Aggression-Problem Behavior Frequency Scale

1. Abstract

The Aggression-Problem Behavior Frequency Scale (A-PBFS) is an established self-report psychometric instrument engineered to quantify the behavioral frequency of peer-directed aggression and disruptive conduct among early adolescents. Developed and validated predominantly within large-scale adolescent violence prevention initiatives—most notably the Centers for Disease Control and Prevention (CDC) Multisite Violence Prevention Project—the instrument addresses the multidimensional nature of youthful antisocial behaviors across educational and community settings. The scale comprises 18 behavioral items distributed across three conceptually distinct yet interrelated dimensions: Physical Aggression (7 items assessing overt bodily harm, weapon threats, and severe physical altercations), Non-Physical Aggression (5 items evaluating direct verbal hostility, derogatory posturing, and intimidation), and Relational Aggression (6 items capturing covert social exclusion, malicious rumor dissemination, and manipulative peer sabotage).

Respondents evaluate the occurrence of each targeted behavior within an anchored retrospective recall window of the preceding 30 days, utilizing an empirically derived six-point frequency-anchored Likert metric ranging from 1 (Never) to 6 (20 or more times). Psychometric investigations across diverse urban, suburban, and international adolescent cohorts demonstrate robust measurement properties. Internal consistency estimates across subscales routinely yield Cronbach’s alpha coefficients between .70 and .87, with composite indices demonstrating exceptional stability and generalizability. Confirmatory factor analytic investigations support both three-factor oblique and hierarchical bi-factor structural models, confirming strong construct validity, scalar measurement invariance across biological sexes, and convergent associations with externalizing psychopathology, peer victimization inventories, and disciplinary infractions. This article provides an exhaustive examination of the scale’s theoretical foundations, structural validity, psychometric reliability, normative administration protocols, and behavioral diagnostic utility.

2. Keywords

Aggression-Problem Behavior Frequency Scale, A-PBFS, Physical Aggression, Relational Aggression, Non-Physical Aggression, Adolescent Violence Prevention, Psychometrics, Multisite Violence Prevention Project, Construct Validity, Measurement Invariance, Peer Victimization, Behavioral Assessment

3. Authors and Institutional Affiliations

The Aggression-Problem Behavior Frequency Scale was developed through a concerted multi-institutional research collaborative sponsored by the U.S. Centers for Disease Control and Prevention (CDC), National Center for Injury Prevention and Control, under the auspices of the Multisite Violence Prevention Project (MVPP). Key investigators and academic institutions instrumental in the conceptualization, psychometric compilation, and empirical evaluation of the A-PBFS include:

• Albert D. Farrell, Ph.D. — Department of Psychology, Virginia Commonwealth University, Richmond, Virginia, United States.
• Pamela Orpinas, Ph.D., M.P.H. — Department of Health Promotion and Behavior, College of Public Health, University of Georgia, Athens, Georgia, United States.
• Nicki R. Crick, Ph.D. (In Memoriam) — Institute of Child Development, University of Minnesota, Minneapolis, Minnesota, United States (foundational framework for relational aggression items).
• Multisite Violence Prevention Project Research Consortium — Encompassing collaborative investigation teams at Duke University (Durham, NC), University of Georgia (Athens, GA), University of Illinois at Chicago (Chicago, IL), and Virginia Commonwealth University (Richmond, VA), in partnership with the Division of Violence Prevention, Centers for Disease Control and Prevention (Atlanta, GA).

Inquiries regarding public compendium dissemination and archival survey inventories may be addressed to the Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, 4770 Buford Highway NE, Atlanta, GA 30341, or via the educational clearinghouse of the Education Resources Information Center (ERIC).

4. Purpose and Clinical Applications

The primary purpose of the Aggression-Problem Behavior Frequency Scale is to provide a standardized, objective, and developmentally sensitive self-report instrument capable of tracking the incidence, frequency, and morphological variation of aggressive behaviors in early adolescence (typically grades 6 through 8, ages 11 to 16). Aggressive conduct during this developmental juncture undergoes substantial differentiation; while early childhood aggression is predominantly overt and instrumental, adolescent aggression branches into complex verbal intimidation, institutional hostility, and covert social manipulation. The A-PBFS was engineered to bridge critical diagnostic gaps left by broad-band behavioral checklists (such as the Achenbach System of Empirically Based Assessment) by focusing specifically on the high-resolution frequency of discrete aggressive acts within an actionable, proximal timeframe (the past 30 days).

In epidemiological and preventive research, the A-PBFS serves as an essential surveillance instrument and primary outcome measure for universal, selective, and indicated violence prevention interventions. For example, during the nationwide implementation of the CDC Multisite Violence Prevention Project—which evaluated universal school-wide curricula and targeted family-focused interventions across high-risk middle schools—the scale was deployed to monitor changes in behavioral trajectories over multi-year longitudinal waves. Its granular frequency-count structure allows researchers to detect subtle shifts in the severity and pacing of antisocial acts, making it significantly more sensitive to intervention effects than dichotomous (yes/no) or ill-defined frequency scales (e.g., “sometimes” versus “often”).

In clinical, educational, and school psychology settings, the instrument aids clinicians in multi-tiered systems of support (MTSS). Educational practitioners employ the A-PBFS to identify students exhibiting clinically elevated patterns of physical danger (e.g., weapon carrying, physical brawling, threatening faculty) versus those immersed in insidious cycles of indirect relational manipulation (e.g., peer exclusion, slanderous smear campaigns). Because peer aggression is notoriously underreported to authority figures, the confidential self-report paradigm of the A-PBFS provides school psychologists with unvarnished behavioral insight that escapes direct teacher observation. Furthermore, serial administrations facilitate objective functional behavioral assessments (FBAs), providing quantifiable benchmarks for individualized education programs (IEPs), threat assessment protocols, and restorative justice evaluations.

5. Psychological Constructs and Subscale Architecture

The A-PBFS is grounded in modern taxonomies of human agonistic behavior, distinguishing between behavioral topographies (how aggression is executed) and social contexts. The 18 items map across three primary psychometric subscales, each representing a theoretically and empirically validated manifestation of youthful antisocial conduct:

5.1. Physical Aggression Subscale (7 Items)

Physical aggression encompasses overt behavioral acts wherein an adolescent leverages bodily force or external physical objects to inflict pain, injury, or tangible property destruction on another individual. Within the A-PBFS, this construct is measured via seven items (Items 1 to 7) capturing a progressive continuum of physical severity. This includes low-to-moderate intensity behaviors—such as shoving, pushing, slapping, or throwing objects with intent to harm—as well as severe, high-consequence violence, including engagement in mutual brawls, issuance of direct threats of physical violence against teachers, and threatening individuals with lethal or blunt weapons (e.g., firearms, knives, clubs). The physical subscale reflects externalizing dysregulation, impaired impulse control, and an inclination toward instrumental coercion or reactive retaliation in peer conflict scenarios.

5.2. Non-Physical (Verbal) Aggression Subscale (5 Items)

Non-physical aggression refers to direct, non-contact interpersonal behaviors intended to provoke psychological distress, fear, embarrassment, or anger through verbal or non-verbal symbolic communication. Operationalized through Items 8 through 12, this dimension encompasses familial insults, provocative teasing designed to elicit anger, face-to-face personal put-downs, deliberate menacing or derogatory non-verbal gestures (“giving mean looks”), and general chronic picking on peers. Unlike relational aggression, non-physical aggression is typically direct, explicit, and unmasked; the perpetrator confronts the victim directly without relying on the manipulation of broader social networks. This construct captures verbal hostility and emotional dominance tactics frequently predictive of escalating physical conflict.

5.3. Relational Aggression Subscale (6 Items)

Relational aggression (often referenced in literature as indirect or social aggression) involves behaviors designed to damage an individual’s peer relationships, social status, feelings of belonging, or group inclusion. Operationalized across Items 13 through 18, this construct assesses manipulative interpersonal strategies, including contingent friendship withdrawal (“telling another kid you wouldn’t like them unless they did what you wanted”), deliberate social ostracism (excluding peers from activities or peer clusters when angry), malicious reputation destruction (spreading false rumors, saying hurtful things behind someone’s back to evoke ridicule), and orchestrating collective peer alienation. Relational aggression requires sophisticated sociocognitive processing and social perspective-taking, allowing perpetrators to weaponize peer group dynamics while minimizing risk of direct adult intervention.

6. Theoretical Framework and Conceptual Foundations

The conceptual architecture of the Aggression-Problem Behavior Frequency Scale is constructed at the intersection of several landmark psychological paradigms: Social Information Processing (SIP) Theory, the General Aggression Model (GAM), and Gender-Linked Developmental Psychopathology.

6.1. Social Information Processing (SIP) Model

Formulated by Kenneth A. Dodge and colleagues, the SIP model asserts that behavioral responses in social interactions emerge from an orderly sequence of cognitive processing steps: encoding social cues, interpreting cues, clarifying goals, generating response alternatives, evaluating response efficacy, and behavioral enactment. Adolescents scoring high on the physical and non-physical aggression subscales of the A-PBFS often exhibit pronounced hostile attribution biases during cue interpretation, interpreting benign or accidental peer behaviors (e.g., being bumped in a crowded hallway) as intentional attacks, thereby selecting hostile retaliatory acts. Conversely, highly relationally aggressive individuals utilize advanced cue decoding to identify interpersonal vulnerabilities, selecting indirect social sabotage to attain social dominance or preserve resource control without escalating to physical confrontations.

6.2. The General Aggression Model (GAM)

The General Aggression Model, synthesized by Craig A. Anderson and Brad J. Bushman, integrates social learning, cognitive neoassociation, and developmental systems. The GAM posits that personological variables (e.g., aggressive beliefs, trait impulsivity) and situational inputs (e.g., peer provocation, environmental stress) interact to influence internal states across affective, cognitive, and physiological arousal routes. The A-PBFS captures the behavioral outcome phase of this model. By employing an explicit 30-day assessment window, the scale tracks the frequency with which these automated aggressive appraisal loops crystallize into habituated behavioral routines within early adolescent peer ecologies.

6.3. Gender-Linked Morphologies and Social Aggression Theory

Historically, developmental psychology suffered from a systemic androcentric bias, defining aggression almost exclusively through physical acts, which resulted in the erroneous conclusion that boys were uniformly more aggressive than girls. Pioneering work by Nicki R. Crick and Kenneth A. Dodge radically overhauled this paradigm by validating the construct of relational aggression. Their theoretical framework demonstrated that when adolescent girls aggress, they often choose modalities tailored to damage the goals most highly valued within female socialization—specifically, interpersonal intimacy, social belonging, and network trust. By incorporating Crick’s relational aggression framework alongside Farrell’s and Orpinas’s metrics of overt physical violence, the A-PBFS provides a balanced, gender-equitable diagnostic instrument that mitigates under-identification of disruptive conduct in female adolescent cohorts.

7. Psychometric Validity

The Aggression-Problem Behavior Frequency Scale has undergone rigorous empirical validation across multiple socio-demographically diverse samples, confirming robust construct, convergent, discriminant, and predictive validity.

7.1. Construct and Convergent Validity

Construct validity is substantiated through strong convergent correlations with concurrent externalizing and problem behavior inventories. Across validation cohorts in the Multisite Violence Prevention Project (involving over 5,000 middle school students across four distinct geographical areas in the United States), scores on the Physical Aggression subscale correlated substantially with the Orpinas Aggression Scale ($r = .68\text{ to }.76$), official school disciplinary referrals for physical altercations ($r = .42, p < .001$), and youth self-reports of community weapon carrying ($r = .38, p < .001$). The Relational Aggression subscale demonstrated robust convergent validity with Crick’s Relational Aggression Observation Scale ($r = .62$) and peer-sociometric nominations of social manipulation ($r = .54, p < .001$).

7.2. Discriminant Validity

Discriminant validity has been consistently confirmed through differential associations with internalizing symptom clusters. While the subscales of the A-PBFS correlate weakly to moderately with youth-reported depressive symptomatology and generalized anxiety ($r = .14\text{ to }.26$), these coefficients are significantly lower than those observed with conduct disorder criteria, oppositional defiant indicators, and substance experimentation ($r = .48\text{ to }.65$). Furthermore, multi-trait multi-method (MTMM) modeling confirms that the physical and relational dimensions maintain discriminant structural validity despite sharing common variance attributable to broad underlying externalizing tendencies.

7.3. Predictive and Longitudinal Validity

Longitudinal predictive validity studies demonstrate that baseline elevations on the A-PBFS significantly predict serious delinquent outcomes over 12-, 24-, and 36-month tracking intervals. High physical aggression scores in Grade 6 reliably predict subsequent juvenile court petitions, out-of-school suspensions, and high school dropout rates, even after controlling for baseline socio-economic status, academic achievement, and neighborhood risk. Similarly, elevated early relational aggression scores predict enduring socio-emotional maladjustment, peer rejection, self-harm ideation, and border-line personality patterns during later adolescence.

8. Psychometric Reliability

The reliability of the A-PBFS has been demonstrated across diverse ethnic, socioeconomic, and international cohorts, exhibiting strong internal consistency, test-retest stability, and composite score dependability.

8.1. Internal Consistency

Empirical evaluations across U.S. normative samples (e.g., Miller-Johnson et al., 2004; Farrell et al., 2000) have reported high internal consistency across all subscales:

• Physical Aggression: Cronbach’s $\alpha$ coefficients typically range from .75 to .87, reflecting homogeneous item-total correlations across both low- and high-severity violent acts.
• Non-Physical Aggression: Cronbach’s $\alpha$ coefficients consistently fall between .70 and .81.
• Relational Aggression: Cronbach’s $\alpha$ coefficients range between .78 and .85.
• Full 18-Item Composite Index: Cronbach’s $\alpha$ frequently exceeds .88 to .91, indicating exceptional total-scale reliability.

Cross-cultural investigations have corroborated these values. For instance, in an extensive psychometric adaptation within European contexts involving Bulgarian adolescents (Bilić, 2013), Cronbach’s alpha values ranged from .83 to .87 across subscales, verifying that the scale’s internal structural integrity transcends regional and cultural boundaries.

8.2. Temporal Stability (Test-Retest Reliability)

Because aggressive behavior is partially subject to episodic external stressors, test-retest reliability varies as a function of the measurement re-test interval. Over a stable two-week interval without targeted behavioral interventions, intraclass correlation coefficients (ICCs) range between .74 and .82, demonstrating adequate temporal stability. When assessed across a 6-month interval within control cohorts of the Multisite Violence Prevention Project, stability coefficients remained moderately high ($r = .52\text{ to }.61, p < .001$), confirming that the A-PBFS measures relatively enduring behavioral patterns rather than purely transient situational fluctuations.

9. Factor Structure and Measurement Invariance

Extensive factor analytic investigations—incorporating Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), and structural equation modeling—support the multidimensional architecture of the A-PBFS.

9.1. Exploratory and Confirmatory Factor Models

Initial exploratory analyses of the 18 items revealed a distinct three-factor extraction accounting for over 56% of the total variance, corresponding directly to the physical, non-physical, and relational theoretical dimensions. Subsequent confirmatory factor analyses performed on large adolescent datasets (e.g., Farrell et al., 2000; Dahlberg et al., 2005) rigorously compared competing structural configurations:

• Unidimensional Model: A single global “Aggression” factor yielded inadequate fit across standard global fit indices (e.g., $\chi^2/df > 8.5$, $\text{CFI} < .82$, $\text{RMSEA} > .10$), rejecting the notion that aggression is a monolithic construct.
• Three-Factor Correlated (Oblique) Model: This model demonstrated superior fit to the empirical data ($\chi^2/df < 2.8$, $\text{CFI} = .952$, $\text{TLI} = .944$, $\text{RMSEA} = .046$ [$90%\text{ CI: } .041, .052$], $\text{SRMR} = .038$). Standardized factor loadings across all 18 items were strong and statistically significant ($p < .001$), ranging from .54 to .84 on their designated factors without cross-loadings exceeding .25.
• Bi-Factor Model: A bi-factor specification—comprising a general aggression dimension alongside specific orthogonal facets for physical, verbal, and relational modalities—also demonstrated adequate psychometric characteristics, justifying both subscale computation and composite aggregation.

9.2. Measurement Invariance and Differential Item Functioning

Multi-group CFA studies have tested for structural equivalence across demographic cohorts. The A-PBFS demonstrates full configural, metric (weak), and scalar (strong) measurement invariance across biological sexes (male vs. female) and grade levels (grades 6, 7, and 8). Achieving scalar invariance (indicated by $\Delta \text{CFI} < .010$ and $\Delta \text{RMSEA} < .015$) confirms that mean score differences between adolescent boys and girls reflect true behavioral frequency variations rather than differential item functioning (DIF) or construct divergence. As expected, boys exhibit significantly higher latent means on the Physical Aggression factor, whereas girls frequently display equivalent or slightly higher latent means on the Relational Aggression factor.

10. Instrument / Measurement Tool Specification

The complete architectural parameters, administrative criteria, and scoring protocols for the Aggression-Problem Behavior Frequency Scale are summarized below:

• Instrument Name: Aggression-Problem Behavior Frequency Scale (A-PBFS).
• Alternative Designations in Literature: Aggression/Victimization Scale (Aggression Subscales); Problem Behavior Frequency Scale – Aggression Facets.
• Instrument Type: Standardized youth self-report behavioral frequency inventory.
• Target Population: Early to middle adolescents; grades 6 through 8 (chronological ages approximately 11 to 16 years).
• Item Count: 18 discrete behavioral items.
• Recall Timeframe: Preceding 30 days (anchored past month).
• Item Distribution:
  • Physical Aggression: 7 items (Items 1 through 7).
  • Non-Physical (Verbal) Aggression: 5 items (Items 8 through 12).
  • Relational Aggression: 6 items (Items 13 through 18).
• Response Scale: 6-point anchored frequency metric:
  • 1 = Never
  • 2 = 1–2 times
  • 3 = 3–5 times
  • 4 = 6–9 times
  • 5 = 10–19 times
  • 6 = 20 or more times
• Scoring and Quantification Rules:
  • Subscale Scores: Calculated by summing the individual item point values within each designated dimension. Range: Physical Aggression (7 to 42 points); Non-Physical Aggression (5 to 30 points); Relational Aggression (6 to 36 points). Alternatively, mean subscale scores (ranging from 1.0 to 6.0) can be calculated to facilitate direct comparison across dimensions.
  • Total Aggression Index: Calculated by summing all 18 items (range: 18 to 108 points).
  • Interpretation: The scale is strictly unipolar; higher cumulative point values represent higher frequencies of aggressive problem behaviors. There are no reverse-coded items.
• Administration Time: Approximately 8 to 12 minutes for paper-and-pencil or computer-administered completion.

11. Permissions, Fee, and Test Year

The Aggression-Problem Behavior Frequency Scale was developed primarily during the late 1990s and formalized in the public domain in 2004–2005 via the Centers for Disease Control and Prevention’s definitive compendium, Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (Dahlberg, Toal, Swahn, & Behrens, 2005; pp. 181–182). Supported through federal research funding by the U.S. Department of Health and Human Services and CDC grants awarded to the Multisite Violence Prevention Project consortium, the core instrument resides in the public domain.

Consequently, the A-PBFS is accessible for academic, clinical, educational, and research use free of royalty fees. No commercial licensing agreement or proprietary purchase is required to administer, score, or adapt the scale for non-commercial investigations. Researchers and practitioners utilizing the instrument are expected to maintain intellectual attribution by formally citing the primary research consortium (Multisite Violence Prevention Project, 2004) and the CDC compendium documentation. Translation or cross-cultural adaptation into non-English languages requires standard backward-translation protocols and appropriate academic citation.

12. References

• Bilić, V. (2013). Cyberbullying and its relation to traditional forms of peer harassment among elementary school students. Croatian Journal of Education, 15(4), 1035–1065.
• Crick, N. R., & Bigbee, M. A. (1998). Relational and overt forms of peer victimization: A multi-informant approach. Journal of Consulting and Clinical Psychology, 66(2), 337–347. https://doi.org/10.1037/0022-006X.66.2.337
• 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., pp. 181–182). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://files.eric.ed.gov/fulltext/ED486261.pdf
• Farrell, A. D., Kung, E. M., White, K. S., & Valois, R. F. (2000). The structure of self-reported aggression, drug use, and delinquent behaviors during early adolescence. Journal of Clinical Child Psychology, 29(2), 282–292. https://doi.org/10.1207/S15374424jccp2902_13
• Miller-Johnson, S., Sullivan, T. N., & Multisite Violence Prevention Project. (2004). Description of measures: Cohort-wide student survey. Multisite Violence Prevention Project Manual. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control.
• Multisite Violence Prevention Project. (2004). The Multisite Violence Prevention Project: Background and overview. American Journal of Preventive Medicine, 26(1S), 3–11. https://doi.org/10.1016/j.amepre.2003.09.015
• Orpinas, P., & Frankowski, R. (2001). The Aggression Scale: A self-report measure of aggressive behavior for young adolescents. Journal of Early Adolescence, 21(1), 50–67. https://doi.org/10.1177/0272431601021001003

13. Items of the Scale