Aggression/Victimization Scale

Abstract

The Aggression/Victimization Scale (AVS) is a standardized, self-report psychometric instrument designed to assess the behavioral frequency of peer-directed aggression perpetration and peer victimization among youth and early adolescents within school contexts. Developed out of foundational youth violence measurement paradigms established by Pamela Orpinas and Ralph Frankowski and compiled within the Centers for Disease Control and Prevention (CDC) Compendium of Assessment Tools, the instrument comprises 12 items organized into two correlated 6-item subscales: Victimization and Aggression. Each subscale systematically captures three primary behavioral typologies: overt physical aggression, direct verbal aggression, and indirect/relational aggression. Respondents report the precise occurrence frequency of each behavior during a concrete recall window of the preceding 7 days using a 7-point response scale ranging from 0 to 6 or more times (0, 1, 2, 3, 4, 5, 6+). Individual subscales yield summed scores from 0 to 36, producing a composite aggression-victimization score ranging from 0 to 72 points. Psychometric evaluations across diverse elementary and middle school cohorts (predominantly grades 4 through 8) demonstrate high internal consistency, with Cronbach’s alpha coefficients regularly spanning .83 to .90 for the perpetration dimension and .78 to .87 for the victimization dimension. Confirmatory factor analyses corroborate a robust two-factor structure delineating perpetration from victimization while preserving second-order loadings for physical, verbal, and relational expressions. The scale exhibits strong convergent validity with disciplinary infractions, teacher behavioral ratings, and emotional distress markers, making it a gold-standard instrument for school climate evaluations, youth violence epidemiology, and the rigorous pre-post evaluation of school-based violence prevention programs.

Keywords

Aggression/Victimization Scale, peer victimization, youth violence, school bullying, relational aggression, physical aggression, verbal aggression, behavioral psychometrics, adolescent assessment, violence prevention

Authors

The underlying psychometric architecture of the Aggression/Victimization Scale was established by Pamela Orpinas, Ph.D., M.P.H., and Ralph Frankowski, Ph.D. Dr. Orpinas is a Professor of Health Promotion and Behavior in the College of Public Health at the University of Georgia, whose career has focused on youth violence prevention, social-emotional development, and longitudinal trajectories of aggressive behaviors. Dr. Frankowski is a Professor Emeritus of Biostatistics at the University of Texas Health Science Center at Houston School of Public Health, specializing in longitudinal methodology, clinical trial design, and psychometrics.

The operationalized 12-item dual-construct instrument was further consolidated, standardized, and disseminated by the Division of Violence Prevention within the National Center for Injury Prevention and Control at the Centers for Disease Control and Prevention (CDC) under senior research syntheses authored by Linda L. Dahlberg, Susan B. Toal, Monica Swahn, and Christopher B. Behrens.

Purpose

Youth violence, bullying, and interpersonal victimization in primary and secondary educational institutions constitute significant public health concerns that predict chronic psychological maladjustment, academic decrement, and externalizing psychiatric outcomes. The primary purpose of the Aggression/Victimization Scale is to provide researchers, educational evaluators, school psychologists, and child clinicians with a brief, methodologically robust, and objective measurement tool to quantify behavioral manifestations of violence perpetration and peer victimization.

Historically, research on childhood interpersonal conflict suffered from measurement bifurcations: scales evaluated either aggressive behavior in isolation or focused exclusively on passive peer victimization. This artificial conceptual separation obscured the well-documented phenomenon of the “bully-victim” subgroup—individuals who both initiate aggressive actions against peers and simultaneously experience recurrent peer victimization. The AVS was intentionally designed with structural behavioral symmetry, capturing parallel manifestations of physical harm, verbal hostility, and relational exclusion across both perpetration and victimization dimensions.

In clinical and educational settings, the AVS serves multiple vital functions:

• Epidemiological Surveillance: Establishing baseline prevalence rates of peer aggression and victim status across classrooms, schools, and educational districts.
• Targeted Screening: Identifying youths at elevated risk for chronic externalizing disorders, social rejection, school avoidance, depressive symptoms, or retaliatory violence.
• Programmatic Evaluation: Serving as a highly sensitive outcome metric in randomized controlled trials and quasi-experimental evaluations of school-based violence prevention curricula (such as positive behavioral interventions, social-emotional learning platforms, and restorative justice frameworks).

By constraining the reporting timeframe to the immediate past 7 days, the AVS markedly minimizes the cognitive biases and recall inaccuracies common in children when queried about expansive timeframes (such as “over the past school year”), rendering it exceptionally well-suited for repeated-measures designs, continuous monitoring, and longitudinal time-series investigations.

Psychological Construct

The Aggression/Victimization Scale measures two fundamental, interrelated interpersonal behavioral constructs in youth: Aggression Perpetration and Peer Victimization. Within both macro-constructs, the scale operationalizes three distinct behavioral subtypes: physical, verbal, and relational/psychological.

1. Aggression Perpetration

Aggression perpetration is defined as intentional, non-accidental behavior directed toward a peer with the immediate intent to cause physical pain, psychological distress, social degradation, or reputational damage. Within the AVS, aggressive behaviors are assessed across three qualitative domains:

• Overt Physical Aggression: Direct somatic assault against another peer, manifested through behaviors such as pushing, shoving, hitting, or physically assaulting another student (Item 8).
• Direct Verbal Aggression: The explicit use of spoken language or vocalizations intended to belittle, insult, dominate, or terrorize another child. This includes name-calling (Item 9), malicious teasing (Item 7), and direct verbal threats of imminent physical harm (Item 10).
• Relational and Psychological Aggression: Non-physical behaviors explicitly intended to damage peer relationships, compromise social standing, or undermine social inclusion within the school ecosystem. It is expressed through active, deliberate social ostracism (Item 11: leaving another kid out on purpose) and the fabrication and dissemination of malicious rumors or false statements designed to manipulate peer group alliances (Item 12).

2. Peer Victimization

Peer victimization reflects the subjective experience of being the target or recipient of hostile, aggressive, or exclusionary actions initiated by peers within the school setting. Mirroring the perpetration construct, the victimization subscale assesses three symmetric facets:

• Physical Victimization: Suffering bodily assault, rough handling, shoving, or striking by school peers (Item 2).
• Verbal Victimization: Being the direct target of derogatory epithets, insulting labels (Item 3), mocking or derisive teasing (Item 1), or intimidating verbal threats of physical violence (Item 4).
• Relational and Psychological Victimization: Experiencing intentional, systemic social marginalization (Item 5: being excluded on purpose) and suffering damage to peer reputation via malicious falsehoods, slander, or social manipulation engineered by other students (Item 6).

Theoretical Framework

The conceptual framework underpinning the Aggression/Victimization Scale synthesizes several foundational models from developmental psychopathology, cognitive psychology, and social-ecological theory.

The Social-Ecological Model of Youth Violence

In accordance with the ecological systems paradigm proposed by Urie Bronfenbrenner and adapted for violence prevention by Dahlberg and Krug, aggressive behavior and victimization cannot be understood as isolated intrapsychic phenomena. Rather, they are nested within proximal microsystems, most notably the school peer group. The AVS concentrates on the proximal interpersonal transactions occurring within this immediate school ecology, recognizing that classroom norms, peer reinforcement, and playground dynamics directly mediate individual behavioral trajectories.

Social Information Processing (SIP) Theory

The cognitive underpinnings of youth aggression are extensively illuminated by the Social Information Processing model formulated by Kenneth A. Dodge and Nicki R. Crick. According to SIP theory, children navigate interpersonal interactions through a continuous six-step sequence: (1) encoding external cues, (2) interpreting social cues, (3) clarifying goals, (4) accessing behavioral responses, (5) selecting a response, and (6) enacting the chosen behavior. Children demonstrating high aggression scores on the AVS frequently exhibit a hostile attribution bias—the systematic tendency to interpret ambiguous peer behaviors (such as an accidental bump in a hallway) as intentionally malevolent—triggering immediate, reactive physical or verbal hostility. Conversely, chronically victimized youth often develop hyper-vigilance toward threatening cues, which leads to withdrawal, anxiety, and impaired social cue decoding.

The General Aggression Model (GAM)

Formulated by Craig A. Anderson and Brad J. Bushman, the General Aggression Model provides a comprehensive integrative framework explaining how personal traits (e.g., impulsivity, trait hostility) and situational inputs (e.g., peer provocation, social exclusion) influence an individual’s internal state—affecting cognition, physiological arousal, and emotional appraisals. The AVS captures recurring behavioral outcomes resulting from these internal states, operationalizing aggression as an ongoing interactional script reinforced through either instrumental outcomes (e.g., social dominance, peer compliance) or emotional catharsis.

Validity

The Aggression/Victimization Scale has undergone rigorous psychometric evaluation across multiple large-scale developmental and epidemiologic cohorts, confirming robust construct, convergent, discriminant, and predictive validity.

Construct and Factorial Validity

Construct validity is substantiated by structural equation modeling and confirmatory factor analyses showing that aggression and victimization form distinct yet moderately correlated dimensions ($r = .35$ to $.52$, $p < .001$). This confirms that while perpetration and victimization are conceptually separate experiences, they frequently co-occur within school settings, reflecting the empirical realities of retaliatory aggression and bully-victim dynamics.

Convergent Validity

The AVS demonstrates strong convergent associations with multiple independent criteria:

• School Disciplinary Data: Scores on the Aggression subscale correlate significantly with official disciplinary referrals for fighting, insubordination, and classroom disruptions ($r = .38$ to $.49$, $p < .001$).
• Teacher and Peer Sociometric Ratings: Peer nominations of aggressive behavior and teacher-rated externalizing problems on instruments such as the Child Behavior Checklist (CBCL) or the Teacher Report Form (TRF) correlate strongly with self-reported AVS aggression ($r = .42$ to $.56$).
• Internalizing Symptoms: Scores on the Victimization subscale demonstrate substantial convergent correlations with self-reported depressive symptomatology on the Children’s Depression Inventory (CDI), generalized anxiety, and feelings of loneliness and school alienation ($r = .40$ to $.55$, $p < .001$).

Discriminant Validity

Discriminant validity is supported by modest or non-significant correlations with unrelated cognitive and demographic variables, including standardized academic achievement test scores (after controlling for socio-economic status) and general cognitive ability. Furthermore, the Victimization subscale cleanly dissociates from non-interpersonal behavioral disturbances, such as purely somatic complaints without emotional distress.

Predictive and Evaluative Validity

Longitudinal evaluations demonstrate that elevated AVS aggression scores in elementary school (grades 4 and 5) significantly predict subsequent delinquency, substance initiation, and suspension rates in middle and high school. In intervention research, such as the Multisite Violence Prevention Project and various CDC-evaluated programs, the AVS has demonstrated sensitivity to change, capturing statistically significant reductions in weekly aggressive incidents following targeted socio-behavioral interventions.

Reliability

The reliability of the Aggression/Victimization Scale has been documented across numerous psychometric studies encompassing socio-demographically diverse youth populations.

Internal Consistency

In standard validation cohorts of fourth- through eighth-grade students, the instrument demonstrates high internal consistency:

• Aggression Subscale: Cronbach’s alpha ($lpha$) coefficients consistently range between .83 and .90. In the original validation work conducted by Orpinas and Frankowski (2001) evaluating adolescent aggression, alpha coefficients of .87 to .89 were observed across multi-wave longitudinal assessments.
• Victimization Subscale: Internal consistency estimates for the victimization scale typically range from .78 to .87, reflecting strong inter-item relatedness despite covering distinct physical, verbal, and relational domains.
• Full Composite Scale: When utilized as an overall 12-item index of school violence involvement, composite Cronbach’s alphas routinely exceed .88.

Test-Retest Reliability

Because the AVS evaluates behavioral occurrences over a concrete 7-day recall window, test-retest reliability estimates are inherently sensitive to real weekly fluctuations in interpersonal encounters. Over short intervals (1 to 2 weeks), intraclass correlation coefficients (ICCs) and Pearson stability coefficients range from .70 to .82, indicating substantial temporal stability while remaining sensitive to situational shifts and intervention effects.

Factor Analysis

Extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have established the structural dimensionality of the 12-item Aggression/Victimization Scale.

Structural Architecture: Two-Factor vs. Hierarchical Models

Initial principal axis factoring with promax (oblique) rotation cleanly delineates a two-factor solution accounting for greater than 55% of the total variance:

• Factor 1 (Victimization): Items 1 through 6 load heavily on this factor (factor loadings ranging from .58 to .81) with minimal cross-loadings (< .20) on perpetration items.
• Factor 2 (Aggression Perpetration): Items 7 through 12 demonstrate primary loadings on this factor (factor loadings ranging from .61 to .84).

Confirmatory factor analyses utilizing robust maximum likelihood (MLR) or weighted least squares means and variance adjusted (WLSMV) estimation confirm acceptable to excellent goodness-of-fit indices for the correlated two-factor model:

• Comparative Fit Index (CFI): $ge .95$
• Tucker-Lewis Index (TLI): $ge .94$
• Root Mean Square Error of Approximation (RMSEA): $le .052$ (90% CI [.041, .063])
• Standardized Root Mean Square Residual (SRMR): $le .045$

Alternative second-order hierarchical models—which introduce three lower-order latent factors (Physical, Verbal, Relational) nested within the two overarching dimensions of Aggression and Victimization—also exhibit strong psychometric fit, demonstrating that the AVS adequately reflects both global violence involvement and modality-specific behavioral expressions.

Instrument / Measurement Tool

• Instrument Name: Aggression/Victimization Scale (AVS)
• Primary Constructs Measured: Peer Aggression Perpetration (Physical, Verbal, Relational); Peer Victimization (Physical, Verbal, Relational)
• Target Population: Upper elementary and middle school students (primarily grades 4 to 8, ages 9 to 14); adaptable for high school populations
• Administration Format: Self-report questionnaire; paper-and-pencil or computerized/digital administration
• Estimated Completion Time: 5 to 10 minutes
• Item Count: 12 items total (6 Victimization items, 6 Aggression items)
• Recall Window: Explicitly constrained to the past 7 days (“during the last 7 days” / “during the previous week”)
• Response Options & Scoring Values: A 7-point ratio-frequency scale:
  • 0 times = 0 points
  • 1 time = 1 point
  • 2 times = 2 points
  • 3 times = 3 points
  • 4 times = 4 points
  • 5 times = 5 points
  • 6 or more times (6+) = 6 points
• Score Calculation:
  • Subscale Scores: Computed by summing respective items. Victimization Subscale (Items 1–6) ranges from 0 to 36 points. Aggression Subscale (Items 7–12) ranges from 0 to 36 points.
  • Total Composite Score: Sum of all 12 items, yielding a theoretical range of 0 to 72 points.
  • Interpretation: Each point reflects one distinct behavioral event reported during the preceding week. Higher scores correspond directly to elevated frequencies of aggressive behavior or victimization experiences.
• Missing Data Imputation Protocol:
  • If 4 or more items are missing across the instrument, the total score cannot be reliably computed and the protocol must be deemed invalid.
  • If 3 or fewer items are missing, missing values are replaced by the respondent’s mean item score across the answered items before calculating the final summed score.

Permissions & Fee and Test Year

The underlying psychometric groundwork for the scale was established in 2001 by Pamela Orpinas and Ralph Frankowski in their seminal validation of the Aggression Scale published in the Journal of Early Adolescence. The dual Aggression/Victimization Scale format was formally cataloged in 2005 by the Centers for Disease Control and Prevention (CDC) within Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools.

Because the compilation and publication of this instrument were funded and disseminated by the United States federal government via the CDC (National Center for Injury Prevention and Control), the tool resides within the public domain. It may be used, reproduced, and administered by researchers, educators, evaluators, and clinicians free of royalty or licensing fees. Users are expected to cite the foundational validation articles and the CDC compendium in all scholarly publications and evaluation reports.

References

• Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology, 53(1), 27-51. https://doi.org/10.1146/annurev.psych.53.100901.135231
• Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing mechanisms in children’s social adjustment. Psychological Bulletin, 115(1), 74-101. https://doi.org/10.1037/0033-2909.115.1.74
• Dahlberg, L. L., Toal, S. B., Swahn, M., & Behrens, C. B. (2005). Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (2nd ed.). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/YV_Compendium.pdf
• Dodge, K. A., & Coie, J. D. (1987). Social-information-processing factors in reactive and proactive aggression in children’s peer groups. Journal of Personality and Social Psychology, 53(6), 1146-1158. https://doi.org/10.1037/0022-3514.53.6.1146
• Orpinas, P., & Frankowski, R. (2001). The Aggression Scale: A self-report measure of aggressive behavior for young adolescents. The Journal of Early Adolescence, 21(1), 50-67. https://doi.org/10.1177/0272431601021001003
• Orpinas, P., Horne, A. M., & Project PACT Team. (2004). A schoolwide approach to preventing violence in schools: Results of a 3-year intervention. Journal of School Health, 74(6), 231-238. https://doi.org/10.1111/j.1746-1561.2004.tb08278.x
• Salmivalli, C., Lagerspetz, K., Björkqvist, K., Österman, K., & Kaukiainen, A. (1996). Bullying as a group process: Participant roles and their relations to social status within the group. Aggressive Behavior, 22(1), 1-15. https://doi.org/10.1002/(SICI)1098-2337(1996)22:1<1::AID-AB1>3.0.CO;2-T

Items of the Scale