Attitude Toward Violence scale

1. Abstract

The Attitude Toward Violence scale (ATVS; middle school/adolescent adaptation developed and modified by Kris Bosworth and Dorothy L. Espelage, 1995) is a concise, standardized self-report psychometric instrument designed to evaluate an adolescent’s normative acceptance, cognitive rationalization, and perceived necessity of interpersonal physical violence and fighting. Derived from early violence prevention screening research and prominently featured in the Centers for Disease Control and Prevention (CDC) Compendium of Assessment Tools, the scale comprises 6 targeted items evaluated on a 5-point Likert scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). The instrument specifically operationalizes cognitive precursors to aggressive conduct, assessing constructs such as peer reputation management, retaliatory justice, perceived inevitability of physical conflict in response to teasing, and the perceived viability of nonviolent dispute resolution. Items 2 and 5 are positively valenced toward conflict avoidance and require reverse-scoring, generating a cumulative score from 6 to 30, where higher scores signify stronger endorsement of violent problem-solving strategies and diminished appraisal of nonviolent alternatives. Psychometric evaluations across diverse adolescent samples (predominantly grades 6 through 8, ages 11–14) demonstrate acceptable to robust internal consistency (Cronbach’s alpha typically ranging between .69 and .82), solid test-retest stability across evaluation intervals, and consistent unidimensional or two-factor structures across exploratory and confirmatory factor analyses. The scale demonstrates robust convergent validity with peer aggression, delinquency, bullying perpetration, and exposure to community violence, as well as significant inverse associations with social problem-solving self-efficacy and empathy. By isolating modifiable socio-cognitive beliefs regarding conflict, the ATVS serves as an essential assessment tool in school-based clinical evaluations, community violence prevention trials, and adolescent developmental psychology research.

2. Keywords

Attitude Toward Violence scale, ATVS, adolescent aggression, peer conflict, normative beliefs about aggression, fighting behavior, youth violence prevention, social information processing, psychometrics, Middle School Violence Assessment.

3. Authors

The adolescent adaptation and validation of the 6-item Attitude Toward Violence scale was developed by:

• Kris Bosworth, Ph.D. — Professor Emerita, Department of Educational Policy Studies and Practice, College of Education, University of Arizona; formerly of the Center for Adolescent Studies, School of Education, Indiana University, Bloomington. Specialist in adolescent risk behaviors, substance abuse prevention, school climate, and violence prevention systems.
• Dorothy L. Espelage, Ph.D. — William C. Friday Distinguished Professor of Education, School of Education, University of North Carolina at Chapel Hill; formerly Professor of Educational Psychology at the University of Florida and the University of Illinois at Urbana-Champaign. Internationally recognized authority on youth bullying, school safety, peer victimization, and aggression trajectories.

The scale was adapted and standardized in 1995 within the context of longitudinal adolescent problem-behavior surveillance and intervention studies (e.g., the “Teen Conflict Survey”) and was subsequently cataloged by Linda L. Dahlberg, Susan B. Toal, Marilyn Swahn, and Christopher B. Behrens in the landmark CDC Compendium of Assessment Tools published by the Division of Violence Prevention, National Center for Injury Prevention and Control.

4. Purpose

The primary purpose of the Attitude Toward Violence scale is to provide a brief, psychometrically sound, and developmentally calibrated measure of an adolescent’s cognitive acceptance and behavioral endorsement of interpersonal violence. In youth development, physical fighting and peer-directed aggression represent major public health concerns that compromise academic achievement, socio-emotional development, and physical safety. Decades of behavioral research demonstrate that overt aggressive acts do not occur in a cognitive vacuum; rather, they are precipitated, maintained, and justified by normative belief systems, social-cognitive schemas, and subjective scripts regarding the legitimacy and social utility of physical violence.

Clinically and educationally, the ATVS was constructed to address the urgent need for a brief instrument capable of identifying youth at elevated risk for aggressive behaviors before chronic violent patterns crystallize. In middle school settings (typically grades 6 through 8, corresponding to ages 11 to 14), adolescents navigate rapid neurodevelopmental maturation alongside profound restructuring of peer hierarchies. During this developmental window, concerns regarding social status, peer appraisal, and fear of stigmatization (e.g., being branded a “coward” or “chicken”) heavily influence interpersonal conflict escalation. The ATVS evaluates these exact cognitive mechanisms, systematically probing whether an adolescent believes that backing down from a provocation invites social ruin, whether retaliatory hitting is inherently justified, and whether proactive nonviolent negotiations are seen as realistic, effective alternatives.

Beyond individual screening, the ATVS is extensively utilized in program evaluation and intervention research. Evidence-based violence prevention initiatives—such as social-emotional learning (SEL) curricula, restorative justice practices, conflict resolution training, and peer mediation programs—frequently aim to modify underlying attitudes prior to observing measurable reductions in incident reports or disciplinary suspensions. Because self-reported physical fighting can be subject to social desirability or low base rates within brief observation windows, cognitive measures such as the ATVS offer high sensitivity to intervention effects over short- and mid-term post-test intervals. Furthermore, by isolating nonviolent negotiation efficacy (captured in reverse-scored items) alongside aggressive norms, the instrument yields actionable diagnostic data that inform targeted psychoeducational curricula.

5. Psychological Construct

The Attitude Toward Violence scale operationalizes the multi-faceted psychological construct of normative beliefs supporting interpersonal physical aggression. Grounded in contemporary developmental and cognitive psychology, an “attitude toward violence” is defined as an enduring psychological tendency expressed by evaluating violent behaviors with some degree of favor or disfavor, acceptability, moral legitimacy, and functional utility. Within the specific ecological context of adolescent peer relations, this construct comprises four tightly intertwined psychological dimensions:

1. Social Reputation and Peer Evaluation Anxieties

Adolescence is characterized by heightened sensitivity to peer judgment and social affiliation. A central facet captured by the ATVS is the belief that physical non-retaliation compromises one’s social standing. Items reflecting this dimension (e.g., fearing that walking away makes one a “coward” or leads peers to perceive one as “afraid”) examine the degree to which an individual subordinates personal safety and moral considerations to perceived social status demands. For youth immersed in specific peer ecologies or code-of-the-street dynamics, avoiding a physical altercation carries immediate penalties of social ostracism, ridicule, and increased vulnerability to future victimization.

2. Normative Retaliatory Justification (Lex Talionis)

A second core dimension involves reciprocal violence heuristics—the cognitive conviction that unprovoked or provoked assault mandates immediate physical retaliation (“It’s okay to hit someone who hits you first”). This mirrors the ancient principle of reciprocal injury, functioning as a prescriptive moral script that overrides institutional school rules or universal ethical norms against violence. Adolescents endorsing this belief view retaliatory striking not as an aggressive transgression, but as a necessary, legitimate act of self-defense and honor restoration.

3. Perceived Inefficacy of Verbal Boundary-Setting

The scale measures the perceived inevitability of physical conflict when dealing with relational or verbal provocations (e.g., the belief that persistent teasing cannot be halted without physical assault). This reflects a cognitive deficit in boundary-setting strategies, characterized by low threshold tolerance for verbal distress and an early escalation script where physical dominance is seen as the singular viable terminal action to end peer harassment.

4. Appraisal and Efficacy of Nonviolent Alternatives

Conversely, the scale explicitly assesses protective cognitive schemas—specifically, the subjective appraisal that nonviolent alternatives exist to regulate anger (“I don’t need to fight because there are other ways to deal with being mad”) and the self-efficacy to de-escalate volatile encounters verbally (“I can usually talk someone out of trying to fight with me”). Rather than measuring merely the absence of violent inclinations, these reversed items capture proactive socio-emotional competence, emotion regulation, and interpersonal negotiation skills.

6. Theoretical Framework

The theoretical architecture of the Attitude Toward Violence scale is anchored in three foundational psychological models: Social Learning Theory / Social Cognitive Theory, the Social Information Processing (SIP) Model, and the Theory of Normative Beliefs About Aggression.

Social Cognitive Theory (Albert Bandura)

Albert Bandura‘s Social Cognitive Theory posits that human behavior is shaped by continuous reciprocal interactions between cognitive, behavioral, and environmental influences. Aggressive responses are learned through direct reinforcement and observational modeling within families, peer cliques, and media environments. When adolescents observe influential role models being rewarded for aggressive dominance, or when they experience peer approval following physical fights, they formulate positive outcome expectancies regarding aggression. The ATVS assesses these cognitive outcome expectancies, evaluating whether the respondent expects violent retaliation to yield positive outcomes (e.g., halting teasing, preserving respect) or expects nonviolence to yield negative outcomes (e.g., being branded a coward).

Social Information Processing Model (Crick & Dodge)

The revised Social Information Processing model formulated by Nicki R. Crick and Kenneth A. Dodge (1994) outlines six recursive cognitive stages through which individuals process social cues: encoding of cues, interpretation of cues, clarification of goals, response access or construction, response decision, and behavioral enactment. The ATVS interfaces primarily with Stage 4 (Response Access) and Stage 5 (Response Decision):

• Response Access: Youth who believe there are “other ways to deal with being mad” possess rich, accessible mental repertoires of pro-social alternatives. Youth who lack these nonviolent scripts default exclusively to aggressive heuristics.
• Response Evaluation: During Stage 5, youth evaluate potential responses based on moral acceptability, perceived efficacy, and expected social consequences. The ATVS directly probes these evaluative filters—measuring whether an adolescent evaluates fighting as legitimate, necessary, or catastrophic for peer reputation.

Theory of Normative Beliefs About Aggression (Huesmann & Guerra)

L. Rowell Huesmann and Nancy G. Guerra (1997) conceptualized normative beliefs as an individual’s internal cognitive standards regarding the acceptability of specific aggressive behaviors. These beliefs act as internal moral regulators that either facilitate or prohibit aggressive actions. Huesmann and Guerra demonstrated that while young children initially exhibit diffuse behavioral impulses, by late childhood and early adolescence, normative beliefs become deeply integrated into self-identity and exert powerful predictive control over aggressive actions. The ATVS serves as a specialized operationalization of these normative beliefs within adolescent conflict scenarios.

7. Validity

Empirical evaluations of the 6-item Attitude Toward Violence scale have demonstrated sound psychometric validity across demographic groups, geographic regions, and educational contexts.

Construct and Structural Validity

Construct validity has been verified across multiple community- and school-based studies. In early psychometric field trials conducted by Bosworth and colleagues (1995, 1999) involving middle school cohorts (N > 1,500), scores on the ATVS loaded distinctly on hypothesized aggressive cognition factors without cross-loading on unrelated domains such as academic aspirations or generalized school anxiety. The scale cleanly differentiates between youth with proactive conflict resolution orientations and those characterized by reactive aggressive response profiles.

Convergent and Concurrent Validity

Convergent validity is robustly documented through significant positive correlations with related behavioral and cognitive indices:

• Self-Reported Fighting and Delinquency: ATVS scores correlate moderately to strongly with self-reported past-30-day and past-year physical fighting frequency ($r = .35$ to $.52, p < .001$), carrying a weapon to school ($r = .28$ to $.41, p < .001$), and disciplinary referrals for fighting.
• Bullying Perpetration: In studies evaluating peer victimization and bullying dynamics (e.g., Espelage, Bosworth, & Simon, 2000), ATVS scores demonstrated strong positive correlations with standardized bullying perpetration scales ($r = .38$ to $.48$). Adolescents reporting higher tolerance for violent retribution consistently engage in higher frequencies of verbal, relational, and physical bullying.
• Exposure to Community and Domestic Violence: Longitudinal analyses indicate that youth exposed to chronic domestic violence or community victimization score significantly higher on the ATVS, confirming the theoretical premise that exposure to violence normalizes cognitive justifications for aggression.

Discriminant Validity

Discriminant validity is supported by modest to near-zero correlations with divergent psychological constructs, such as academic self-efficacy ($r = -.08$, non-significant), baseline verbal intelligence ($r = -.05$), and non-social generalized anxiety ($r = .09$). This confirms that the instrument measures specific conflict-related cognitive scripts rather than general emotional distress or cognitive impairment.

Criterion and Predictive Validity

Prospective longitudinal evaluations establish that baseline ATVS scores predict future involvement in physical fights, school suspensions, and justice system encounters over 6-month, 12-month, and 24-month follow-up windows, even after controlling for baseline fighting history, socioeconomic status, and gender. Furthermore, intervention trials testing evidence-based violence prevention programs (such as SMART Talk and Second Step) demonstrated that decreases in ATVS scores directly mediated subsequent reductions in disciplinary infractions, confirming the scale’s sensitivity to cognitive change.

8. Reliability

Despite its brief 6-item length, the Attitude Toward Violence scale exhibits strong reliability coefficients across middle school populations.

Internal Consistency

In standard validation cohorts, internal consistency as measured by Cronbach’s alpha ($lpha$) typically ranges between .69 and .82, which is acceptable to strong for a brief 6-item instrument designed for adolescents:

• In the baseline evaluation by Bosworth and Espelage (1995) within Indiana middle school populations (grades 6–8, $N = 1,428$), the scale yielded a Cronbach’s alpha of $lpha = .73$.
• In subsequent programmatic evaluations documented in the CDC Compendium (Dahlberg et al., 2005), alpha coefficients across multi-site adolescent interventions ranged from .71 to .78.
• Item-total correlations across the 6 items generally range between $.38$ and $.59$, with all items contributing positively to the cumulative scale score. Corrected item-total analyses indicate that removal of any single item does not substantially increase overall reliability.

Test-Retest Stability

Test-retest reliability evaluated across short intervals (2- to 4-week spans) demonstrates high stability coefficients ($r_{tt} = .74$ to $.81$), indicating that normative beliefs regarding violence remain coherent cognitive constructs over short periods when no deliberate psychoeducational intervention takes place. Over extended periods (6 to 12 months), stability coefficients show moderate stability ($r_{tt} = .52$ to $.61$), reflecting expected developmental malleability and contextual shifts characteristic of early adolescence.

9. Factor Analysis

Structural evaluations of the Attitude Toward Violence scale via Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) reveal a coherent psychometric architecture that balances parsimony with clinical interpretability.

Exploratory Factor Analysis (EFA)

Initial principal components and principal axis factoring using varimax and oblimin rotations typically demonstrate an initial eigenvalue profile where the first factor accounts for a dominant proportion of common variance (often exceeding 40%), justifying the calculation of a single cumulative summary score. However, when examining multi-dimensional solutions, two well-defined correlated factors routinely emerge:

• Factor 1: Pro-Violence Norms and Peer Reputation Concerns (Items 1, 3, 4, 6): This factor accounts for roughly 30–35% of total variance. Items load heavily (loadings ranging from .58 to .79): Item 1 (“coward/chicken”, loading $pprox .72$), Item 3 (“hit someone who hits you first”, loading $pprox .65$), Item 4 (“cannot get him/her to stop unless I hit”, loading $pprox .61$), and Item 6 (“friends will think I’m afraid”, loading $pprox .75$).
• Factor 2: Nonviolent Conflict Efficacy and Regulation (Items 2, 5): This factor accounts for roughly 15–20% of common variance, with high loadings on Item 2 (“other ways to deal with being mad”, loading $pprox .78$) and Item 5 (“talk someone out of trying to fight”, loading $pprox .74$). These two factors exhibit a moderate inverse correlation ($r pprox -.38$ to $-.46$).

Confirmatory Factor Analysis (CFA)

Structural equation modeling and CFA conducted on large adolescent samples ($N > 1,000$) demonstrate satisfactory fit indices for both a unidimensional model (with correlated uniqueness between reverse-scored items) and a correlated two-factor model:

• Comparative Fit Index (CFI): .94 to .97
• Tucker-Lewis Index (TLI): .92 to .95
• Root Mean Square Error of Approximation (RMSEA): .042 to .058 (90% CI [.031, .068])
• Standardized Root Mean Square Residual (SRMR): .035 to .044

Measurement invariance analyses across gender groups (boys vs. girls) establish full metric and scalar invariance, confirming that differences in observed mean scores reflect genuine differences in normative violent attitudes rather than gender-biased measurement artifacts.

10. Instrument / Measurement Tool

The operational specifications of the instrument are summarized below:

• Tool Name: Attitude Toward Violence scale (ATVS) [Middle School Adaptation]
• Instrument Type: Self-administered paper-and-pencil or computerized self-report questionnaire
• Target Population: Adolescents and youth, primarily middle school students in grades 6 through 8 (approximate ages 11–14); adaptable for early high school grades 9–10.
• Item Count: 6 items
• Response Format: Standard 5-point Likert scale:
  • 1 = Strongly disagree
  • 2 = Disagree
  • 3 = Neither
  • 4 = Agree
  • 5 = Strongly agree
• Administration Time: Approximately 3 to 5 minutes
• Scoring and Directionality:
  • Reverse Scored Items: Items 2 and 5 must be reverse scored prior to summing ($1
    ightarrow 5$, $2
    ightarrow 4$, $3
    ightarrow 3$, $4
    ightarrow 2$, $5
    ightarrow 1$).
  • Direct Scored Items: Items 1, 3, 4, and 6 are scored directly ($1 = 1$ to $5 = 5$).
  • Total Score Range: Minimum score is 6; maximum score is 30.
  • Interpretation: Higher cumulative scores reflect a stronger positive attitude toward violent conflict resolution strategies, heightened sensitivity to peer reputation costs of nonviolence, and a diminished appraisal/use of nonviolent dispute alternatives.

11. Permissions & Fee and Test Year

The original adaptations and additions of this 6-item scale were finalized in 1995 by Dr. Kris Bosworth and Dr. Dorothy L. Espelage at Indiana University. The measure was subsequently curated in 2005 by the Centers for Disease Control and Prevention (CDC) in the public-domain reference volume Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools.

• Cost / Fee: Free (no fee noted). The tool resides in the public domain as an educational and public health evaluation resource.
• Permissions: Researchers, clinical psychologists, school counselors, and program evaluators may freely administer, adapt, and score the instrument for non-commercial research, academic inquiry, and educational evaluation without written copyright clearance, provided appropriate academic attribution is accorded to the primary developers (Bosworth & Espelage, 1995; Dahlberg et al., 2005).

12. References

The following references provide the empirical, theoretical, and psychometric foundations for the scale:

• Bandura, A. (1973). Aggression: A social learning analysis. Prentice-Hall.
• Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall.
• Bosworth, K., & Espelage, D. L. (1995). Teen conflict survey. Center for Adolescent Studies, Indiana University, Bloomington, IN.
• Bosworth, K., Espelage, D. L., & Simon, T. R. (1999). Factors associated with bullying behavior in middle school students. The Journal of Early Adolescence, 19(3), 341–362. https://doi.org/10.1177/0272431699019003003
• 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://files.eric.ed.gov/fulltext/ED486261.pdf
• Espelage, D. L., Bosworth, K., & Simon, T. R. (2000). Examining the social context of bullying behaviors in early adolescence. Journal of Counseling & Development, 78(3), 326–333. https://doi.org/10.1002/j.1556-6676.2000.tb01914.x
• Huesmann, L. R., & Guerra, N. G. (1997). Children’s normative beliefs about aggression and aggressive behavior. Journal of Personality and Social Psychology, 72(2), 408–419. https://doi.org/10.1037/0022-3514.72.2.408
• Slaby, R. G., & Guerra, N. G. (1988). Cognitive mediators of aggression in adolescent offenders: 1. Assessment. Developmental Psychology, 24(4), 580–588. https://doi.org/10.1037/0012-1649.24.4.580

13. Items of the Scale (Questionnaire)