Delinquent Peers—Rochester Youth Development Study

Abstract

The Delinquent Peers—Rochester Youth Development Study (RYDS) scale is an 8-item psychometric instrument designed to assess an adolescent’s perception of delinquent behavior among their close peer network. Originating from the seminal longitudinal project established in 1988 by Terence P. Thornberry and colleagues, this instrument was engineered to operationalize peer deviance within the framework of Interactional Theory. The scale surveys peer involvement across eight delinquent behaviors spanning three major domains: violent offenses (e.g., weapon use, physical assault), property crimes (e.g., grand theft, petty theft, property destruction, unauthorized motor vehicle operation), and status or minor normative infractions (e.g., truancy). Each item is evaluated across a 4-point Likert-type response scale ranging from 1 (None of them) to 4 (Most of them), framed within a 30-day retrospective recall window. The total score is computed via a summative composite ranging from 8 to 32, with higher scores reflecting elevated immersion within deviant peer structures.

Extensive psychometric investigations across diverse demographic cohorts within the RYDS sample (initially consisting of 1,000 panel members followed longitudinally from early adolescence to age 31) demonstrate robust internal consistency, with Cronbach’s alpha coefficients consistently spanning .82 to .89 across multiple assessment waves. Confirmatory factor analytic investigations demonstrate an essentially unidimensional latent construct of perceived peer delinquency, while supporting measurement invariance across sex, race, and socioeconomic strata. The scale exhibits outstanding criterion, convergent, and predictive validity, establishing peer delinquency as one of the most potent proximal predictors of self-reported juvenile delinquency, substance abuse, and official juvenile arrest records. As an open-access measure cataloged by the Centers for Disease Control and Prevention (CDC), the instrument remains a foundational assessment tool in developmental criminology, forensic psychology, and adolescent violence prevention research.

Keywords

Delinquent Peers Scale, Rochester Youth Development Study, RYDS, Interactional Theory, Peer Deviance, Adolescent Delinquency, Juvenile Violence, Psychometrics, Differential Association, Longitudinal Criminology, Behavioral Assessment, Peer Socialization, Risk Assessment.

Authors

The Delinquent Peers Scale was developed as an integral measurement component of the Rochester Youth Development Study (RYDS) by a multidisciplinary team of developmental criminologists and sociologists:

• Terence P. Thornberry, Ph.D. — Principal Investigator of the Rochester Youth Development Study; Distinguished Professor Emeritus, Department of Criminology and Criminal Justice, University of Maryland, College Park; formerly Professor at the School of Criminal Justice, University at Albany, State University of New York (SUNY).
• Alan J. Lizotte, Ph.D. — Co-Principal Investigator; Distinguished Professor Emeritus, School of Criminal Justice, University at Albany, SUNY.
• Marvin D. Krohn, Ph.D. — Co-Principal Investigator; Professor, Department of Sociology and Criminology & Law, University of Florida; formerly at University at Albany, SUNY.
• Margaret Farnworth, Ph.D. — Investigator, Department of Criminal Justice, Sam Houston State University; formerly research faculty at University at Albany, SUNY.
• Sung Joon Jang, Ph.D. — Research Professor of Criminology, Institute for Studies of Religion, Baylor University.

The research was conducted under the administrative and methodological umbrella of the Hindelang Criminal Justice Research Center, University at Albany, State University of New York, supported through substantial funding by the Office of Juvenile Justice and Delinquency Prevention (OJJDP), the National Institute on Drug Abuse (NIDA), and the National Science Foundation (NSF).

Purpose

The primary purpose of the Delinquent Peers Scale is to capture an empirical index of an adolescent’s exposure to, and immersion within, a peer ecology characterized by rule-breaking and criminal activity. In developmental and life-course criminology, association with deviant peers represents one of the single most robust correlates and predictors of individual delinquency, youth violence, and antisocial conduct. However, measuring peer influence poses unique methodological challenges; obtaining direct network-wide sociometric data from all nominated peers is logistically difficult, expensive, and frequently hindered by non-response from delinquent affiliates. The RYDS Delinquent Peers Scale was formulated as an efficient, self-administered target-respondent instrument capable of capturing the respondent’s proximal perception of their immediate peer circle’s illegal behavior.

Clinically and pragmatically, the scale serves critical diagnostic, risk-stratification, and evaluative purposes:

• Forensic and Juvenile Justice Risk Assessment: Within juvenile diversion programs, probation monitoring, and detention settings, identifying whether a youth’s social milieu is dominated by deviant peers provides actionable guidance for intervention planning, such as implementing cognitive-behavioral programs that disrupt delinquent peer networks.
• School-Based Prevention Screening: The instrument can be deployed within secondary school environments to identify high-risk adolescents who are affiliating with anti-social peer networks prior to the escalation of severe violence or persistent truancy.
• Theoretical Research on Delinquency Etiology: In longitudinal criminological research, the measure enables structural equation modeling and cross-lagged panel analyses to disentangle the bidirectional mechanisms of “peer selection” (deviant individuals selecting deviant friends) versus “peer socialization” (conforming individuals adopting antisocial patterns through peer contagion).
• Intervention Efficacy Monitoring: Community violence-reduction programs use the scale to monitor shifts in an adolescent’s peer affiliations over time, evaluating whether therapeutic foster care, multi-systemic therapy (MST), or family counseling successfully detaches the youth from delinquent networks.

Psychological Construct

The psychological and sociological construct measured by this instrument is perceived peer delinquency—defined as the perceived prevalence and concentration of illegal, aggressive, and non-normative behaviors enacted by an adolescent’s self-defined circle of close friends. This construct operates at the intersection of social cognition, peer contextual socialization, and criminology.

Perceived peer delinquency is not merely an objective count of peer crimes; it reflects the adolescent’s mental representation of the behavioral norms governing their peer reference group. Psychologists conceptualize this construct as having two core elements:

1. Social Modeling and Vicarious Reinforcement: When an individual observes friends engaging in weapon use, assault, theft, or truancy without experiencing immediate adverse consequences (or receiving peer admiration and elevated status), the perceived utility and permissibility of antisocial behavior increase.
2. Normative Climate and Social Expectancy: The presence of friends who violate legal norms signals to the adolescent that deviant acts are endorsed or expected within the micro-culture of the group.

The scale encompasses three behavioral manifestations of peer deviance:

• Interpersonal Violence and Weapon-Related Offending: Captured by items querying whether friends used a weapon or force to extract money or possessions, attacked someone with a weapon or intention of serious injury, or physically struck someone with harmful intent. These items measure exposure to instrumental and expressive aggression, which strongly predict severe adolescent violence.
• Property Crime and Material Acquisition: Captured by items regarding grand theft (property valued over $100), minor theft (property valued between$5 and $50), purposeful destruction of external property (vandalism), and unauthorized joyriding or motor vehicle theft. These behaviors reflect peer involvement in covert delinquency, economic offending, and reckless sensation seeking.
• Status Offenses and Institutional Non-Compliance: Exemplified by unexcused school truancy (“skipped classes without an excuse”), this component evaluates low-level institutional detachment and defiance of school authority, which often serves as a gateway to escalated delinquent trajectories.

While some criminologists differentiate between “perceived” peer delinquency and “actual” peer delinquency (measured by interviewing the peers directly), psychometric research demonstrates that perceived peer delinquency often exerts an even stronger proximal influence on an adolescent’s subjective decision-making than objective peer actions, because individuals act upon their internal interpretations of reality.

Theoretical Framework

The development of the Delinquent Peers Scale is anchored directly in Interactional Theory, formulated by Terence P. Thornberry (1987) and empirically tested in landmark papers such as Thornberry, Lizotte, Krohn, Farnworth, and Jang (1994). Interactional Theory represents a comprehensive synthesis of Social Control Theory (Hirschi, 1969) and Differential Association Theory / Social Learning Theory (Sutherland, 1947; Akers, 1973).

Prior to Interactional Theory, developmental criminology was divided by unidirectional explanatory models:

• Social Control Theory argued that weak bonds to conventional society (attachment to parents, commitment to school, involvement in prosocial activities, belief in conventional values) free the individual to pursue deviant desires. In this view, association with delinquent peers is a mere epiphenomenon—delinquent individuals choose like-minded peers because they already lack conventional bonds (“birds of a feather flock together”).
• Differential Association and Social Learning Theories argued that delinquency is entirely learned through social interaction in intimate groups. Individuals become delinquent through exposure to definitions favorable to violation of law, imitation of deviant models, and differential reinforcement (“peer socialization”).

Thornberry’s Interactional Theory bridged this divide by positing that human behavior occurs within dynamic, bidirectional, and reciprocal feedback loops across the life course. Weakened conventional bonds do not automatically cause delinquency; rather, they attenuate the adolescent’s protective barriers, facilitating entry into delinquent peer networks. Once immersed in a delinquent peer group, social learning mechanisms activate: peers reinforce antisocial definitions, provide practical opportunities for crime, and punish conventional conformity. In turn, engaging in delinquent behavior further solidifies affiliation with delinquent peers while eroding remaining parental and academic bonds.

The 8-item scale was constructed specifically to capture the dynamic peer input within this non-recursive system. By using a 30-day recall timeframe, the instrument was calibrated to be administered repeatedly across multiple semi-annual and annual waves of data collection, enabling panel analyses that isolate how changes in perceived peer delinquency predict subsequent individual offending while controlling for prior baseline offending.

Validity

The Delinquent Peers Scale has undergone extensive psychometric validation within the Rochester Youth Development Study and subsequent independent replications across international samples.

Construct Validity

Construct validity is substantiated through structural equation modeling (SEM) and factor analytic procedures confirming that the 8 items load heavily on a coherent latent factor representing peer deviance. Factor loadings across longitudinal panels consistently exceed .60, indicating that the items share substantial common variance. Furthermore, measurement invariance tests across male and female adolescents, as well as across African American, Hispanic, and Caucasian subgroups, confirm metric and scalar invariance, demonstrating that the construct is interpreted equivalently across diverse demographic cohorts.

Predictive and Criterion Validity

The predictive power of the scale is exceptionally well-documented. Longitudinal analyses from the RYDS cohort demonstrated that scores on the Delinquent Peers Scale at Wave 2 and Wave 3 were among the single strongest predictors of an adolescent’s self-reported delinquent offending at subsequent waves, even after controlling for baseline delinquency, family attachment, and socio-demographic factors (Thornberry et al., 1994). Furthermore, high scores on the scale significantly predict:

• Subsequent police contact and official juvenile arrest records (OJJDP reports).
• Adolescent gang involvement and sustained gang membership.
• Substance initiation, polysubstance abuse, and severe alcohol intoxication.
• Violent offending trajectories extending into emerging adulthood.

Convergent and Discriminant Validity

Convergent validity is illustrated by strong, statistically significant positive correlations (typically $r = .45$ to $.65$) with allied constructs, including individual antisocial attitudes, neutralization of guilt, moral disengagement, and school disciplinary sanctions. Discriminant validity is evidenced by moderate-to-low negative correlations with prosocial constructs, such as parental attachment ($r = -.25$ to $-.35$), commitment to schooling ($r = -.30$ to $-.42$), and conventional religious involvement, showing that the scale does not simply measure generalized social alienation, but specifically captures antisocial peer immersion.

Reliability

The scale possesses high internal consistency across multiple developmental periods, racial and ethnic groups, and socioeconomic strata:

• Internal Consistency: In the primary validation studies of the RYDS cohort (Thornberry et al., 1994; Dahlberg et al., 2005), Cronbach’s alpha coefficients consistently ranged from .82 to .89 across waves 1 through 9. Analyses focusing specifically on the violent items yield subscale alphas around .78 to .83, while the comprehensive 8-item composite routinely exceeds the .80 threshold recommended for empirical research.
• Composite Reliability: Confirmatory factor analytic evaluations have reported composite reliability (McDonald’s $\omega$) coefficients exceeding .85, confirming that measurement error remains low across the full spectrum of peer behaviors.
• Stability and Test-Retest Considerations: In short-term assessments (e.g., 2- to 4-week intervals in methodological sub-studies), the instrument demonstrates high test-retest reliability ($r > .75$). Over longer longitudinal spans (e.g., 6 months to 1 year), stability coefficients are moderate ($r = .40$ to $.60$), accurately reflecting developmental flux and real-world turnover in adolescent friendship networks rather than psychometric instability.

Factor Analysis

Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) conducted on the 8 items indicate a robust, predominantly unidimensional factor structure, with all eight items loading onto a single underlying peer delinquency dimension.

Principal axis factoring with oblique rotation typically reveals a primary eigenvalue exceeding 3.50, accounting for 45% to 55% of the total item variance. While some secondary models reveal a bifactor or two-factor structure distinguishing violent delinquency (items 1, 2, 3) from non-violent/property/status delinquency (items 4, 5, 6, 7, 8), the correlation between these two latent dimensions routinely exceeds $r = .70$. Consequently, developmental criminologists strongly advocate the use of the unified unidimensional sum or mean score.

In CFA investigations, a single-factor model yields excellent goodness-of-fit indices across adolescent samples when minor correlated error terms between similarly phrased violent items (e.g., items 2 and 3) are specified:

• Comparative Fit Index (CFI): $ge .95$
• Tucker-Lewis Index (TLI): $ge .94$
• Root Mean Square Error of Approximation (RMSEA): $le .055$ (90% CI: .042–.068)
• Standardized Root Mean Square Residual (SRMR): $le .040$

Standardized factor loadings across all eight items generally range from .60 to .83. Notably, items involving direct violence (“Hit someone with the idea of hurting them”) and property theft (“Stole something worth more than $5 but less than$50″) consistently demonstrate the highest loadings (.72–.82), confirming their centrality to the latent construct of peer delinquency.

Instrument / Measurement Tool

• Instrument Name: Delinquent Peers Scale — Rochester Youth Development Study (RYDS)
• Authors / Originators: Terence P. Thornberry, Alan J. Lizotte, Marvin D. Krohn, Margaret Farnworth, and Sung Joon Jang (1994)
• Administration Format: Self-administered paper-and-pencil questionnaire, structured interview, or computer-assisted self-interview (CASI)
• Target Population: Adolescents and young adults (approximately ages 11 to 21)
• Item Count: 8 items
• Recall Window: Preceding 30 days (“In the past 30 days, how many of your friends …”)
• Response Scale: 4-point Likert-type frequency scale:
  • 1 = None of them
  • 2 = A few of them
  • 3 = Some of them
  • 4 = Most of them
• Scoring Rules:
  • Each of the 8 items is assigned an integer value from 1 to 4 corresponding to the chosen response option.
  • Total score is calculated by summing all 8 items: $\text{Total Score} = \sum_{i=1}^{8} \text{Item}_i$.
  • Possible score range: 8 to 32.
  • Alternatively, researchers frequently compute a mean composite score (ranging from 1.00 to 4.00) by dividing the sum by 8, which preserves the original response metric.
  • There are no reverse-coded items.
  • Interpretation: Higher scores denote a higher proportion of delinquent peers in the respondent’s proximal peer group. Scores of 8 indicate zero reported peer delinquency; intermediate scores (9–16) indicate low to moderate peer deviance (typically concentrated in minor status infractions like truancy); scores above 17 reflect significant exposure to serious violent and property-offending peer circles.
• Completion Time: Approximately 3 to 5 minutes

Permissions & Fee and Test Year

The Delinquent Peers Scale was conceptualized during the inception of the Rochester Youth Development Study in 1988–1989 and formally detailed in landmark peer-reviewed criminological publications in 1994 (e.g., Thornberry et al., 1994). Because the instrument was developed under public research grants awarded by the federal government—specifically the Office of Juvenile Justice and Delinquency Prevention (OJJDP, Grant #86-JN-CX-0007) and the National Institute on Drug Abuse (NIDA, Grant #DA-05512)—it resides in the public domain.

The scale was subsequently anthologized by the Centers for Disease Control and Prevention (CDC) in their public compendium, Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (Dahlberg et al., 2005, p. 216). As such, there are no licensing fees, royalties, or formal permissions required for academic, clinical, or non-commercial research use. Researchers and clinicians are expected to cite the original authors (Thornberry et al., 1994) and the CDC compendium in their publications.

References

• Akers, R. L. (1973). Deviant behavior: A social learning approach. Wadsworth Publishing Company.
• 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., p. 216). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/YV_Compendium.pdf
• Hirschi, T. (1969). Causes of delinquency. University of California Press. https://doi.org/10.1525/9780520310247
• Krohn, M. D., Lizotte, A. J., Thornberry, T. P., Smith, C. A., & McDowall, D. (Eds.). (2019). Handbook on crime and deviance (2nd ed.). Springer. https://doi.org/10.1007/978-3-030-20779-3
• Sutherland, E. H. (1947). Principles of criminology (4th ed.). J.B. Lippincott.
• Thornberry, T. P. (1987). Toward an interactional theory of delinquency. Criminology, 25(4), 863–892. https://doi.org/10.1111/j.1745-9125.1987.tb00823.x
• Thornberry, T. P., Lizotte, A. J., Krohn, M. D., Farnworth, M., & Jang, S. J. (1994). Delinquent peers, beliefs, and delinquent behavior: A longitudinal test of interactional theory. Criminology, 32(1), 47–83. https://doi.org/10.1111/j.1745-9125.1994.tb01146.x
• Thornberry, T. P., Lizotte, A. J., Krohn, M. D., Smith, C. A., & Porter, P. K. (2003). Causes and consequences of delinquency: Findings from the Rochester Youth Development Study. In T. P. Thornberry & M. D. Krohn (Eds.), Taking stock of delinquency: An overview of findings from contemporary longitudinal studies (pp. 11–46). Kluwer Academic/Plenum Publishers. https://doi.org/10.1007/0-306-47945-1_2

13. Items of the Scale (Questionnaire)

Below are the authentic scale items in their original language as published in the standard psychometric validation studies, without modification or translation to preserve instrument validity and reliability:
1
Used a weapon or force to get money or things from people?
2
Attacked someone with a weapon or with the idea of seriously hurting them?
3
Hit someone with the idea of hurting them?
4
Stole something worth more than $100?
5
Stole something worth more than $5 but less than$50?
6
Damaged or destroyed someone else’s property on purpose?
7
Took a car for a ride or drive without the owner’s permission?
8
Skipped classes without an excuse?
★
Rate This Scale
5.0 / 5 • 1 vote