is not an artifact of tau-equivalence assumptions.<\/p>\n Temporal stability has been verified across multiple time intervals. Spence et al. (2001) evaluated 12-month test-retest reliability in a non-clinical cohort, documenting an intraclass correlation coefficient (ICC) of $r = .62$<\/span> ($p < .001$<\/span>) for the total score. While this coefficient reflects reasonable stability, it appropriately mirrors the dynamic developmental shifts and contextual fluctuations characteristic of preschool socio-emotional growth. Shorter-term test-retest evaluations over a 2- to 4-week window (Broeren & Muris, 2008) yielded substantially higher stability coefficients, with test-retest correlations ranging between $r = .79$<\/span> and $r = .86$<\/span> for the total scale, and between .68 and .82 for the individual subscales, indicating that the instrument is resistant to transient measurement error.<\/p>\n Inter-parent agreement between mothers and fathers has been documented as moderate to strong. Spence et al. (2001) reported mother-father concordance correlations ranging from $r = .53$<\/span> to $.65$<\/span> ($p < .001$<\/span>) across the subscales, with the total score showing an inter-rater agreement of $r = .64$<\/span>. These figures compare favorably with general meta-analytic averages for cross-informant internalizing ratings in child psychopathology, where maternal-paternal agreement typically hovers around .50 to .60.<\/p>\n The structural composition of the Spence Preschool Anxiety Scale has been extensively evaluated using both exploratory factor analysis (EFA) and confirmatory factor analysis<\/a> (CFA), cementing its position as one of the most structurally sound preschool emotional screening instruments available.<\/p>\n During initial scale development, Spence and colleagues conducted an exploratory principal axis factoring with oblimin rotation on an item pool derived from DSM-IV diagnostic conceptualizations. The resulting empirical solution cleanly extracted five primary factors displaying eigenvalues greater than 1.0, accounting for 48.7% of the total variance. Crucially, items loaded heavily on their theoretically designated dimensions (factor loadings > .40), with minimal cross-loadings observed<\/span>.<\/p>\n To substantiate this structure, Spence et al. subsequently tested competing structural models using Confirmatory Factor Analysis (maximum likelihood estimation) in a broader validation cohort ($N = 1,008$<\/span>). Four structural models were systematically compared:<\/p>\n Within the optimal higher-order structural model, standard standardized factor loadings ($lambda$<\/span>) for individual items onto their respective first-order constructs were uniformly robust:<\/p>\n Furthermore, second-order factor loadings of the five first-order latent constructs onto the broad anxiety construct were uniformly high: Generalized Anxiety ($\\gamma = .93$<\/span>), Separation Anxiety ($\\gamma = .86$<\/span>), Social Anxiety ($\\gamma = .79$<\/span>), Obsessive-Compulsive Disorder ($\\gamma = .68$<\/span>), and Physical Injury Fears ($\\gamma = .58$<\/span>). This empirical distribution demonstrated that Generalized and Separation Anxiety form the core structural nexus of early childhood anxiety, while specific fears and compulsive traits diverge slightly into more specialized phenotypes.<\/p>\n Subsequent psychometric investigations internationally have verified measurement invariance across sex and cultural backgrounds. Confirmatory analyses conducted in Dutch cohorts (Broeren & Muris, 2008), Spanish cohorts (Orgiles et al., 2013), and Chinese cohorts (Zhao et al., 2016) confirmed configural and metric invariance, demonstrating that the five-factor higher-order construct is cross-culturally stable and operates uniformly across boys and girls in early childhood.<\/p>\n The Spence Preschool Anxiety Scale is a standardized, parent-completed assessment protocol. Its operational parameters and administration guidelines are outlined below:<\/p>\n The Spence Preschool Anxiety Scale was initially published in 2001 following comprehensive validation studies by Professor Susan H. Spence and colleagues. In accordance with Professor Spence\u2019s commitment to facilitating global child mental health research and accessible clinical practice, the instrument is made available as an open-access, free-to-use<\/strong> assessment tool.<\/p>\n Licensing and Usage Conditions:<\/strong><\/p>\nKeywords<\/h2>\n
Authors<\/h2>\n
\n
Purpose<\/h2>\n
\n
Psychological Construct<\/h2>\n
1. Generalized Anxiety (5 Items: 1, 4, 8, 14, 28)<\/h3>\n
2. Social Anxiety (6 Items: 2, 5, 11, 15, 19, 23)<\/h3>\n
3. Obsessive-Compulsive Disorder Symptoms (5 Items: 3, 9, 18, 21, 27)<\/h3>\n
4. Physical Injury Fears (7 Items: 7, 10, 13, 17, 20, 24, 26)<\/h3>\n
5. Separation Anxiety (5 Items: 6, 12, 16, 22, 25)<\/h3>\n
6. Post-Traumatic Stress Symptoms (Items 29\u201334)<\/h3>\n
Theoretical Framework<\/h2>\n
Developmental Psychopathology and Ontogenic Adaptation<\/h3>\n
Barlow’s Triple Vulnerability Model<\/h3>\n
\n
Attachment Theory and Separation Dynamics<\/h3>\n
Validity<\/h2>\n
Construct and Structural Validity<\/h3>\n
Convergent Validity<\/h3>\n
Discriminant Validity<\/h3>\n
Criterion and Known-Groups Validity<\/h3>\n
Reliability<\/h2>\n
Internal Consistency<\/h3>\n
\n
Test-Retest Reliability<\/h3>\n
Inter-Rater Reliability<\/h3>\n
Factor Analysis<\/h2>\n
Foundational Factor Analyses (Spence et al., 2001)<\/h3>\n
\n
Item Factor Loadings<\/h3>\n
\n
Cross-Cultural and Longitudinal Invariance<\/h3>\n
Instrument \/ Measurement Tool<\/h2>\n
\n
\n
\n
\n
\n
Permissions & Fee and Test Year<\/h2>\n
\n
References<\/h2>\n
\n
13. Items of the Scale (Questionnaire)<\/h2>\n