Participant Role Questionnaire

Abstract

The Participant Role Questionnaire (PRQ) is a seminal psychometric assessment instrument designed to capture the multifaceted, group-based dynamics of school-age bullying. Developed originally by Christina Salmivalli and colleagues (1996) and subsequently adapted into streamlined and self-report variants (Salmivalli & Voeten, 2004), the PRQ shifted the psychological paradigm of peer harassment from a purely dyadic framework (perpetrator vs. victim) to a comprehensive systemic group-process model. The standard peer-assessment version evaluates five distinct social participant roles across 15 behavioral items: the Bully (the initiator and leader of harassment), the Assistant (who directly aids the primary perpetrator), the Reinforcer (who provides social reward through laughter or incitement), the Defender (who provides emotional support or active intervention on behalf of the victim), and the Outsider (who disengages, remains passive, or avoids the conflict scene). Administered using peer-rating methodologies, respondents evaluate each classroom peer on a 3-point Likert-type behavioral frequency scale ranging from 0 (Never) to 2 (Often). Psychometric investigations consistently confirm robust internal consistency across subscales, with Cronbach’s alpha coefficients typically ranging between α = .88 and α = .95 (Bully: .93; Assistant: .95; Reinforcer: .90; Defender: .89; Outsider: .88). Factor analytic studies demonstrate clear five-factor dimensional structures with high convergent and discriminant validity across socio-metric statuses, longitudinal anti-bullying interventions (such as the KiVa program), and behavioral observations. The PRQ remains the global gold standard for diagnostic assessment, epidemiological research, and empirical evaluation of classroom social ecology.

Keywords

Participant Role Questionnaire, school bullying, peer victimization, participant roles, bystander effect, peer nominations, KiVa anti-bullying program, social ecology, aggressive behavior, defender behavior

Authors

The Participant Role Questionnaire was conceived, developed, and validated through the research laboratory of developmental psychologist Dr. Christina Salmivalli at the University of Turku, Finland, in collaboration with eminent developmental researchers Kirsti Lagerspetz, Kaj Björkqvist, Karin Österman, and Ari Kaukiainen, alongside psychometric refinements in subsequent studies with Marinus Voeten.

• Christina Salmivalli, PhD — Professor of Psychology, Department of Psychology and INVEST Flagship Research Center, University of Turku, Turku, Finland. (Lead developer of the participant role approach and co-founder of the KiVa Anti-Bullying Program; Contact: [email protected]).
• Kirsti M. J. Lagerspetz, PhD (Late) — Department of Psychology, University of Turku, Finland.
• Kaj Björkqvist, PhD — Professor Emeritus of Developmental Psychology, Åbo Akademi University, Vasa, Finland.
• Karin Österman, PhD — Department of Social Sciences, Åbo Akademi University, Vasa, Finland.
• Ari Kaukiainen, PhD — Department of Psychology, University of Turku, Finland.
• Marinus J. M. Voeten, PhD — Associate Professor of Psychometrics and Research Methodology, Radboud University Nijmegen, The Netherlands.

Purpose

The fundamental purpose of the Participant Role Questionnaire (PRQ) is to operationalize, quantify, and empirically disentangle the complex social-ecological architecture of bullying within peer groups. For decades, developmental psychology and psychiatric epidemiology conceptualized school harassment almost exclusively as a closed, bidirectional dyad comprised solely of an aggressive perpetrator (“bully”) and an exploited target (“victim”). While that dyadic framework illuminated individual personality deficits, family risk factors, and acute trauma profiles, it failed to explain why bullying behaviors persist, escalate, or terminate within stable peer social environments such as school classrooms.

Grounded in the landmark insight that peer harassment is inherently a group process maintained by peer audience dynamics, social rewards, and systemic social status striving, the PRQ was constructed to fulfill multiple clinical, developmental, and evaluative purposes:

• Deconstructing Bystander Behavior: The PRQ systematically differentiates seemingly homogeneous “bystanders” into functionally distinct behavioral cohorts: active assistants who join physical or verbal attacks, passive or vocal reinforcers who supply positive reinforcement through social approbation or amusement, prosocial defenders who alleviate victim distress or confront aggressors, and outsiders who withdraw or practice active indifference.
• Socio-Ecological Classroom Diagnostics: In educational and school psychological contexts, the tool allows educators, school psychologists, and intervention teams to diagnose the micro-culture and normative climate of specific classrooms. Rather than targeting only identified perpetrators, professionals utilize PRQ profiles to understand which students maintain the bully’s power base and identify latent defender resources that can be mobilized to support targets.
• Theoretical Modeling of Social Striving: In academic research, the instrument serves to examine structural equations connecting bullying roles with sociological indices such as perceived popularity, sociometric popularity, peer rejection, social dominance, empathy, moral disengagement, and emotional regulation.
• Intervention Efficacy Benchmarking: The PRQ serves as the bedrock measurement engine for evidence-based anti-bullying programs, most notably the internationally validated KiVa program. By evaluating pre- and post-intervention changes in Defender, Reinforcer, and Assistant scores, researchers can determine whether behavioral shifts in the broader peer group drive long-term reductions in target victimization.

Psychological Construct

The participant role construct posits that school bullying is not an isolated psychiatric pathology or an uncontrolled release of anger; rather, it is a deliberate, instrumental form of proactive social aggression intended to gain, secure, and display social status and peer group dominance. Because proactive aggression requires an audience to confer status rewards, the peer collective becomes intrinsically complicit, segregating itself into systematic behavioral positions. The PRQ evaluates five distinct psychological constructs, alongside the target/victim status:

1. The Bully (Ring-Leader / Initiator)

The Bully construct reflects intentional, unprovoked, and repetitive aggression characterized by an asymmetric power imbalance. Bullies are social architects who initiate hostile interactions, organize group assaults, and invent varied modalities of emotional, verbal, relational, or physical torment. Psychologically, high scores on this construct reflect high proactive aggression, heightened social intelligence harnessed for manipulative aims, moderate to high perceived popularity, and low affective empathy alongside advanced cognitive theory of mind.

2. The Assistant (Follower / Active Henchman)

The Assistant construct captures secondary perpetration. Unlike the primary bully, assistants rarely organize or inaugurate hostile actions independently. Instead, they act as cooperative executioners once the initial assault has been initiated. They physically hold down victims, repeat abusive epithets, or surround targets to block egress. This construct reflects high social conformity, peer-directed submissiveness, desire for affiliation with dominant peers, and diffused responsibility within group contagion.

3. The Reinforcer (Audience / Social Rewarder)

The Reinforcer construct captures behavioral feedback loops that incentivize the perpetrator. Reinforcers provide the immediate social currency that motivates bullying: laughter, active gathering around the altercation, smiling, and overt verbal encouragement (“Show him!” or “Keep going!”). Even when remaining physically non-violent, reinforcers validate the bully’s perceived social dominance and signal to the victim that the classroom consensus condones their degradation. Psychometrically, this dimension correlates strongly with moral disengagement, high fear of negative evaluation, and low moral courage.

4. The Defender (Prosocial Protector / Resister)

The Defender construct embodies prosocial intervention, active altruism, and social justice within the peer network. Defenders engage in high-risk moral actions aimed at ending the harassment or comforting the victim. This entails directly confronting perpetrators, ordering aggressors to stop, seeking institutional support from teachers, and emotionally supporting the victim during or post-incident. Psychologically, defenders are characterized by elevated affective empathy, strong moral identity, high self-efficacy for intervention, and resilient personal sociometric status that withstands potential social retaliation.

5. The Outsider (Passive Bystander / Withdrawn Observer)

The Outsider construct represents deliberate detachment, physical withdrawal, or apparent neutrality during peer victimization episodes. Outsiders routinely pretend not to notice ongoing abuse, walk away from the scene, or claim a position of absolute non-involvement (“It is not my business”). Psychologically, outsider behavior does not necessarily denote indifference; extensive research demonstrates that many outsiders experience severe internal conflict, acute anxiety, and vicarious guilt, yet remain paralyzed by social anxiety, fear of becoming the next target, or the classical bystander effect.

Theoretical Framework

The Participant Role Questionnaire is theoretically anchored at the convergence of three seminal paradigms in developmental, cognitive, and social psychology:

1. Social-Ecological Systems Theory

Grounded in Urie Bronfenbrenner’s Ecological Systems Theory, bullying is conceptualized as an emergent property of the micro-systemic classroom ecology. The classroom represents an enclosed, captive social collective where adolescents must navigate peer hierarchy and social visibility. Bullying cannot occur in an ecological vacuum; it is maintained by the distributed actions, non-actions, and feedback signals of all interacting members in the ecosystem.

2. Social Learning and Reinforcement Theory

Drawing from Albert Bandura’s Social Cognitive Theory and operant conditioning principles, bullying is understood as behavior governed by anticipated social reinforcement. When bullies assault peers, the tangible rewards are not merely the material items extorted, but the elevated status, attention, and compliance conferred by peers. By cheering or laughing, reinforcers deliver positive reinforcement; by looking away or remaining silent, outsiders deliver negative reinforcement by eliminating obstacles to aggression and confirming social impunity.

3. Theory of Moral Disengagement

Bandura’s construct of moral disengagement plays a critical explanatory role in the PRQ framework. The differentiated roles illustrate how children deactivate self-regulatory moral sanctions across varied degrees of psychological distancing. Assistants and reinforcers practice diffusion and displacement of responsibility (“I was just following along”; “Everyone was laughing”), while outsiders practice distorting consequences or cognitive blunting (“It wasn’t that bad”; “They were just playing”). Conversely, defenders possess high moral engagement, overcoming pluralistic ignorance and diffusion of responsibility.

Validity

The Participant Role Questionnaire has undergone rigorous empirical validation across diverse cross-cultural settings, consistently demonstrating robust construct, convergent, discriminant, and predictive validity.

Construct and Structural Validity

In their landmark validation study, Salmivalli et al. (1996) examined 573 Finnish sixth-grade students (ages 12–13). Multitrait-multimethod analyses demonstrated that the PRQ distinctively isolated the theoretical dimensions with minimal cross-loadings. Peer-ratings demonstrated marked discriminability between active aggression (Bully, Assistant) and audience behavior (Reinforcer), validating the conceptual bifurcation between leading aggression and maintaining social reinforcement.

Convergent and Discriminant Validity

Extensive investigations have correlated PRQ dimension scores with established clinical and personality inventories:

• Bully and Assistant Scales: Correlate strongly and positively with direct, indirect, and physical aggression scales (e.g., Direct and Indirect Aggression Scales [DIAS]; r = .65 to .78), hyperactive-impulsive behaviors, and measures of callous-unemotional traits. They correlate negatively with affective empathy and prosocial orientation.
• Defender Scale: Exhibits strong positive correlations with standardized empathy measures (such as the Interpersonal Reactivity Index; r = .52 to .64), social competence inventories, and prosocial nomination metrics. It demonstrates strong negative correlations with moral disengagement and Machiavellianism.
• Outsider Scale: Correlates positively with social withdrawal, internalized shyness, and trait anxiety, while showing near-zero or non-significant correlations with proactive aggression indices, establishing clear discriminant validity from the bully/assistant cluster.

Predictive and Ecological Validity

Observational validation conducted by researchers utilizing naturalistic playground video recordings and wireless audio transmitters (e.g., Pepler, Craig, & O’Connell, 1999; Salmivalli et al., 2004) revealed striking concordance between PRQ peer-nomination role classifications and real-world behavioral occurrences during live bullying incidents. Students peer-identified as reinforcers were observed joining the bullying circle and smiling or cheering in over 80% of recorded incidents, confirming exceptional ecological validity.

Reliability

The PRQ consistently demonstrates outstanding psychometric reliability across short, long, peer-reported, and self-reported iterations:

Internal Consistency

In standard peer-rating administrations, the 15-item short-form PRQ yields high Cronbach’s alpha coefficients, reflecting tightly bound latent constructs within each 3-item subscale:

• Bully Scale: α = .93
• Assistant Scale: α = .95
• Reinforcer Scale: α = .90
• Defender Scale: α = .89
• Outsider Scale: α = .88

The composite reliability across the broader 49-item and 23-item validation forms developed across Nordic and pan-European cohorts demonstrates alpha values consistently exceeding α = .85 for all operational subscales.

Test-Retest Stability

Longitudinal stability evaluations over 6-month and 12-month intervals have documented significant test-retest reliability. Stability coefficients for the Bully subscale typically range from r = .60 to .75 across an academic year, reflecting the enduring social position of bullies within intact classroom groups. Defender behaviors similarly show stable intraclass correlation coefficients (ICC = .68 to .74). Outsider and Reinforcer roles show slightly lower but solid temporal stability (r = .48 to .62), reflecting shifts in peer group structure, classroom seating arrangements, and teacher supervision vigilance.

Factor Analysis

Both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) provide conclusive empirical support for the multi-dimensional structure of the PRQ.

Exploratory Factor Analysis

In initial scale construction using principal axis factoring with oblimin (oblique) rotation, eigenvalues supported a 5-factor solution accounting for over 68% of the total variance. The five factors corresponded precisely to the theoretical roles: Bully, Assistant, Reinforcer, Defender, and Outsider. While the Bully and Assistant dimensions shared moderate to high inter-factor correlation (r ≈ .62 to .71), and Bully and Reinforcer correlated positively (r ≈ .55), item loadings loaded distinctly onto their primary factors (> .65), with cross-loadings remaining below .25.

Confirmatory Factor Analysis and Model Fit

Subsequent psychometric evaluations utilizing robust maximum likelihood estimation (MLR) across multi-national samples (e.g., Salmivalli & Voeten, 2004; Goossens et al., 2006) confirmed that the hypothesized 5-factor correlated model provided superior fit over competing unidimensional, two-factor (Aggressor vs. Non-Aggressor), or three-factor models:

• Comparative Fit Index (CFI): .96 to .98
• Tucker-Lewis Index (TLI): .95 to .97
• Root Mean Square Error of Approximation (RMSEA): .042 (90% CI [.036, .049])
• Standardized Root Mean Square Residual (SRMR): .038

These fit indices establish that while participant roles are interrelated within the ecological network, they represent distinct, non-redundant behavioral manifestations that cannot be collapsed into crude dichotomies.

Instrument / Measurement Tool

The standard peer-rating Participant Role Questionnaire is administered to intact classroom cohorts where students have spent substantial time together (typically a minimum of 2 to 3 months) to ensure reliable familiarity with peer behaviors.

• Test Type: Peer-nomination / Peer-rating behavioral socio-metric questionnaire.
• Target Population: Elementary, middle, and early secondary school students (Grades 3 through 10; ages 9 to 16).
• Administration Format: Classroom-based administration, available via printed rosters or computer/tablet digital testing portals.
• Number of Items: 15 items (comprising five 3-item subscales: Bully, Assistant, Reinforcer, Defender, Outsider).
• Classroom Roster Design: The names of every participating student in the classroom are pre-printed along one axis, while the behavioral items are displayed on the other axis. Each respondent assesses every peer in the room on each item.
• Response Format: 3-point behavioral frequency rating scale:
  • 0 = Never
  • 1 = Sometimes
  • 2 = Often
• Scoring and Transformation:
  • Individual Item Peer-Score: For each student, raw ratings received from all responding classmates are summed and divided by the number of nominators (N − 1, excluding self-evaluations) to yield an item peer-mean between 0.00 and 2.00.
  • Subscale Scale Score: The mean of the three corresponding item scores is calculated for each of the five subscales (Bully, Assistant, Reinforcer, Defender, Outsider).
  • Standardization: Subscale raw scores are standardized into Z-scores within the classroom or within gender cohorts across the sample (mean = 0, standard deviation = 1).
  • Categorical Assignment (Optional): A student is classified into a specific dominant participant role if their standardized Z-score on that scale exceeds a predefined threshold (conventionally Z > +1.00) and is higher than their Z-score on any other subscale. Students who do not meet the threshold on any scale are categorized as “Unclassified” or “No Specific Role.”

Permissions & Fee and Test Year

The original Participant Role Questionnaire was published in 1996 in Aggressive Behavior by John Wiley & Sons, Inc., with psychometric adaptations and peer-norm connection studies published in 2004 in the International Journal of Behavioral Development. The instrument and its items are protected under copyright (© 1996 John Wiley & Sons, Inc.).

Licensing and Academic Accessibility: For independent academic, non-commercial scientific research and doctoral dissertations, the PRQ is widely accessible for research use. Researchers must secure formal authorization and follow copyright guidelines prior to full-scale reproduction or digital deployment. Portions of the tool, including the 15-item short-form, have been cataloged for diagnostic reference in the United States Centers for Disease Control and Prevention (CDC) compendium: Measuring Bullying Victimization, Perpetration, and Bystander Experiences: A Compendium of Assessment Tools (Hamburger et al., 2011, p. 101).

For research permissions, institutional translations, or operational inquiries regarding the implementation of the PRQ within comprehensive prevention programs, inquiries should be directed to the primary developer:

Christina Salmivalli, PhD
Department of Psychology, University of Turku
FIN-20014 Turku, Finland
Email: [email protected]

References

• Goossens, F. A., Olthof, T., & Dekker, P. H. (2006). New participant roles in bullying: An exploratory study. Aggressive Behavior, 32(3), 247–257. https://doi.org/10.1002/ab.20130
• Hamburger, M. E., Vivolo, A. M., & O’Flaherty, C. M. (2011). Measuring Bullying Victimization, Perpetration, and Bystander Experiences: A Compendium of Assessment Tools. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/bullycompendiumbk-a.pdf
• Pepler, D. J., Craig, W. M., & O’Connell, P. (1999). Understanding bullying from a dynamic systems perspective. In A. Slater & D. Muir (Eds.), The Blackwell Reader in Developmental Psychology (pp. 440–453). Blackwell.
• Salmivalli, C., Lagerspetz, K., Björkqvist, K., Österman, K., & Kaukiainen, A. (1996). Bullying as a group process: Participant roles and their relations to social status within the group. Aggressive Behavior, 22(1), 1–15. https://doi.org/10.1002/(SICI)1098-2337(1996)22:1<1::AID-AB1>3.0.CO;2-T
• Salmivalli, C., & Voeten, M. (2004). Connections between attitudes, group norms, and behaviors associated with bullying in schools. International Journal of Behavioral Development, 28(3), 246–258. https://doi.org/10.1080/01650250344000488
• Sutton, J., & Smith, P. K. (1999). Bullying as a group process: An adaptation of the Salmivalli et al. (1996) Participant Role Questionnaire. Aggressive Behavior, 25(2), 97–111. https://doi.org/10.1002/(SICI)1098-2337(1999)25:2<97::AID-AB3>3.0.CO;2-7

13. Items of the Scale (Questionnaire)