Norms for Aggression and Alternatives

1. Abstract

The Norms for Aggression and Alternatives (NAA) is a comprehensive psychometric instrument engineered to evaluate middle school adolescents’ social normative perceptions and personal moral evaluative standards concerning interpersonal aggression and constructive nonviolent alternatives. Adapted and operationalized within large-scale prevention trials—most notably the Multisite Violence Prevention Project (MVPP) coordinated by the Centers for Disease Control and Prevention (CDC)—the assessment builds upon Jay Jackson’s Return Potential Model of social norm measurement. Comprising 36 self-report items tailored to youth in grades 6 through 8, the NAA measures behavioral situations across physical, verbal, and passive-avoidant conflict domains. The instrument employs a parallel structural architecture that bisects behavioral events across two independent normative referent dimensions (Individual Norms versus Perceived School Norms) and two functional behavioral typologies (Aggression versus Alternatives to Aggression). This orthogonal design generates four core psychometric subscales: Individual Norms for Aggression (10 items), Individual Norms for Alternatives to Aggression (8 items), School Norms for Aggression (10 items), and School Norms for Alternatives to Aggression (8 items). Respondents evaluate how they personally would feel, or how their school peers would feel, regarding specific conflict behaviors utilizing a 3-point anchored response continuum: Not like it (1 point), Not care (2 points), and Like it (3 points). Subscale scores represent mean approval ratings ranging from 1.00 to 3.00, where higher scores reflect greater personal or perceived collective approval of the target actions. Extensively evaluated across diverse socioeconomic and multiethnic cohorts of young adolescents, the NAA exhibits robust psychometric integrity, demonstrating high internal consistency reliability coefficients (Cronbach’s alpha values regularly spanning 0.77 to 0.88 across subscales), robust confirmatory factor structures establishing cross-informant metric invariance, and well-documented convergent and predictive validity with externalizing problem behavior, peer sociometric status, and school climate indices. The scale remains a seminal tool for developmental psychologists, school mental health specialists, and evaluation researchers designing targeted youth violence intervention frameworks.

2. Keywords

Norms for Aggression and Alternatives, social norms, peer normative beliefs, adolescent aggression, school violence prevention, conflict resolution, Return Potential Model, middle school behavioral assessment, social-cognitive evaluation, antisocial behavior.

3. Authors

The operationalization and psychometric establishment of the Norms for Aggression and Alternatives instrument emerged from the cooperative collaborative framework of the Multisite Violence Prevention Project (MVPP) in conjunction with the Centers for Disease Control and Prevention (CDC), National Center for Injury Prevention and Control, Division of Violence Prevention, Atlanta, Georgia. Foundational theoretical and structural modeling originates from the sociological and social-psychological formulations of Jay Jackson (1966) at the University of Oregon, alongside cross-cultural adaptations of group norm structures advanced by Kunihiro Sasaki (1979). Major research consortium investigator teams facilitating the scale’s empirical validation include interdisciplinary research groups at Duke University, the University of Georgia, the University of Illinois at Chicago, and Virginia Commonwealth University.

4. Purpose

The primary purpose of the Norms for Aggression and Alternatives (NAA) is to quantify the cognitive, prescriptive, and evaluative architectures that govern youth interpersonal conflict. During early adolescence (typically spanning grades 6 through 8, ages 11 to 14), youth experience rapid neurobiological, social, and emotional transitions. These transitions are marked by a heightened sensitivity to peer status, group conformity, and evolving personal codes of social conduct. Within school environments, aggression is frequently governed not merely by individual impulse control deficits, but by entrenched social norms that tacitly validate retaliation, verbal escalation, or dominance postures while disparaging or devaluing constructive dispute resolution. The NAA was designed to resolve a long-standing gap in behavioral measurement: the precise disentanglement between what an adolescent personally approves of (individual prescriptive norms) versus what they believe their broader school peer collective endorses or condones (descriptive and perceived collective school norms).

From an applied research and clinical prevention standpoint, measuring these twin normative planes is essential for multiple reasons. First, clinical psychology and behavioral epidemiology have revealed the pervasive phenomenon of pluralistic ignorance within student populations. Students frequently overestimate peer approval of delinquent, violent, and retaliatory conduct (such as physical retaliation after a slight or physical strike) while simultaneously underestimating their peers’ true private support for prosocial de-escalation strategies (such as seeking adult help or actively listening to an opposing perspective). By systematically measuring both individual standards and perceived school norms in parallel, the NAA enables school psychologists, behavioral interventionists, and prevention researchers to detect gaps between personal values and collective perceptions. This diagnostic capability is critical for deploying targeted “social norms marketing” campaigns, restorative justice programs, and cognitive-behavioral curricula.

Second, the scale provides programmatic utility as an outcome evaluation instrument. Comprehensive school violence prevention initiatives—such as the MVPP’s universal and targeted intervention packages (e.g., Responding in Peaceful and Positive Ways, or RIPP)—require sensitive, reliable metrics capable of capturing subtle shifts in social-cognitive processing before overt behavioral repertoires shift. Reductions in weapon-carrying, physical altercations, and disciplinary suspensions are distal outcomes; proximal outcomes consist of cognitive restructuring, specifically the delegitimization of aggressive retaliation and the elevation of perceived normative acceptability regarding peaceful nonviolent alternatives. The NAA delivers standardized, granular measurement across specific instigation contexts (e.g., unprovoked hostility, provoked physical attack, rumors, and accidental collisions), thereby providing evaluators with fine-grained diagnostic insight into which behavioral domains are resistant or responsive to intervention.

5. Psychological Construct

The psychological construct assessed by the NAA is grounded in social-cognitive theories of moral evaluation, behavioral intention, and normative influence. Broadly defined, a norm constitutes a shared expectation or evaluative standard regarding how individuals ought or ought not to behave in specified social situations. The NAA specifically isolates the construct of normative evaluation across a 2 × 2 matrix crossing the referent source (Individual vs. School) with the behavioral typology (Aggression vs. Alternatives to Aggression). This structural matrix yields four distinct yet interrelated psychometric dimensions:

1. Individual Norms for Aggression

This subscale comprises 10 items assessing an adolescent’s personal moral endorsement, approval, or affective sanctioning of aggressive behavioral responses. The measured behaviors span physical aggression (e.g., hitting someone who said something mean, hitting someone who hit first, or hitting someone for no reason), verbal aggression (e.g., yelling at someone who yelled first or yelling unprovoked), and relational/threatening behaviors (e.g., threatening someone who said something mean or threatened unprovoked). Rather than measuring behavioral frequency (how often the youth actually fights), this dimension measures the cognitive legitimacy and personal justification assigned to violence. For instance, endorsing hitting someone in retaliation reflects a high internal normative valuation of retributive justice.

2. Individual Norms for Alternatives to Aggression

Consisting of 8 items, this subscale quantifies an individual’s personal endorsement of constructive, prosocial, and self-regulatory conflict resolution mechanisms. Items evaluate actions such as ignoring a rumor, actively asserting to a peer that alternative conflict resolution options exist, enlisting the support of a teacher or school adult when challenged to fight, offering an apology following an accidental hallway bump, verbally intervening with a peer to “stop and calm down,” taking physical detours to actively avoid an impending altercation, listening to an opponent’s perspective, and engaging physiological self-regulation (taking deep breaths). High scores on this subscale reflect high personal valuing of peaceful de-escalation, emotional self-control, and prosocial problem-solving.

3. School Norms for Aggression

Mirroring the 10 aggressive situations presented in the individual subscale, this 10-item subscale captures the student’s perception of the collective peer climate within their school building. Items require the respondent to assess: “How would the kids in your school feel if a kid…” engaged in unprovoked, retaliatory, physical, or verbal hostility. This dimension assesses the perceived descriptive and injunctive climate of the peer ecology. It measures whether violence is perceived to confer peer status, social dominance, and communal admiration, or whether it results in communal opprobrium and social marginalization within the broader school culture.

4. School Norms for Alternatives to Aggression

Mirroring the 8 prosocial scenarios, this 8-item subscale measures the adolescent’s subjective assessment of the school peer ecology’s respect for peaceful resolution. In middle school settings, nonviolent alternatives (such as reporting threats to teachers or walking away from a physical dare) can be interpreted as signs of maturity and moral integrity, or stigmatized as cowardice, weakness, or violating peer solidarity codes (“snitching”). This subscale measures whether youth believe their school peer group validates, respects, and rewards peaceful de-escalation, or treats nonviolence with social disdain.

6. Theoretical Framework

The conceptual foundation of the NAA integrates three major paradigms in sociology, social psychology, and developmental behavioral science: Jay Jackson’s Return Potential Model (RPM), Social Cognitive Theory as formulated by Albert Bandura, and Social Information Processing (SIP) theory formulated by Kenneth Dodge and Nicki Crick.

Jay Jackson’s Return Potential Model (RPM)

Historically formulated by Jay Jackson in 1966, the Return Potential Model provides the direct structural blueprint for the NAA. Jackson posited that a social norm cannot be understood merely as a single modal behavior; instead, it is a two-dimensional curve plotting a continuum of behavior (horizontal axis) against an evaluative dimension of approval versus disapproval (the “return potential” vertical axis). The height and shape of this return curve quantify the degree of group consensus, the intensity of normative feeling, and the latitude of acceptable behavior. In 1979, Kunihiro Sasaki expanded the RPM to evaluate group peer dynamics and structural norms within educational settings. The NAA operationalizes this framework by presenting standardized points along a behavioral continuum—ranging from high physical violence to proactive verbal mediation—and requiring respondents to evaluate each behavioral act along an approval scale (disapproval/”not like it”, indifference/”not care”, and approval/”like it”).

Bandura’s Social Cognitive Theory & Moral Disengagement

Albert Bandura’s Social Cognitive Theory underscores that human action is motivated and regulated by internal standards and self-evaluative consequences. When an individual adopts moral standards that sanction aggression as legitimate, violence becomes a socially and cognitively permissible avenue for securing self-worth, social dominance, and status. Furthermore, Bandura’s construct of moral disengagement demonstrates how cognitive mechanisms—such as advantageous comparison, blaming the victim, and minimizing the consequences of harm—allow aggressive actions to occur without self-censure. In the NAA, items isolating reactive hostility (e.g., hitting someone who hit first or who said something mean) capture the adolescent’s threshold for moral disengagement under perceived provocation.

Dodge and Crick’s Social Information Processing (SIP) Model

The reformulated Social Information Processing model (Crick & Dodge, 1994) conceptualizes adolescent social interaction through a sequential six-step cognitive model: (1) encoding of cues, (2) interpretation of cues, (3) clarification of goals, (4) response access or construction, (5) response decision, and (6) behavioral enactment. Step 5—the response decision phase—is where normative beliefs exert direct regulatory influence. When selecting among potential behavioral scripts generated in Step 4 (such as retaliating vs. walking away), adolescents evaluate each response according to two criteria: outcome expectations (what will happen if I act this way?) and response evaluation / moral acceptability (is this behavior right, and will my peers approve?). The NAA evaluates this fifth step by quantifying whether nonviolent alternatives or violent responses are deemed cognitively and socially normative.

7. Validity

The psychometric validity of the NAA has been established through extensive empirical trials, most prominently within the multisite cohort studies undertaken by the Multisite Violence Prevention Project (2004, 2008), the CDC Division of Violence Prevention compendia, and subsequent peer-reviewed investigations into school violence etiology.

Construct and Structural Validity

Construct validity was initially established through structural equation modeling (SEM) and hierarchical confirmatory factor analysis (CFA) across diverse middle school student cohorts. Multi-group confirmatory analyses supported a distinct four-factor oblique model over alternative unidimensional or two-factor models (e.g., merging individual standards with school norms), demonstrating that adolescents maintain psychologically differentiated representations of their own moral standards versus the perceived ethos of their peer collective. Structural invariance tests confirmed that the four-factor construct demonstrates strong measurement invariance across biological sex (boys vs. girls) and grade levels (6th through 8th grades), indicating that the scale functions equivalently across adolescent developmental bands.

Convergent and Discriminant Validity

The convergent validity of the NAA is demonstrated through its statistical correlations with theoretical benchmarks. Individual Norms for Aggression correlates positively and significantly with self-reported physical fighting, disciplinary infraction records, hostile attributional bias, and externalizing behavioral checklists (e.g., Youth Self-Report externalizing scales), with observed correlation coefficients typically falling between $r = 0.35$ and $r = 0.52$ ($p < .001$). Conversely, Individual Norms for Alternatives to Aggression correlates positively with prosocial competence scales, empathy assessments, emotional self-regulation indices, and peer nominations for cooperative behavior ($r = 0.30$ to $0.45$). Discriminant validity is supported by low correlations with unrelated constructs, such as academic intelligence scores and standardized test performance, confirming that the subscales do not merely reflect generalized cognitive ability or test-taking competence.

Predictive and Longitudinal Validity

Longitudinal structural modeling conducted over multi-year evaluation trials demonstrates that NAA baseline scores predict subsequent behavioral trajectories. Elevated baseline endorsements on the Individual Norms for Aggression and School Norms for Aggression subscales predict increases in adolescent self-reported violent behaviors and official school suspension rates 12 to 24 months later, controlling for baseline delinquent involvement. Furthermore, students reporting high personal disapproval of aggression who simultaneously perceived high School Norms for Aggression displayed elevated internalizing distress symptoms (anxiety, somatic complaints, and feelings of school alienation), confirming the pathogenic effect of experiencing moral dissonance within a perceived aggressive peer environment.

8. Reliability

The NAA has demonstrated consistent reliability coefficients across varied adolescent samples, high-risk community cohorts, and multi-school evaluation cohorts.

Internal Consistency Reliability

Extensive psychometric investigations conducted by the Multisite Violence Prevention Project documented high Cronbach’s alpha ($lpha$) coefficients across all four subscales. In representative normative middle school samples (grades 6–8, $N > 4,000$):

• School Norms for Aggression (10 items): Cronbach’s $lpha$ ranges from 0.82 to 0.88, indicating high internal coherence among perceptions of collective peer hostility.
• School Norms for Alternatives to Aggression (8 items): Cronbach’s $lpha$ ranges from 0.77 to 0.83.
• Individual Norms for Aggression (10 items): Cronbach’s $lpha$ consistently demonstrates strong reliability, spanning 0.80 to 0.86 across study cohorts.
• Individual Norms for Alternatives to Aggression (8 items): Cronbach’s $lpha$ ranges from 0.76 to 0.82.

McDonald’s omega ($\omega$) coefficients, reported in contemporary re-analyses, mirror these values, ranging from $0.78$ to $0.87$, confirming that internal consistency is not an artifact of tau-equivalence violations.

Temporal Stability (Test-Retest Reliability)

Test-retest stability was evaluated across short-term non-intervention control cohorts over intervals spanning 4 to 8 weeks. Intra-class correlation coefficients (ICC) and Pearson product-moment correlations showed substantial temporal stability, ranging from $r_{tt} = 0.68$ to $r_{tt} = 0.77$ for the individual norm subscales, and $r_{tt} = 0.61$ to $r_{tt} = 0.72$ for the perceived school norm subscales. The slightly lower stability for perceived school norms reflects the expected developmental sensitivity of peer climate perceptions to ongoing contextual dynamics within middle school environments.

9. Factor Analysis

The latent structural architecture of the NAA has been evaluated using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

Initial principal axis factoring with promax (oblique) rotation was conducted during scale construction to verify the conceptual orthogonality of individual versus school referents and aggressive versus prosocial behaviors. Analysis of eigenvalues (> 1.00 criteria) and scree plot examination identified four clear latent factors accounting for more than 54% of the cumulative total variance. Factor loadings demonstrated clean simple structure:

• All 10 items assessing personal evaluation of aggression loaded onto the Individual Norms for Aggression factor (loadings ranging from 0.48 to 0.74) without cross-loading on prosocial dimensions.
• All 10 items assessing school peer evaluations of aggression loaded onto the School Norms for Aggression factor (loadings ranging from 0.52 to 0.76).
• The 8 individual alternative items and 8 school alternative items cleanly formed their respective factors, with standardized factor loadings spanning 0.44 to 0.71.

Confirmatory Factor Analysis (CFA) and Model Fit

Subsequent structural modeling in large validation samples cross-validated this 4-factor configuration against alternative structural models. Competing models included: (a) a single general “normative acceptance” factor, (b) a 2-factor model separating only Aggression from Alternatives regardless of referent, and (c) a 2-factor model separating Individual from School referents regardless of behavioral valence.

The hypothesized 4-factor model demonstrated superior fit across standard structural equation modeling indices:

• Comparative Fit Index (CFI): 0.93 to 0.96 (exceeding the standard 0.90 cut-off for acceptable fit).
• Tucker-Lewis Index (TLI): 0.92 to 0.95.
• Root Mean Square Error of Approximation (RMSEA): 0.038 to 0.048 (90% Confidence Interval: [0.034, 0.051]), well within the ≤ 0.05 threshold indicating close fit.
• Standardized Root Mean Square Residual (SRMR): 0.041.

Correlations among the latent factors revealed that Individual Norms for Aggression and School Norms for Aggression were moderately positively intercorrelated ($r pprox 0.42$), while Individual Norms for Aggression and Individual Norms for Alternatives were inversely related ($r pprox -0.36$), confirming construct divergence alongside meaningful theoretical relatedness.

10. Instrument / Measurement Tool

• Instrument Name: Norms for Aggression and Alternatives (NAA)
• Original Authors / Developers: Multisite Violence Prevention Project (CDC); based conceptually on models by Jay Jackson (1966) and Kunihiro Sasaki (1979)
• Target Population / Age Range: Middle school students (typically grades 6 through 8; ages 11–14 years)
• Administration Format: Self-administered paper-and-pencil or computerized questionnaire; administered individually or in group classroom settings
• Administration Time: Approximately 15 to 20 minutes
• Total Number of Items: 36 items
• Subscale Breakdown:
  • Individual Norms for Aggression (10 items): Items 3, 7, 13, 17, 19, 21, 27, 29, 31, 33
  • School Norms for Aggression (10 items): Items 4, 8, 14, 18, 20, 22, 28, 30, 32, 34
  • Individual Norms for Alternatives to Aggression (8 items): Items 1, 5, 9, 11, 15, 23, 25, 35
  • School Norms for Alternatives to Aggression (8 items): Items 2, 6, 10, 12, 16, 24, 26, 36
• Response Format and Numerical Scoring: Items utilize a categorical 3-point evaluative scale scored as:
  • Not like it = 1
  • Not care = 2
  • Like it = 3
• Scoring and Computational Procedures:
  • For each of the four subscales, sum the point values of the answered items.
  • Divide the subscale sum by the total number of items in that subscale (10 for the aggression subscales; 8 for the alternatives subscales).
  • The resulting score is a mean composite spanning from 1.00 to 3.00 for each dimension.
  • Higher scores on the Aggression subscales reflect greater approval/legitimization of violence. Higher scores on the Alternatives subscales reflect greater approval/endorsement of prosocial conflict resolution.

11. Permissions & Fee and Test Year

The Norms for Aggression and Alternatives (NAA) was developed under federal grant support coordinated by the Centers for Disease Control and Prevention (CDC), National Center for Injury Prevention and Control, and documented in its cohort-wide survey protocols in 2004. As an instrument compiled and distributed within federal violence prevention compendia, the NAA resides in the public domain for research, academic, educational, and clinical evaluation purposes. No royalty fees, purchase licensing agreements, or commercial software activations are required to administer, score, or adapt the assessment. Researchers and clinicians are expected to maintain instrument integrity and properly cite the original source compendium and the Multisite Violence Prevention Project foundational publications in all research output.

12. References

Bandura, A. (1991). Social cognitive theory of moral thought and action. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development (Vol. 1, pp. 45–103). Lawrence Erlbaum Associates.

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

Guerra, N. G., & Slaby, R. G. (1990). Evaluative factors in social problem solving by aggressive boys. Developmental Psychology, 26(2), 269–277. https://doi.org/10.1037/0012-1649.26.2.269

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

Jackson, J. (1966). A conceptual and measurement model for norms and roles. Pacific Sociological Review, 9(1), 35–47. https://doi.org/10.2307/1388303

Multisite Violence Prevention Project. (2004). Description of measures: Cohort-wide student survey. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control.

Multisite Violence Prevention Project. (2008). The Multisite Violence Prevention Project: Impact of a universal school-based violence prevention program on social-cognitive outcomes in eighth grade. Prevention Science, 9(4), 231–244. https://doi.org/10.1007/s11121-008-0104-5

Sasaki, K. (1979). Present status of research on group norms in Japan. American Journal of Community Psychology, 7(2), 147–158. https://doi.org/10.1007/BF00894381

13. Items of the Scale