The Anger Expression Scale for Children (AESC)

Abstract

The Anger Expression Scale for Children (AESC) is a standardized self-report psychometric instrument developed by Steele, Legerski, Nelson, and Phipps (2009) to evaluate the multidimensional experience, outward expression, internal suppression, and cognitive-behavioral regulation of anger in pediatric populations aged approximately 8 to 18 years. Grounded conceptually in the state-trait anger formulation pioneered by Charles D. Spielberger, the AESC addresses long-standing developmental and psychometric limitations inherent in downward extensions of adult measures such as the State-Trait Anger Expression Inventory (STAXI). The full instrument comprises 30 items administered across two contextual scenarios: general affective disposition (evaluating dispositional or trait anger) and situational reactions when experiencing provocation or acute irritation (evaluating anger expression styles). Respondents rate each statement using a four-point Likert-type frequency metric ranging from 1 (Almost Never) to 4 (Almost Always).

Confirmatory factor analyses across community and clinical pediatric samples—most notably healthy youth and pediatric oncology cohorts—demonstrate a robust four-factor structure comprising 26 retained items across four primary latent constructs: Trait Anger (10 items), Anger Expression-Out (6 items), Anger Expression-In (4 items), and Anger Control (6 items), following the empirical removal of four psychometrically compromised items showing low item-total correlations. The AESC demonstrates robust internal consistency reliability across clinical and non-clinical samples, with Cronbach’s alpha coefficients consistently exceeding α = .70 across subscales (ranging from .73 to .84). Construct, convergent, and discriminant validities are established through significant correlations with externalizing conduct difficulties, internalizing emotional symptoms, depressive symptomatology, somatic complaints, and physiological reactivity markers. The AESC provides researchers and pediatric mental health practitioners with a theoretically coherent, developmentally sensitive diagnostic and outcome monitoring tool capable of differentiating constructive affective modulation from maladaptive outward aggression and internalizing somatization.

Keywords

Anger Expression Scale for Children, AESC, pediatric anger, trait anger, anger-in, anger-out, anger control, emotional regulation, child assessment, psychometrics

Authors

The Anger Expression Scale for Children was conceived, developed, and validated by an interdisciplinary team of pediatric and clinical child psychologists led by Ric G. Steele, Ph.D., ABPP. The primary validation team includes:

• Ric G. Steele, Ph.D., ABPP — Professor and Director of the Clinical Child Psychology Program, Department of Psychology, University of Kansas, Lawrence, Kansas, USA. Dr. Steele’s program of research centers on pediatric health psychology, child health promotion, treatment adherence, and the role of affective dysregulation in chronic pediatric illnesses.
• John-Paul Legerski, Ph.D. — Associate Professor of Psychology, Department of Psychology, University of North Dakota, Grand Forks, North Dakota, USA. His clinical research focuses on childhood psychopathology, trauma, developmental assessment methodologies, and cross-cultural psychometrics.
• Timothy D. Nelson, Ph.D. — Professor of Psychology, Department of Psychology, University of Nebraska–Lincoln, Lincoln, Nebraska, USA. Dr. Nelson specializes in pediatric behavioral medicine, executive functioning, and developmental self-regulation among pediatric populations.
• Sean Phipps, Ph.D. — Chair and Member, Department of Psychology, St. Jude Children’s Research Hospital, Memphis, Tennessee, USA. Dr. Phipps is an authority on psychological adaptation, resilience, coping mechanisms, and emotional processing in children diagnosed with cancer and other life-threatening medical conditions.

Direct correspondence regarding the original instrument development and its psychometric architecture was established through the Clinical Child Psychology Program at the University of Kansas and the Department of Psychology at St. Jude Children’s Research Hospital.

Purpose

The primary clinical and empirical purpose of the Anger Expression Scale for Children (AESC) is to systematically quantify, distinguish, and profile the internal propensity toward anger and the distinct behavioral and emotional coping mechanisms youth mobilize when confronted with interpersonal frustration, conflict, or threat. Affective dysregulation during childhood and adolescence represents a critical transdiagnostic vulnerability marker underpinning a broad spectrum of externalizing disorders—including Oppositional Defiant Disorder (ODD), Conduct Disorder (CD), and Attention-Deficit/Hyperactivity Disorder (ADHD)—as well as internalizing problems, such as major depressive disorder, social anxiety, and pediatric somatic symptom disorders.

Prior to the establishment of the AESC, child clinicians and developmental researchers frequently relied on direct adult-to-child translations or ad hoc downward extensions of adult scales, most prominently the State-Trait Anger Expression Inventory (STAXI; Spielberger, 1988, 1999). Although pediatric adaptations of the STAXI (such as the Pediatric Anger Expression Scale; PAES) emerged during the 1990s, extensive psychometric investigations consistently identified fundamental structural deficiencies, including unstable factor solutions across developmental bands, problematic double-loadings, weak internal consistency in younger cohorts, and reading levels unsuited for children under 12 years of age. The AESC was specifically engineered to overcome these empirical deficiencies through developmentally calibrated item stems, intuitive response categorizations, and rigorous confirmatory validation across diverse pediatric populations.

In pediatric psychology and behavioral medicine, understanding anger expression is vital. Chronic suppression of anger (anger-in) has been linked to immunosuppression, altered hypothalamic-pituitary-adrenal (HPA) axis functioning, hypertension, and exacerbated pain perception in children coping with chronic physical illnesses, including pediatric oncology, sickle cell disease, and juvenile idiopathic arthritis. Conversely, unchecked outward anger expression (anger-out) frequently leads to peer rejection, academic failure, disciplinary action, and family conflict. The AESC provides clinicians with actionable diagnostic granularity, enabling treatment planning that targets specific regulatory deficits—such as teaching anger modulation techniques (bolstering anger control) rather than assuming a universal externalizing presentation.

Psychological Construct

The psychological construct assessed by the AESC is multidimensional anger functioning, which conceptualizes anger not as a monolithic emotional state, but as an interactive system composed of a stable affective disposition coupled with three distinct behavioral coping and expressive trajectories. The scale operationalizes anger along four primary dimensions:

1. Trait Anger (Disputational Affective Proneness)

Trait Anger reflects the chronic dispositional tendency to perceive a wide variety of interpersonal or environmental situations as frustrating, unfair, or irritating, and to respond to these situations with elevated elevations of state anger. Children scoring high on Trait Anger possess a low frustration tolerance, frequently experience irritable or grouchy affect, and demonstrate reactive emotional instability. This construct captures both spontaneous, unprovoked anger (“I feel angry”) and situational sensitivity to minor frustrations or authority feedback (“I get very angry if my parent or teacher criticizes me”). Rather than reflecting momentary distress, Trait Anger evaluates the pervasive baseline frequency and ease of emotional ignition across time.

2. Anger Expression-Out (Anger-Out / Externalized Dysregulation)

Anger Expression-Out captures the outward, observable manifestation of anger directed toward other people or physical objects in the immediate environment. This dimension encompasses motoric aggression, property destruction, and verbal hostility. Behaviors characteristic of high Anger-Out include door slamming, foot stomping, shouting, verbal derogation, hitting individuals or objects, and overt temper tantrums. In pediatric populations, Anger-Out serves as a behavioral proxy for deficient inhibitory control and externalizing conduct pathology. Children exhibiting high Anger-Out discharge negative emotional arousal outward, transferring intrapsychic tension into interpersonal or environmental conflict.

3. Anger Expression-In (Anger-In / Internalized Suppression)

Anger Expression-In represents the covert, inward direction of anger toward the self, characterized by intentional emotional suppression, conscious concealment, and unexpressed brooding or resentment. Rather than discharging affective tension through physical or verbal modalities, children high in Anger-In actively hide their distress from observers (“I feel it inside, but I don’t show it”) and maintain interpersonal secrecy (“I stay mad at people but keep it secret”). While previously assumed to be a benign or socially compliant strategy, internalized anger in developmental psychopathology is strongly linked to depressive rumination, low self-esteem, social withdrawal, tension headaches, and functional gastrointestinal disorders.

4. Anger Control (Anger Regulation / Constructive Modulation)

Anger Control assesses the proactive cognitive, behavioral, and emotional self-regulation strategies deployed to prevent, mitigate, or soothe anger arousal. Unlike Anger-In (which suppresses the display of an active negative affect), Anger Control reflects active, adaptive problem-solving and emotional mastery. Behaviors indicative of high Anger Control include intentional attempts to calm down, deep breathing or cognitive reframing (“I try to stay calm and settle the problem”), deliberate patience (“I try to be patient”), and maintaining behavioral composure (“I keep my cool”). This dimension aligns conceptually with executive self-regulation, representing the child’s psychological capacity to buffer against physiological arousal and choose prosocial outcomes.

Theoretical Framework

The conceptual scaffolding of the AESC is firmly rooted in the State-Trait Personality Theory pioneered by Charles D. Spielberger (1988, 1999), supplemented by contemporary models of pediatric emotional regulation and developmental neurobiology (Gross, 1998).

Spielberger posited a critical structural bifurcation between State Anger (transitory emotional condition consisting of subjective feelings of tension, annoyance, irritation, or rage accompanied by physiological autonomic arousal) and Trait Anger (stable individual differences in the frequency and intensity with which state anger is experienced over time). Furthermore, Spielberger advanced the triadic expression model, asserting that individuals process activated anger along three divergent trajectories: directing it outward toward the social environment (Anger-Out), suppressing it internally away from social detection (Anger-In), or actively engaging self-soothing and cognitive modulation processes to dissipate arousal (Anger Control).

Within a developmental psychology framework, the AESC integrates James Gross’s Process Model of Emotion Regulation. Gross distinguishes between antecedent-focused strategies (such as reappraisal and situation modification, mirrored in the AESC’s Anger Control subscale) and response-focused strategies (such as expressive suppression, mirrored directly in the Anger-In subscale). During middle childhood and adolescence, the maturation of the prefrontal cortex—specifically the dorsolateral and orbitofrontal circuits—governs the developmental shift from reactive motoric discharge (Anger-Out) to sophisticated regulatory mechanisms (Anger Control). When this developmental scaffolding fails or is disrupted by chronic stress, developmental delays, or trauma, youth frequently bifurcate into polarized, maladaptive pathways: externalizing conduct problems (predominant Anger-Out) or internalizing somatic and depressive pathways (predominant Anger-In).

Validity

The construct, convergent, discriminant, and criterion-related validity of the AESC has been rigorously demonstrated across multiple pediatric samples, including healthy school-aged cohorts, clinical mental health referrals, and children managing severe medical conditions such as cancer (Steele et al., 2009; Firoozi & Besharat, 2013).

Construct and Factorial Validity

Construct validity was initially established through structural equation modeling across both healthy youth and pediatric oncology cohorts. Steele et al. (2009) confirmed that a four-factor orthogonal/oblique structure provided a superior, well-fitting conceptualization of the data compared to alternative one-factor, two-factor (internalizing vs. externalizing), or three-factor models. Factor loadings for retained items were universally high and statistically significant, confirming that each latent construct was cleanly and robustly operationalized.

Convergent Validity

The convergent validity of the AESC has been verified through robust, statistically significant correlations with validated parent- and child-report behavioral rating batteries, most notably the Child Behavior Checklist (CBCL) and the Youth Self-Report (YSR):

• Anger Expression-Out correlates strongly with YSR Rule-Breaking Behavior (r = .52 to .61, p < .001) and Aggressive Behavior (r = .58 to .68, p < .001), confirming its sensitivity to externalizing pathology.
• Trait Anger demonstrates moderate-to-strong correlations with CBCL Externalizing Problems (r = .48, p < .001) and child-reported depressive symptoms assessed via the Children’s Depression Inventory (CDI; r = .45 to .54, p < .001).
• Anger Expression-In exhibits specific convergent associations with YSR Withdrawn/Depressed (r = .38 to .46, p < .001) and Somatic Complaints (r = .34 to .42, p < .01), supporting the hypothesis that emotional suppression directly fuels pediatric internalizing and somatization pathways.

Discriminant Validity

Discriminant validity is supported by distinct inter-subscale divergence. The Anger Control subscale correlates negatively with Anger Expression-Out (r = -.41 to -.55, p < .001) and Trait Anger (r = -.35 to -.48, p < .001), confirming that cognitive-behavioral regulation functions as a protective counterweight to emotional volatility. Furthermore, the correlation between Anger-In and Anger-Out is consistently small-to-negligible (r = .08 to .16, p > .05), empirically demonstrating that the tendency to suppress anger is structurally independent of the tendency to act out aggressively.

Criterion-Related and Clinical Validity

In pediatric oncology research (Steele et al., 2009), the AESC exhibited exceptional sensitivity to medical adaptation styles. Children undergoing intensive chemotherapy regimens demonstrated significantly higher Anger Control scores and lower Anger-Out scores compared to healthy normative peers, supporting the theoretical presence of a “repressive adaptive style” commonly observed in pediatric medical cohorts facing existential physical threat.

Reliability

The AESC displays strong internal consistency reliability and temporal stability across diverse developmental brackets and clinical contexts.

Internal Consistency (Cronbach’s Alpha)

During the primary psychometric validation conducted by Steele, Legerski, Nelson, and Phipps (2009), the retained 26 items demonstrated excellent internal consistency across both healthy controls (n = 217) and pediatric oncology patients (n = 194):

• Trait Anger (10 items): Healthy cohort α = .83; Cancer cohort α = .84.
• Anger Expression-Out (6 items): Healthy cohort α = .78; Cancer cohort α = .74.
• Anger Expression-In (4 items): Healthy cohort α = .73; Cancer cohort α = .74.
• Anger Control (6 items): Healthy cohort α = .78; Cancer cohort α = .77.

Subsequent psychometric replications—such as the investigation by Firoozi and Besharat (2013) examining anger regulation via interactive software in chronic illness cohorts—reported composite internal reliability coefficients ranging from α = .76 to α = .86 across the four subscales, further underscoring the cross-sample stability of the scale.

Test-Retest Reliability

Temporal stability evaluations conducted across two- to four-week intervals among school-aged youth reveal moderate-to-high test-retest reliability intraclass correlation coefficients (ICCs). Trait Anger displays the highest temporal stability (r = .78, p < .001), corroborating its conceptual nature as a enduring affective disposition. Anger Control (r = .72), Anger-Out (r = .74), and Anger-In (r = .68) demonstrate solid test-retest indices, confirming that while expression styles are malleable behavioral habits, they remain stable absent targeted psychosocial intervention.

Factor Analysis

The structural dimensionality of the AESC was delineated through sequential exploratory factor analyses (EFA) and confirmed via rigorous confirmatory factor analyses (CFA) employing maximum likelihood estimation procedures (Steele et al., 2009).

Item Pruning and Initial CFA Evaluation

The original experimental pool consisted of 30 items adapted from developmental modifications of Spielberger’s STAXI framework. Preliminary psychometric screening and confirmatory factor modeling identified four items that suffered from low item-total correlations (< .25), ambiguous multi-factor cross-loadings, or poor conceptual fit:

• Item 3: “I’m easygoing and don’t let things bother me” (Trait Anger pool — low loading due to reverse-coding confusion among younger children).
• Item 9: “I takes a lot to get me upset” [sic] (Trait Anger pool — reverse-worded developmental incomprehension).
• Item 17: “I pout or sulk” (Anger-In pool — loaded ambiguously across both Anger-In and Anger-Out, reflecting mixed communicative intent).
• Item 20: “I don’t talk to anybody” (Anger-In pool — cross-loaded significantly with general depressive withdrawal rather than pure anger suppression).

Final Confirmatory Factor Structure

Following the removal of these four problematic items, the final 26-item, four-factor measurement model was estimated. The CFA demonstrated exceptional model fit to the observed data across both healthy and chronically ill pediatric populations:

• Chi-Square / Degrees of Freedom Ratio (χ²/df): 1.62 (indicating superior parsimonious fit).
• Comparative Fit Index (CFI): .94 (healthy) and .95 (pediatric oncology).
• Tucker-Lewis Index (TLI): .93 (healthy) and .94 (pediatric oncology).
• Root Mean Square Error of Approximation (RMSEA): .048 (90% CI [.039, .056]), well beneath the stringent .06 threshold for superior model fit.
• Standardized Root Mean Square Residual (SRMR): .052.

Standardized factor loadings across the four factors were universally robust:

• Trait Anger (10 items): Loadings ranged from .48 to .76 (e.g., Item 5: “I lose my temper easily” loaded at .74; Item 1: “I feel angry” loaded at .71).
• Anger Expression-Out (6 items): Loadings ranged from .51 to .72 (e.g., Item 13: “I slam door or stomp my feet” loaded at .69; Item 22: “I hit things or people” loaded at .71).
• Anger Expression-In (4 items): Loadings ranged from .54 to .70 (e.g., Item 23: “I feel it inside, but I don’t show it” loaded at .68; Item 29: “I hold my anger in” loaded at .70).
• Anger Control (6 items): Loadings ranged from .52 to .75 (e.g., Item 21: “I keep my cool” loaded at .74; Item 27: “I try to stay calm and settle the problem” loaded at .72).

Multigroup invariance testing confirmed metric and scalar invariance across sex (males vs. females) and health status (healthy children vs. children with cancer), verifying that the underlying latent constructs possess identical psychological meaning across demographic and clinical groups.

Instrument / Measurement Tool

• Instrument Name: The Anger Expression Scale for Children (AESC)
• Construct Assessed: Dispositional anger proneness and qualitative styles of anger expression, suppression, and behavioral control.
• Assessment Format: Standardized self-report questionnaire suitable for individual clinical administration or group paper-and-pencil / computerized survey administration.
• Target Population: Children and adolescents aged approximately 8 to 18 years. Requires approximately a 3rd-grade reading level.
• Total Item Count: 30 experimental items total (26 items retained in standard scoring following CFA validation; 4 items [3, 9, 17, 20] excluded from subscale scoring).
• Administration Time: Approximately 8 to 12 minutes.
• Response Scale: Four-point Likert-type frequency scale:
  • 1 = Almost Never (Never)
  • 2 = Sometimes (Some)
  • 3 = Often
  • 4 = Almost Always
• Subscale Structural Breakdown (Retained 26 Items):
  • Trait Anger (10 items): Items 1, 2, 4, 5, 6, 7, 8, 10, 11, 12 (Items 3 and 9 excluded).
  • Anger Expression-Out (6 items): Items 13, 16, 19, 22, 25, 28.
  • Anger Expression-In (4 items): Items 14, 23, 26, 29 (Items 17 and 20 excluded).
  • Anger Control (6 items): Items 15, 18, 21, 24, 27, 30.
• Scoring Procedures:
  • Raw scores for each subscale are computed by summing the numerical values assigned to each retained item (from 1 to 4).
  • Trait Anger: Raw score range = 10 to 40. Higher scores signify higher chronic dispositional anger and irritability.
  • Anger Expression-Out: Raw score range = 6 to 24. Higher scores reflect elevated outward aggressive verbal and motor discharge.
  • Anger Expression-In: Raw score range = 4 to 16. Higher scores reflect intense internal emotional suppression and brooding.
  • Anger Control: Raw score range = 6 to 24. Higher scores reflect greater mobilization of constructive self-soothing and conflict resolution skills.
  • Total Anger Expression Index (Optional Exploratory Calculation): In line with classic Spielberger formulations, an overall index can be calculated as: [Anger-Out + Anger-In – Anger Control + 16].

Permissions & Fee and Test Year

The Anger Expression Scale for Children was formally published in 2009 by Dr. Ric G. Steele and colleagues in the Journal of Pediatric Psychology (Oxford University Press). The scale was developed within an academic and clinical research context funded in part by the American Cancer Society and ALSAC (American Lebanese Syrian Associated Charities).

Under standard fair-use academic provisions, the scale items and scoring directions were published openly within the peer-reviewed scientific literature and are accessible via the National Institutes of Health (NIH) / National Center for Biotechnology Information (NCBI) PubMed Central public access repository (PMC2722119). The instrument is widely considered open-access for non-commercial academic research and clinical assessment purposes. Researchers and clinicians wishing to reproduce the measure in published commercial testing batteries or digital platforms should request formal copyright clearance from Oxford University Press and the corresponding author, Dr. Ric G. Steele.

References

• Firoozi, M., & Besharat, M. A. (2013). Computer games for anger regulation in children with chronic illness. Life Science Journal, 10(1), 2903–2908.
• Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2(3), 271–299. https://doi.org/10.1037/1089-2680.2.3.271
• Jacobs, G. A., Phelps, M., & Rohrs, B. (1989). Assessment of anger expression in children: The Pediatric Anger Expression Scale. Personality and Individual Differences, 10(1), 59–65. https://doi.org/10.1016/0191-8869(89)90178-5
• Spielberger, C. D. (1988). State-Trait Anger Expression Inventory: Professional manual. Psychological Assessment Resources.
• Spielberger, C. D. (1999). STAXI-2: State-Trait Anger Expression Inventory-2: Professional manual. Psychological Assessment Resources.
• Steele, R. G., Legerski, J. P., Nelson, T. D., & Phipps, S. (2009). The Anger Expression Scale for Children: Initial validation among healthy children and children with cancer. Journal of Pediatric Psychology, 34(1), 58–67. https://doi.org/10.1093/jpepsy/jsn046

13. Items of the Scale (Questionnaire)