Normative Beliefs about Aggression

1. Abstract

The Normative Beliefs About Aggression Scale (NOBAGS) is an influential self-report psychometric instrument designed to evaluate an individual’s internal cognitive standards regarding the moral acceptability and social appropriateness of aggressive behavior. Developed by L. Rowell Huesmann, Nancy G. Guerra, and colleagues, the scale operationalizes the cognitive constructs embedded within social-cognitive models of aggressive conduct, specifically the Social Information Processing (SIP) model and cognitive script theory. The 20-item version of the NOBAGS assesses normative cognitive approval across two overarching domains: Retaliation Beliefs (12 items evaluating physical and verbal retaliation under systematic variations of verbal or physical provocation across peer gender dyads) and General Beliefs (8 items assessing context-free approval of verbal and physical aggression).

Each item utilizes a 4-point Likert response format measuring degree of endorsement or condemnation (ranging from It’s really wrong to It’s perfectly OK). Psychometrically, the instrument demonstrates robust internal consistency across diverse developmental cohorts ranging from early childhood and elementary school through adolescence, emerging adulthood, and collegiate samples, with typical Cronbach’s alpha coefficients spanning α = .72 to .91 across subscales. Extensive structural equation modeling and confirmatory factor analyses support both correlated two-factor and hierarchical models, delineating general normative acceptance from context-specific retaliatory justifications. Longitudinal investigations have firmly established the scale’s construct, convergent, discriminant, and predictive validity, revealing that pro-aggressive normative beliefs significantly predict subsequent aggressive acts, mediate the longitudinal effects of media violence exposure and community violence exposure on antisocial conduct, and serve as primary mechanisms of change in school-based cognitive-behavioral prevention programs.

2. Keywords

Normative Beliefs About Aggression Scale, NOBAGS, L. Rowell Huesmann, Nancy G. Guerra, Social Information Processing, Aggression, Cognitive Scripts, Retaliatory Aggression, Moral Evaluation, Social-Cognitive Theory

3. Authors

The primary architects and psychometric developers of the Normative Beliefs About Aggression Scale are:

• L. Rowell Huesmann, Ph.D. — Amos N. Tversky Collegiate Professor Emeritus of Communication Studies and Psychology, and Research Professor Emeritus at the Institute for Social Research (ISR), Research Center for Group Dynamics, University of Michigan, Ann Arbor, Michigan, USA. Dr. Huesmann is an internationally distinguished scholar in the etiology of aggression, media violence, cognitive script acquisition, and social-cognitive development.
• Nancy G. Guerra, Ed.D. — Professor of Psychological and Brain Sciences, University of Delaware, Newark, Delaware, USA. Dr. Guerra is a prominent clinical and developmental psychologist specializing in child development, youth violence prevention, social-emotional learning, and international child rights.
• Collaborating Co-Authors: Laurie Miller and Arnaldo Zelli, who collaborated with Huesmann and Guerra on foundational investigations examining peer social norms, cognitive mediators, and the developmental epidemiology of childhood aggression.

Primary Administrative Contact:
L. Rowell Huesmann, Ph.D.
Research Center for Group Dynamics
Institute for Social Research, University of Michigan
426 Thompson Street, Ann Arbor, MI 48106-1248
Email: [email protected]

4. Purpose

The fundamental purpose of the Normative Beliefs About Aggression Scale is to quantify individual differences in normative cognitions regarding peer-directed aggression. Normative beliefs are defined as self-regulating cognitive standards, moral appraisals, or social expectations concerning what constitutes acceptable, tolerable, or impermissible behavior under specified situational conditions. Within developmental, educational, and clinical psychology, researchers and practitioners historically struggled to separate general behavioral tendencies or aggressive affect (e.g., hostility, irritability, impulsivity) from internal cognitive evaluations that license or inhibit aggressive behaviors. The NOBAGS was specifically formulated to isolate and measure this internal cognitive regulatory mechanism.

In research contexts, the scale addresses the underlying psychological architecture that explains why some individuals engage in recurrent interpersonal aggression while others exercise behavioral inhibition when subjected to interpersonal conflict, frustration, or peer provocation. By systematically varying situational parameters—such as the nature of provocation (verbal insult versus physical assault) and the gender configuration of the peer interaction (boy-to-boy, boy-to-girl, girl-to-girl, and girl-to-boy)—the scale illuminates the boundary conditions and cognitive heuristics that govern an individual’s behavioral threshold for violent conduct.

In applied and clinical settings, the NOBAGS fulfills crucial diagnostic, evaluative, and intervention-monitoring functions. Specifically, it is utilized to:

• Screen and Identify At-Risk Youths: The scale enables school psychologists, counselors, and forensic evaluators to identify children and adolescents who endorse high levels of pro-aggressive normative standards, placing them at heightened developmental risk for conduct disorder, juvenile delinquency, disciplinary exclusion, and peer rejection.
• Baseline Cognitive Assessment in Interventions: Clinicians implementing cognitive-behavioral therapy (CBT), social problem-solving curricula, or violence prevention programs (such as the Metropolitan Area Child Study or anger management regimens) use the instrument to establish pre-treatment cognitive baseline scores.
• Monitor Longitudinal Treatment Outcomes: Because normative beliefs act as cognitive mediators of behavioral change, tracking fluctuations in NOBAGS scores across treatment phases allows researchers and clinicians to assess whether therapeutic interventions successfully restructure maladaptive, retaliatory schemas prior to observable behavioral modifications.
• Examine Environmental and Media Effects: Epidemiologists and social psychologists implement the NOBAGS to delineate how chronic exposure to community violence, harsh parental disciplinary practices, or violent mass media systematically shifts an individual’s normative baseline toward moral desensitization and behavioral endorsement of interpersonal violence.

5. Psychological Construct

The psychological construct evaluated by the NOBAGS centers on normative beliefs, which operate as internal cognitive rules, schemas, and ethical parameters concerning the acceptability of behavioral acts. Normative beliefs are distinct from perceived social norms (what an individual believes peers or authority figures actually do or condone); rather, they represent an internalized moral and cognitive code regarding what is right, justified, or permissible for oneself and others to do under designated conditions. As conceptualized by Huesmann and Guerra (1997), these beliefs are organized along distinct cognitive dimensions:

Retaliation Normative Beliefs

The retaliatory dimension captures an individual’s cognitive endorsement of aggression as an acceptable or necessary response to perceived provocation. Rather than examining unprovoked, proactive antisocial behavior, retaliatory normative beliefs evaluate context-dependent, reactive scripts. In human social interaction, retaliatory justifications often represent the most pervasive and psychologically entrenched rationalizations for aggressive retaliation. The NOBAGS dissects this construct along two operational axes:

• Mode and Intensity of Provocation vs. Response: The instrument assesses responses to verbal provocation (e.g., someone saying something bad) compared to physical provocation (e.g., being hit). It further examines whether the respondent endorses verbal retaliation (screaming, yelling) or escalatory physical retaliation (hitting). Cognitively mature individuals typically recognize proportionality, rejecting physical retaliation in response to verbal insults; individuals with elevated retaliatory normative beliefs exhibit a lower cognitive threshold, approving severe physical violence even for minor verbal transgressions.
• Gender Dyad Configurations: Socialization exerts powerful effects on gendered behavioral scripts. The NOBAGS deliberately varies the gender of the provocateur and the gender of the respondent across four distinct relational dyads: male-on-male, female-on-female, male-on-female, and female-on-male. This multi-dyadic structure allows researchers to detect gender-typed scripts (e.g., chivalry norms inhibiting male-to-female aggression, or symmetric norms tolerating male-to-male peer fighting).

General Normative Beliefs

The general dimension evaluates an individual’s trans-situational, context-free normative evaluation of aggression. Unlike retaliatory items, general normative belief items do not specify external provocation, peer confrontation, or mitigating circumstances. Instead, they probe the baseline moral impermissibility or acceptability of verbal aggression (e.g., yelling, insulting, taking anger out on others verbally) and physical aggression (e.g., hitting, pushing, shoving, physical force). Individuals who endorse general normative beliefs regarding aggression exhibit profound deficits in moral disengagement and social-cognitive inhibition, viewing force and verbal hostility as legitimate instruments for everyday emotional expression and conflict resolution.

Collectively, these dimensions form a cognitive hierarchy. General beliefs represent global, abstract attitudes toward aggression, whereas retaliatory beliefs reflect situationally activated conditional rules (e.g., “If provoked physically by an adversary of the same gender, then striking back is morally obligatory”). Together, they dictate whether an aggressive behavioral script will be filtered out as unacceptable or approved for behavioral execution during emotionally charged interpersonal encounters.

6. Theoretical Framework

The theoretical architecture undergirding the Normative Beliefs About Aggression Scale is rooted primarily in Social Information Processing (SIP) Theory formulated by Kenneth A. Dodge and Nicki R. Crick, synthesized with the Cognitive Script Theory and cumulative developmental models articulated by L. Rowell Huesmann (1988, 1998), as well as Social Cognitive Theory developed by Albert Bandura.

Cognitive Script Theory and the Normative Filter

According to Huesmann’s cognitive script theory, social behavior is largely controlled by cognitive “scripts” stored in memory. Scripts are associative networks that define situational roles, behavioral expectations, and anticipated sequences of events. When an individual encounters an environmental trigger or interpersonal conflict, relevant behavioral scripts are retrieved from long-term memory. Once a script is accessed, however, it is not instantaneously enacted. Instead, it must pass through an internal cognitive evaluation process known as the normative filter.

The normative filter compares the retrieved behavioral script against the individual’s internalized normative beliefs. This evaluative phase incorporates several cognitive appraisals:

• Self-Congruence and Moral Evaluation: “Does this action match my internal sense of right and wrong? Will executing this script make me feel guilty or ashamed?”
• Expected Social Sanctions: “Will my peer group, family, or social environment approve or condemn this action?”
• Self-Efficacy for Script Execution: “Do I possess the capacity to execute this behavior, and is it justified under current circumstances?”

If the normative filter classifies the retrieved aggressive script as impermissible, the script is rejected, prompting the individual to search for alternative nonviolent problem-solving scripts. However, if an individual’s normative beliefs endorse aggression as justifiable (e.g., “Hitting someone who insulted you is the right thing to do”), the script passes the filter unchallenged and proceeds to behavioral execution. The NOBAGS directly measures the permeability of this normative filter.

Integration with Social Information Processing (SIP)

In Crick and Dodge’s (1994) reformulated SIP model, social cognitive processing proceeds through six sequential steps: (1) encoding of external and internal cues; (2) interpretation of cues (including hostile attribution biases); (3) clarification or selection of goals; (4) response access or construction; (5) response decision; and (6) behavioral enactment. Normative beliefs operate predominantly at Step 5: Response Decision.

At Step 5, individuals evaluate generated responses on dimensions of outcome expectations, self-efficacy, and moral appropriateness. Even when an individual erroneously perceives hostile intent from a peer (Step 2) and generates an aggressive retaliation script (Step 4), robust anti-aggressive normative beliefs can override the aggressive impulse during Step 5 evaluation, preventing violent retaliation. Conversely, individuals possessing lenient normative beliefs toward aggression will approve and execute the violent response. Longitudinal empirical research has demonstrated a bidirectional developmental relationship: aggressive behavior and normative beliefs reinforce one another over time in a transactional feedback loop, leading to deeply entrenched cognitive schemas.

7. Validity

The construct, convergent, discriminant, and predictive validity of the NOBAGS has been empirically documented across extensive cross-sectional and multi-wave longitudinal research programs spanning early childhood, adolescence, and emerging adulthood.

Construct and Factorial Validity

Construct validity was initially established by Huesmann, Guerra, and colleagues in large-scale developmental samples, most notably within the multi-site Metropolitan Area Child Study (MACS) involving thousands of urban elementary and middle school students. Confirmatory structural modeling confirmed that items operationalizing general aggression and retaliatory aggression load significantly on distinct, yet interrelated latent dimensions. The constructs demonstrated cross-sample invariance across racial, ethnic, and socioeconomic strata, confirming that the scale measures equivalent psychological constructs across diverse demographics.

Predictive and Longitudinal Validity

The predictive utility of the NOBAGS represents its most notable psychometric strength. In a landmark longitudinal study, Huesmann and Guerra (1997) assessed elementary school children across multiple years, demonstrating that normative beliefs assessed in early grades significantly predicted subsequent peer-nominated and teacher-rated physical aggression in later grades, even after controlling for baseline levels of aggressive behavior. Importantly, this predictive relationship exhibited a distinct developmental trajectory: while aggressive behavior in first grade predicted subsequent changes in normative beliefs, by fourth grade, normative beliefs crystallized and began significantly driving future aggressive behavior, illustrating the developmental consolidation of the normative filter.

Convergent Validity

Convergent validity is verified by robust, statistically significant correlations between NOBAGS scores and conceptually related psychological constructs:

• Hostile Attributional Bias: Significant positive correlations (ranging from r = .28 to .46) are documented between pro-aggressive normative beliefs and the tendency to attribute hostile intent to ambiguous peer actions.
• Moral Disengagement: NOBAGS scores correlate positively (r = .40 to .58) with Bandura’s Mechanisms of Moral Disengagement Scale, particularly subscales assessing moral justification, euphemistic labeling, and attribution of blame.
• Exposure to Violence: Scores correlate moderately to strongly (r = .30 to .52) with measures of chronic community violence exposure, domestic violence victimization, and exposure to violent video games and media.

Discriminant Validity

Discriminant validity has been demonstrated by showing that NOBAGS scores do not merely reflect general cognitive deficits, low intelligence, or global negative affectivity. Correlations between the NOBAGS and standardized IQ measures or academic achievement tests are negligible or weakly negative (typically r = −.08 to −.14). Furthermore, the scale demonstrates clear differentiation from internalizing symptomatology, such as generalized anxiety, somatic complaints, and depressive symptoms, confirming that the instrument measures cognitive standards specific to aggression rather than undifferentiated psychological distress.

8. Reliability

The Normative Beliefs About Aggression Scale displays high internal consistency and temporal stability across developmental age groups and linguistic adaptations.

Internal Consistency

In original validation cohorts comprising elementary school students (Grades 1 through 6) and high school adolescents, Huesmann and Guerra (1997) reported robust internal consistency coefficients:

• Total Scale (20 items): Cronbach’s alpha values typically range between α = .85 and .92 across multi-ethnic community and clinical cohorts.
• Retaliation Subscale (12 items): Cronbach’s alpha coefficients consistently fall between α = .82 and .90.
• General Beliefs Subscale (8 items): Cronbach’s alpha values range from α = .73 to .86.

Subscale reliability tends to increase with participant age, reflecting the developmental stabilization of moral-cognitive schemas. For example, while alpha coefficients for the General Beliefs subscale hover around .72 to .76 among 1st and 2nd graders, they routinely exceed .84 among adolescents and collegiate respondents.

Test-Retest Reliability and Temporal Stability

Test-retest reliability assessments conducted over short intervals (2 to 4 weeks) demonstrate high stability, with Pearson correlation coefficients spanning r = .78 to .86. Over longer longitudinal spans (e.g., 1-year to 2-year testing intervals), stability coefficients remain moderate to strong (ranging from r = .45 to .62), reflecting both the trait-like durability of crystallized normative schemas in youth and their sensitivity to developmental maturation, environmental shifts, and targeted school-based preventive interventions.

9. Factor Analysis

The latent factorial architecture of the NOBAGS has been comprehensively examined using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across diverse international populations.

Factor Structures Identified

Psychometric evaluations generally support two alternative structural models depending on empirical applications:

• Correlated Two-Factor Model: This widely adopted model specifies two distinct latent constructs: Retaliatory Aggression Beliefs (12 items) and General Aggression Beliefs (8 items). The correlation between these two latent factors typically ranges between r = .55 and .70, demonstrating that while individuals who condone unprovoked aggression almost universally endorse retaliatory violence, many individuals who strictly condemn unprovoked aggression nonetheless accept retaliatory striking as morally defensible.
• Hierarchical / Higher-Order Factor Model: In this configuration, a higher-order latent construct (Total Normative Approval of Aggression) accounts for the common variance shared by lower-order first-order factors: Retaliation Beliefs and General Beliefs. Secondary sub-factors can further partition Retaliation into Verbal Provocation/Response versus Physical Provocation/Response, alongside peer gender combinations.

Model Fit and Item Factor Loadings

In confirmatory factor analytic studies, the correlated two-factor model consistently yields excellent fit indices when evaluated against standard psychometric criteria:

• Goodness-of-Fit Statistics: Typical CFA fit metrics include Comparative Fit Index (CFI) ≥ .94 to .97, Tucker-Lewis Index (TLI) ≥ .93 to .96, Root Mean Square Error of Approximation (RMSEA) ≤ .042 to .058 (with 90% confidence intervals between .035 and .065), and Standardized Root Mean Square Residual (SRMR) ≤ .048.
• Factor Loadings: Standardized factor loadings across items are uniformly strong. For the 12 Retaliation items, standardized factor loadings generally range from λ = .54 to .82 on their respective factor. For the 8 General Belief items, standardized loadings span λ = .50 to .79. All loadings are statistically significant at p < .001.
• Measurement Invariance: Multi-group CFA investigations have demonstrated full metric and scalar measurement invariance across biological sex (boys vs. girls) and longitudinal time points, confirming that differences in observed scores reflect true developmental and group differences rather than structural measurement artifacts.

10. Instrument / Measurement Tool

• Name: Normative Beliefs About Aggression Scale (NOBAGS)
• Alternative Titles: Normative Beliefs About Aggression Questionnaire, Huesmann & Guerra Normative Beliefs Scale
• Construct Assessed: Individual cognitive standards regarding the moral acceptability and social permissibility of interpersonal physical and verbal aggression
• Target Population / Age Range: Children (elementary school, Grades 1–6), adolescents (middle and high school), emerging adults, and college students. Suitable for diverse community, educational, and clinical samples.
• Administration Format: Paper-and-pencil self-report or computer-administered digital assessment. Can be administered individually or in group classroom settings. For younger children (Grades 1–3), items are read aloud by a trained examiner.
• Administration Time: Approximately 10 to 15 minutes
• Total Item Count: 20 items
• Subscales:
  • Retaliation Beliefs Subscale: Items 1 through 12 (provocation scenarios varying by verbal vs. physical actions and peer gender combinations)
  • General Beliefs Subscale: Items 13 through 20 (context-free statements regarding physical and verbal aggression)
• Response Scale: 4-point Likert-type response format consisting of:
  • It’s really wrong
  • It’s sort of wrong
  • It’s sort of OK
  • It’s perfectly OK
• Scoring and Coding Rules:
  • Items are scored on a numerical metric ranging from 1 to 4, where higher scores consistently reflect greater acceptance/endorsement of aggression.
  • Directly Scored Items (Pro-aggression framing): Items where an affirmative (“OK”) answer condones aggression are scored as: It’s really wrong = 1, It’s sort of wrong = 2, It’s sort of OK = 3, It’s perfectly OK = 4. This applies to items 1, 2, 5, 6, 10, 12, 14, 15, 16, and 20.
  • Reverse-Scored Items (Anti-aggression framing): Items phrased in terms of whether an act is “wrong” require reverse coding so that moral condemnation yields a lower score and tolerance yields a higher score: It’s really wrong = 1, It’s sort of wrong = 2, It’s sort of OK = 3, It’s perfectly OK = 4 when evaluating the “OK” spectrum, or reversed accordingly when keyed to the “wrong” pole (i.e., endorsement that an action is “really wrong” = 1; “sort of wrong” = 2; “sort of OK” = 3; “perfectly OK” = 4). Specifically, items 3, 4, 7, 8, 9, 11, 13, 17, 18, and 19 are keyed to ensure that choosing “It’s really wrong” indicates low normative acceptance of aggression (= 1) and “It’s perfectly OK” indicates high normative acceptance (= 4).
  • Scale Indices: Composite scores can be computed as a mean score (ranging from 1.00 to 4.00) or total sum score for the entire scale (20 to 80), as well as separate mean or sum scores for the Retaliation subscale (12 to 48) and the General Beliefs subscale (8 to 32).

11. Permissions & Fee and Test Year

• Year of Initial Publication: 1992 (initial developmental book chapter publication by Huesmann, Guerra, Miller, & Zelli); comprehensively standardized and psychometrically validated in 1997 (Huesmann & Guerra).
• Copyright Status: The scale is copyrighted by L. Rowell Huesmann and Nancy G. Guerra.
• Fees and Licensing: The instrument is non-commercial and available free of charge for academic, research, and non-profit educational purposes. Researchers and clinical practitioners may access and administer the scale within academic investigations without royalty fees.
• Permission Protocol: Users are requested to contact the primary author for formal permission and to register their academic usage:
  L. Rowell Huesmann, Ph.D.
  Research Center for Group Dynamics, Institute for Social Research
  University of Michigan, 426 Thompson Street, Ann Arbor, MI 48106-1248
  Telephone: (734) 764-8385 | Fax: (734) 763-1202
  Email: [email protected]
• Public Domain & Compendium Availability: The full instrument, administration procedures, and psychometric documentation are also indexed in the public domain via the Centers for Disease Control and Prevention (CDC) Compendium of Assessment Tools: Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths (Dahlberg et al., 2005).

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., Nucci, L., & Huesmann, L. R. (1994). Moral cognition and aggressive children: Cognition, morality, and violence. In B. P. Sternberg (Ed.), Moral development and character education (pp. 111–133). Duke University Press.

Huesmann, L. R. (1988). An information processing model for the development of aggression. Aggressive Behavior, 14(1), 13–24. https://doi.org/10.1016/B978-012278805-5/50005-5

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

Huesmann, L. R., Guerra, N. G., Miller, L., & Zelli, A. (1992). The role of social norms in the development of aggression. In H. Zumkley & A. Fraczek (Eds.), Socialization and aggression (pp. 139–152). Springer-Verlag. https://doi.org/10.1007/978-3-642-77112-5_9

Metropolitan Area Child Study Research Group. (2002). A cognitive-ecological approach to preventing aggression in urban settings: Initial outcomes for high-risk children. Journal of Consulting and Clinical Psychology, 70(1), 179–194. https://doi.org/10.1037/0022-006X.70.1.179

Zelli, A., Dodge, K. A., Lochman, J. E., & Laird, R. D. (1999). The distinction between beliefs supporting aggression and biased processing of social cues. Journal of Personality and Social Psychology, 77(1), 150–166. https://doi.org/10.1037/0022-3514.77.1.150

13. Items of the Scale (Questionnaire)