Forms of Bullying Scale (FBS)

Abstract

The Forms of Bullying Scale (FBS) is a psychometrically validated, self-report assessment instrument designed to measure the behavioral manifestations, frequency, and dual dimensional structure of peer bullying during adolescence. Developed by Therese M. Shaw, Julian J. Dooley, Donna S. Cross, Stephen R. Zubrick, and Stacey K. Waters (2013), the instrument addresses longstanding measurement challenges in developmental psychopathology and school psychology by systematically disentangling bullying involvement into two distinct sub-instruments: the Forms of Bullying Scale – Victimization (FBS-V) and the Forms of Bullying Scale – Perpetration (FBS-P). Each sub-instrument comprises 10 parallel items scored on an authentic 5-point ordinal frequency scale ranging from 1 (“Never”) to 5 (“Most days”), anchored across a standard recall timeframe of the preceding school term.

Within each 10-item sub-instrument, the FBS measures two correlated yet conceptually independent behavioral dimensions: Direct Bullying (overt physical aggression, verbal harassment, property damage, and direct intimidation) and Indirect Bullying (relational manipulation, social exclusion, peer group alienation, secret disclosure, and malicious rumor spreading). Validated on a large longitudinal cohort of Australian secondary school students, the FBS demonstrates robust psychometric properties. Confirmatory factor analyses (CFA) support a two-factor correlated structure for both victimization and perpetration, outperforming unidimensional models across multiple developmental transitions. Internal consistency estimates are high across adolescent samples, with Cronbach’s alpha typically ranging from .82 to .89 for direct and indirect victimization and .78 to .87 for direct and indirect perpetration. Construct and convergent validity are evidenced by substantial associations with depressive symptomatology, generalized anxiety, academic disengagement, low school connectedness, and externalizing behavioral problems. The FBS provides researchers, educational administrators, and mental health clinicians with a refined, behavioral-based diagnostic tool that avoids arbitrary behavioral definitions and captures the nuanced topography of adolescent peer aggression.

Keywords

Forms of Bullying Scale, FBS, Bullying Victimization, Bullying Perpetration, Direct Bullying, Indirect Bullying, Relational Aggression, Adolescent Peer Victimization, Psychometrics, Factor Structure, School Psychology, Child Development

Authors

The Forms of Bullying Scale was developed and standardized by a multidisciplinary team of researchers affiliated with the Child Health Promotion Research Centre (CHPRC) at Edith Cowan University and the Telethon Kids Institute (formerly the Telethon Institute for Child Health Research) at the University of Western Australia:

• Therese M. Shaw, PhD – Biostatistician and Senior Research Fellow, Child Health Promotion Research Centre, School of Medical and Health Sciences, Edith Cowan University, Joondalup, Western Australia, Australia; and Telethon Kids Institute, University of Western Australia. Expert in longitudinal modeling, latent variable structural equation modeling, and psychometric evaluation in child health.
• Julian J. Dooley, PhD – Research Associate, Child Health Promotion Research Centre, Edith Cowan University, Western Australia; and Department of Psychology, Edith Cowan University. Specializes in developmental clinical psychology, youth cyberbullying, and coercive interpersonal dynamics.
• Donna S. Cross, EdD – Professor of Child and Adolescent Health, Child Health Promotion Research Centre, Edith Cowan University; and Telethon Kids Institute, University of Western Australia. Nationally and internationally recognized scholar in school-based bullying prevention interventions, adolescent mental health promotion, and the development of systemic anti-bullying initiatives (such as the Friendly Schools Project).
• Stephen R. Zubrick, PhD – Senior Principal Research Fellow, Telethon Kids Institute; and Centre for Child Health Research, University of Western Australia. Leading developmental epidemiologist specializing in population mental health, developmental trajectories, and life-course statistical methodologies.
• Stacey K. Waters, PhD – Research Fellow, Child Health Promotion Research Centre, Edith Cowan University. Focuses on developmental psychology, child mental health promotion, family-school partnerships, and health-related quality of life.

Correspondence regarding the original psychometric development of the scale was directed to Dr. Therese M. Shaw at the Child Health Promotion Research Centre, Edith Cowan University, 270 Joondalup Drive, Joondalup, WA 6027, Australia.

Purpose

The overarching purpose of the Forms of Bullying Scale (FBS) is to provide an empirically rigorous, behaviorally anchored, and structurally balanced assessment tool capable of quantifying both the experiences of being targeted (victimization) and engaging in harmful actions toward peers (perpetration). Peer bullying in childhood and adolescence represents a major public health and developmental concern linked with elevated risks of major depressive disorder, generalized anxiety disorder, self-harm, academic failure, and subsequent adult psychiatric morbidity. However, empirical inquiry into peer aggression has historically been complicated by marked variation in measurement approaches, inconsistent definitions of bullying, and over-reliance on single-item global screening questions.

Traditional assessment methodologies in bullying research frequently presented participants with complex, abstract definitions of bullying—such as Dan Olweus’ classic three-part criterion requiring intentional harm, repetition over time, and an observed or perceived imbalance of power—before asking global questions like, “How often have you been bullied?” Empirical investigations have demonstrated that this global-definition approach introduces profound cognitive biases. Children and adolescents often fail to interpret cognitive definitions consistently, struggle to accurately assess power imbalances, or avoid self-identifying as “bullies” or “victims” due to social desirability biases and the stigmatizing nature of these labels. In response, the FBS adopts a behavioral-inventory strategy. Rather than forcing youths to categorize themselves under clinical or legal labels, the scale presents specific, observable actions (e.g., “I was teased in nasty ways,” “Others tried to hurt me by leaving me out”) and measures the objective frequency with which these discrete events occurred within a specified reference timeframe.

From an applied perspective, the FBS serves several distinct clinical, educational, and research functions:

• Epidemiological Surveillance and School Climate Auditing: The scale enables educational institutions and public health agencies to systematically monitor the prevalence and distribution of bullying behaviors across diverse cohorts, school environments, and socioeconomic settings.
• Evaluating Targeted and Universal Interventions: Because the FBS differentiates between direct and indirect modalities of perpetration and victimization, program evaluators can determine whether whole-school interventions (such as social-emotional learning curricula or restorative justice practices) successfully reduce covert relational aggression as effectively as they mitigate overt physical bullying.
• Clinical and Psychoeducational Assessment: Mental health professionals working in child guidance clinics, school psychology practices, or adolescent psychiatric units can use the FBS-V and FBS-P profiles to identify the precise mechanisms of peer torment or aggressive behavior, thereby guiding functional behavioral assessments, cognitive-behavioral interventions, and safety planning.
• Disentangling Bully-Victim Subtypes: By administering both parallel versions, researchers can identify distinct relational subtypes: pure victims (high FBS-V, low FBS-P), pure perpetrators (low FBS-V, high FBS-P), bully-victims (high FBS-V, high FBS-P), and uninvolved youths (low on both), groups known to exhibit markedly divergent developmental etiologies and psychiatric prognoses.

Psychological Construct

The Forms of Bullying Scale is grounded in the operationalization of adolescent bullying as a multidimensional manifestation of interpersonal peer aggression. Drawing from the developmental theories of Dan Olweus, Kaj Björkqvist, Nicki R. Crick, and Kenneth A. Dodge, bullying is conceptualized as repeated, unprovoked aggressive acts executed within a social context characterized by an asymmetrical distribution of power. Rather than treating bullying as an undifferentiated, unitary phenomenon, the FBS conceptualizes bullying along two primary theoretical axes: the Participant Role Axis (victimization vs. perpetration) and the Behavioral Manifestation Axis (direct vs. indirect aggression).

Direct Bullying (FBS-V Direct & FBS-P Direct)

Direct bullying encompasses explicit, overt, and confrontational aggressive behaviors wherein the perpetrator directly faces the target. This dimension involves actions intended to inflict immediate physical suffering, tangible property loss, or overt psychological distress through face-to-face verbal assault. In the FBS, this construct is operationalized through five parallel items capturing:

• Physical Aggression: The intentional infliction of bodily injury, hitting, shoving, or coordinated collective assaults (e.g., being deliberately hurt physically or ganged up on).
• Overt Verbal Harassment: Malicious, non-consensual verbal interactions, including cruel teasing and malicious name-calling intended to demean, humiliate, or provoke the victim.
• Coercive Intimidation and Menace: Explicit verbal threats that induce fear regarding future physical or psychological harm.
• Property Destruction and Extortion: Deliberate theft, vandalism, or destruction of personal possessions to exert dominance and cause economic or emotional distress.

Direct forms of aggression are immediately observable by third-party peers and authority figures. They typically elicit overt behavioral responses (such as retaliation, physical flight, or teacher intervention) and have traditionally shown higher prevalence rates among adolescent males.

Indirect Bullying (FBS-V Indirect & FBS-P Indirect)

Indirect bullying, frequently termed relational aggression or social bullying, encompasses covert, insidious behaviors where the aggressive intent is mediated through the social architecture of the peer group. Rather than attacking the target’s physical person or tangible property, indirect aggression seeks to damage the victim’s social standing, peer relationships, self-concept, and sense of community belonging. In the FBS, indirect bullying is operationalized through five parallel items measuring:

• Social Exclusion and Ostracism: Systematically leaving individuals out of peer groups, ignoring their presence, or actively blocking them from participating in social activities.
• Relational Sabotage and Manipulation: Intentionally driving wedges between friends, orchestrating the dissolution of established friendships, or using conditional friendship as an instrument of control (e.g., stating one will not like the person unless they comply with demands).
• Reputational Defamation: The strategic dispersion of malicious lies, fabrications, and defamatory rumors designed to induce collective peer rejection.
• Breach of Intimacy: Weaponizing personal disclosures by broadcasting confidential secrets to others to generate public humiliation and social marginalization.

Indirect bullying is structurally disguised, allowing perpetrators to maintain plausible deniability while inflicting profound psychological harm. Because indirect bullying relies heavily on sophisticated linguistic processing, social intelligence, and complex peer network dynamics, it becomes increasingly prevalent during the transition into early and middle adolescence.

Theoretical Framework

The conceptual architecture of the Forms of Bullying Scale is situated at the intersection of several prominent theoretical frameworks in developmental psychology, social ecology, and social information processing.

Olweus’ Bully/Victim Paradigm and Behavioral Aggression Theory

The foundational bedrock of the FBS is derived from the pioneering work of Dan Olweus (1993, 1996), who defined bullying as proactive aggressive behavior characterized by intentionality, temporal repetition, and an asymmetry of physical, social, or psychological power. The FBS maintains Olweus’ emphasis on specific behavioral topography while refining measurement methodology. In accordance with behavioral aggression theories, the FBS recognizes that proactive aggression (goal-directed, instrumental behavior designed to acquire social dominance, status, or tangible resources) operates via different psychological mechanisms than reactive aggression (emotionally dysregulated, impulsive retaliation to perceived threats). By evaluating direct and indirect forms across both perpetrator and victim roles, the FBS aligns with Olweus’ structural postulations while eliminating subjective definitions that introduce measurement error.

Social Information Processing (SIP) Theory

The distinction between direct and indirect bullying within the FBS reflects Kenneth A. Dodge and Nicki R. Crick’s Social Information Processing Model (Crick & Dodge, 1994). According to SIP theory, behavioral responses in social settings are dictated through a multi-step cognitive sequence: encoding social cues, interpreting cues, clarifying goals, constructing potential behavioral responses, deciding on an action, and behavioral enactment. Direct bullying often reflects a deficit in the late decision-making stages, where aggressive youth anticipate that overt dominance will yield high social rewards with minimal cost. Conversely, indirect and relational aggression requires advanced mentalizing capacity (Theory of Mind), enabling perpetrators to map out social networks, anticipate interpersonal alliances, and subtly manipulate peer dynamics to maximize damage to the target while minimizing the risk of detection or retaliation.

Social-Ecological Systems Theory

The FBS is embedded within Urie Bronfenbrenner’s Social-Ecological Systems Framework, which posits that child development is shaped by interacting micro-, meso-, and macro-systems. Within this theoretical paradigm, bullying is not merely an isolated dyadic pathology between two individuals; it is an emergent property of social groups, peer ecologies, classroom dynamics, and broader school cultural norms. The FBS captures how peer groups function as dynamic arenas where direct aggression enforces physical hierarchies and indirect aggression enforces social stratification and in-group/out-group boundaries. This ecological perspective is critical for interpreting the psychometric divergence between direct and indirect subscales across different developmental stages and educational environments.

Validity

The psychometric evaluation of the Forms of Bullying Scale has yielded extensive evidence supporting its construct, convergent, discriminant, and criterion-related validity across adolescent cohorts (Shaw et al., 2013). Validation procedures conducted on large, population-representative samples of secondary school students have demonstrated the empirical robustness of the instrument.

Construct and Factorial Validity

Construct validity was established by comparing competing latent variable structural models. Structural equation modeling demonstrated that a correlated two-factor model—distinguishing Direct Bullying from Indirect Bullying—provided a superior statistical fit to the observed data compared to a unidimensional (single-factor) model for both victimization and perpetration across all measurement occasions. The correlation between the latent factors of Direct and Indirect Victimization is typically strong (ranging between r = .65 and .78), indicating shared variance consistent with the broader construct of peer victimization, while remaining sufficiently distinct to demonstrate that direct and indirect victimization represent non-redundant behavioral dimensions. Similarly, the latent correlation between Direct and Indirect Perpetration ranges between r = .60 and .72.

Convergent and Criterion Validity

Convergent validity has been established through strong, statistically significant correlations with established legacy instruments measuring peer relations and emotional distress:

• Olweus Bully/Victim Questionnaire (OBVQ): Scores on the FBS-V and FBS-P exhibit high correlations (r = .68 to .81) with corresponding global items and domain-specific subscales of the Revised Olweus Bully/Victim Questionnaire.
• Peer Relations Questionnaire (PRQ): Significant convergent associations were observed between the FBS subscales and the PRQ (Rigby, 1998), confirming that the FBS captures the broader spectrum of peer conflict and victimization.
• Internalizing Symptoms: In alignment with developmental psychopathology literature, FBS-V scores (both Direct and Indirect) demonstrate strong positive correlations with standardized measures of depressive symptoms (e.g., Center for Epidemiological Studies Depression Scale, CES-D) and anxiety inventories, with indirect victimization demonstrating particularly potent predictive utility for internalizing distress among adolescent females.
• Externalizing Behaviors and School Maladjustment: High scores on the FBS-P show strong convergent associations with conduct problems, antisocial behavioral tendencies, school disciplinary referrals, and substance experimentation.

Discriminant Validity

Discriminant validity is supported by the empirical dissociation between pure victimization and pure perpetration trajectories. Although bully-victims score high on both subscales, the FBS effectively discriminates between pure victims (who exhibit elevated internalizing symptoms, low self-esteem, and social withdrawal in the absence of conduct problems) and pure perpetrators (who exhibit high social dominance, high proactive aggression, and sensation-seeking traits in the absence of elevated victimization). Furthermore, multigroup invariant modeling has demonstrated that the scale differentiates between general peer conflict and chronic, systematic bullying involvement.

Reliability

The Forms of Bullying Scale demonstrates high internal consistency, item-scale metric stability, and test-retest reliability across multiple independent evaluations and longitudinal cohort waves.

Internal Consistency Reliability

In the primary psychometric validation study conducted by Shaw et al. (2013) involving over 3,000 Australian adolescents across Grades 8, 9, and 10, internal consistency was rigorously evaluated using Cronbach’s alpha (α) and composite reliability (ρ) coefficients:

• Victimization Version (FBS-V):
  • Total Victimization Scale (10 items): α = .88 to .91 across measurement waves.
  • Direct Victimization Subscale (5 items): α = .82 to .86.
  • Indirect Victimization Subscale (5 items): α = .83 to .87.
• Perpetration Version (FBS-P):
  • Total Perpetration Scale (10 items): α = .84 to .89 across measurement waves.
  • Direct Perpetration Subscale (5 items): α = .78 to .84.
  • Indirect Perpetration Subscale (5 items): α = .79 to .85.

All values comfortably surpass the standard .70 benchmark recommended for group-level research and the .80 threshold required for basic psychoeducational screening applications. The slightly lower alpha values observed for direct perpetration are consistent with epidemiological trends indicating that overt physical bullying has lower base rates in general adolescent populations.

Test-Retest Stability

Longitudinal stability coefficients across adjacent terms (interval of approximately 10 to 12 weeks) demonstrate moderate-to-high rank-order stability for both victimization (test-retest r ranging from .55 to .68) and perpetration (test-retest r ranging from .52 to .64). These values accurately mirror the theoretical nature of bullying: while peer victimization is moderately stable across developmental stages, it is also influenced by environmental changes, such as shifting classroom composition, peer group reconfiguration, and targeted school interventions.

Factor Analysis

The structural dimensionality of the Forms of Bullying Scale was evaluated through exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) within a structural equation modeling (SEM) framework, utilizing robust weighted least squares estimators (WLSMV) designed for ordinal categorical response distributions.

Confirmatory Factor Analysis (CFA) Specifications

Shaw et al. (2013) tested three competing theoretical models for both the FBS-V and the FBS-P across multiple adolescent developmental cohorts:

1. Model 1 (Unidimensional Model): All 10 items load onto a single global latent factor of bullying (Victimization or Perpetration).
2. Model 2 (Two-Factor Uncorrelated Model): Items load onto two orthogonal latent factors: Direct Bullying (5 items) and Indirect Bullying (5 items).
3. Model 3 (Two-Factor Correlated Model): Items load onto two oblique (correlated) latent factors: Direct Bullying and Indirect Bullying.

Across all longitudinal waves, Model 3 demonstrated superior fit, providing an excellent approximation of the empirical data while outperforming both the unidimensional and orthogonal structures. Standard goodness-of-fit indices consistently fell within accepted standards:

• Comparative Fit Index (CFI) ≥ .96
• Tucker-Lewis Index (TLI) ≥ .95
• Root Mean Square Error of Approximation (RMSEA) ≤ .045 (90% CI [.038, .052])
• Standardized Root Mean Square Residual (SRMR) ≤ .040

Standardized Factor Loadings

In the optimal correlated two-factor CFA solution, all standardized factor loadings were statistically significant (p < .001) and exhibited moderate-to-high magnitudes across both instruments:

• FBS-V Direct Victimization Factor: Standardized loadings ranged from .64 to .85. The strongest loading items were “I was teased in nasty ways” (λ ≈ .82) and “I was called names in nasty ways” (λ ≈ .85), followed by physical harm (λ ≈ .73) and property damage (λ ≈ .68).
• FBS-V Indirect Victimization Factor: Standardized loadings ranged from .68 to .84. The items “Others tried to hurt me by leaving me out, ignoring me or not letting me join in” (λ ≈ .84) and “Lies were told or nasty rumours were spread about me to hurt me” (λ ≈ .81) exhibited the highest construct centrality.
• FBS-P Direct Perpetration Factor: Standardized loadings ranged from .61 to .83. Direct verbal teasing and name-calling exhibited loadings above .78, whereas overt physical aggression loaded at approximately .69.
• FBS-P Indirect Perpetration Factor: Standardized loadings ranged from .65 to .82, with social exclusion and malicious rumor spreading demonstrating the strongest latent affinities.

Measurement invariance testing across sex (boys vs. girls) and across educational transitions (middle school to high school) confirmed full configural and metric invariance, and partial scalar invariance, verifying that the FBS measures identical behavioral constructs across demographic subgroups.

Instrument / Measurement Tool

The Forms of Bullying Scale is a self-administered, structured behavioral inventory. Below are the administrative and structural characteristics of the instrument:

• Test Type: Standardized self-report behavioral rating scale.
• Target Population: Adolescents and young people (typically aged 11 to 18 years; Grades 6 through 12).
• Administration Modality: Available in pencil-and-paper format or via digital assessment platforms. Can be completed individually or in group classroom administrations.
• Administration Time: Approximately 8 to 12 minutes to complete both the FBS-V and FBS-P sub-instruments.
• Total Item Count: 20 items total, divided into two balanced, parallel sub-instruments of 10 items each:
  • Forms of Bullying Scale – Victimization (FBS-V): 10 items.
  • Forms of Bullying Scale – Perpetration (FBS-P): 10 items.
• Recall Reference Timeframe: “Last term” (standardized to reflect the preceding academic school term, roughly an 8- to 12-week operational window).
• Authentic Response Scale: 5-point ordinal frequency scale:
  • 1 = Never
  • 2 = Once or twice
  • 3 = Every few weeks
  • 4 = About once a week
  • 5 = Most days
• Subscale Dimensional Composition: Within each 10-item instrument, items are partitioned into two distinct subscales:
  • Direct Bullying Subscale (5 items):
    • Victimization: Items 1, 4, 5, 6, 8
    • Perpetration: Items 11, 14, 15, 16, 18
  • Indirect Bullying Subscale (5 items):
    • Victimization: Items 2, 3, 7, 9, 10
    • Perpetration: Items 12, 13, 17, 19, 20
• Scoring Methodology and Rules:
  • Item Directionality: All items are scored positively in the direction of the measured construct. There are no reverse-scored items. Higher numerical values reflect higher frequencies of bullying involvement.
  • Subscale Scores: Direct and Indirect subscale scores are computed by summing or averaging the corresponding 5 items (theoretical sum range: 5 to 25; theoretical mean range: 1.0 to 5.0).
  • Total Dimension Scores: A total Victimization score (FBS-V) and total Perpetration score (FBS-P) can be derived by summing all 10 respective items (theoretical sum range: 10 to 50; mean range: 1.0 to 5.0).
  • Categorical Cut-Off / Threshold Operationalization: In accordance with epidemiological standards established by Olweus and adapted by Shaw et al. (2013), frequent or chronic involvement in bullying can be operationalized by identifying individuals who report experiencing or perpetrating any single form of bullying at a frequency of 3 (“Every few weeks”), 4 (“About once a week”), or 5 (“Most days”).

Permissions & Fee and Test Year

The Forms of Bullying Scale was formally published in 2013 in the American Psychological Association journal Psychological Assessment (Shaw et al., 2013). The instrument was developed within the Child Health Promotion Research Centre at Edith Cowan University in Western Australia as part of extensive empirical evaluations of school climate and adolescent well-being.

Licensing and Availability: The FBS is an open-access psychometric instrument available for academic, scientific, clinical, and non-commercial educational research. There are no licensing fees or purchase royalties required for academic use. The scale items, response options, and structural scoring instructions are published within the primary source article and through Edith Cowan University’s institutional research repository (ECU Publications Repository Record 743). Researchers, educators, and clinicians intending to administer, translate, or digitally host the FBS are expected to cite the original validation publication (Shaw et al., 2013) in all subsequent scientific reports and psychoeducational documentation. Commercial exploitation or incorporation into proprietary paid diagnostic platforms requires formal copyright permission from the lead authors and the publisher.

References

• Björkqvist, K., Lagerspetz, K. M., & Kaukiainen, A. (1992). Do girls manipulate and boys fight? Developmental trends in regard to direct and indirect aggression. Aggressive Behavior, 18(2), 117–127. https://doi.org/10.1037/0033-2909.115.1.74
• Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66(3), 710–722. https://doi.org/10.2307/1131945
• Olweus, D. (1993). Bullying at school: What we know and what we can do. Blackwell Publishers.
• Olweus, D. (1996). The Revised Olweus Bully/Victim Questionnaire. Research Centre for Health Promotion, University of Bergen.
• Rigby, K. (1998). Manual for the Peer Relations Questionnaire. The Professional Reading Guide for Educational Administrators.
• Shaw, T. M., Dooley, J. J., Cross, D. S., Zubrick, S. R., & Waters, S. K. (2013). The Forms of Bullying Scale (FBS): Validity and reliability estimates for a measure of bullying victimization and perpetration in adolescence. Psychological Assessment, 25(4), 1045–1057. https://doi.org/10.1037/a0032955

13. Items of the Scale (Questionnaire)