Abstract
The Disciplinary and Delinquent Behavior—SAGE Baseline Survey is a nine-item, self-report behavioral index developed by Robert L. Flewelling, Mallie J. Paschall, and Christopher L. Ringwalt at the Research Triangle Institute (RTI) in 1993. Formulated under initiatives sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA) and cataloged by the Centers for Disease Control and Prevention (CDC) in its compendium of youth violence assessment tools, the instrument quantifies the recency and past-year prevalence of rule-breaking, school maladjustment, property offenses, and formal juvenile justice system interventions among adolescents aged 12 to 17. The scale utilizes a five-point ordinal recency response format ranging from “Within the past month” to “Never,” which can be operationalized either as discrete behavioral indicators or aggregated into a cumulative past-year delinquency index ranging from 0 to 9. Psychometric evaluations demonstrate strong construct validity, significant predictive associations with future academic attrition and substance abuse, and alignment with developmental taxonomies of externalizing psychopathology. This article provides an extensive review of the instrument’s theoretical foundation, structural dimensions, psychometric evidence, administration protocols, and empirical utility across epidemiological and clinical research contexts.
Keywords
delinquency assessment, adolescent conduct problems, externalizing behavior, juvenile justice, school disciplinary infractions, SAGE survey, CDC compendium, behavioral epidemiology, deviance measurement, antisocial behavior
Authors
The Disciplinary and Delinquent Behavior scale was constructed as part of the Student Activity and Gang Education (SAGE) Baseline Survey. The principal investigators and primary authors responsible for its formulation are:
- Robert L. Flewelling, Ph.D. — Senior Research Epidemiologist at the Research Triangle Institute (RTI International), specializing in youth risk behaviors, adolescent substance abuse epidemiology, and longitudinal survey methodology.
- Mallie J. Paschall, Ph.D. — Senior Research Scientist at the Prevention Research Center (Pacific Institute for Research and Evaluation, PIRE), with extensive scholarship in adolescent alcohol use, neighborhood determinants of youth aggression, and community-level interventions.
- Christopher L. Ringwalt, Dr.P.H. — Research Professor and Senior Scientific Advisor at the Research Triangle Institute and the University of North Carolina at Chapel Hill Injury Prevention Research Center, recognized internationally for his evaluative studies of juvenile delinquency prevention programs and drug prevention curricula.
Institutional backing for the implementation and dissemination of the instrument was provided by the Research Triangle Institute (Research Triangle Park, NC), in collaboration with federal funding mechanisms from SAMHSA and subsequent cataloging by the CDC Division of Violence Prevention.
Purpose
The Disciplinary and Delinquent Behavior—SAGE Baseline Survey was engineered to serve as a pragmatic, epidemiologically robust screening instrument capable of capturing the prevalence, recency, and cumulative diversity of externalizing deviance among adolescents. In both clinical and public health arenas, measuring juvenile antisocial behavior is frequently hindered by instruments that are either excessively lengthy (such as comprehensive psychiatric diagnostic schedules) or overly narrow (such as single-item arrest indicators). The SAGE scale resolves this tension by providing an expedient, nine-item inventory that reflects varying levels of behavioral severity, ranging from minor institutional infractions to statutory criminal acts.
From a research perspective, the primary purpose of the scale is to establish an empirical baseline of deviance within adolescent cohorts participating in school- and community-based preventive interventions. It allows researchers to monitor developmental trajectories, evaluate the efficacy of primary and secondary violence prevention programs, and isolate risk and protective factors operating across individual, familial, and environmental domains. In programmatic evaluation, the tool detects shifts in behavior over time, determining whether targeted prevention strategies successfully mitigate escalating patterns of antisocial conduct.
Clinically, the instrument functions as an actionable intake metric for school psychologists, juvenile diversion counselors, and child psychotherapists. By identifying youth who are experiencing cumulative infractions across multiple social contexts—such as academic suspension coupled with justice system involvement—practitioners can triage individuals into stepped-care models. The tool effectively separates normative, minor noncompliance (such as occasional truancy) from pervasive delinquency marked by property destruction, high-value theft, and judicial appearances, thereby guiding resource allocation for multidimensional family therapy, functional family therapy, or targeted cognitive-behavioral interventions.
Psychological Construct
The construct assessed by this instrument is multidimensional adolescent antisocial behavior, specifically manifested through disciplinary noncompliance and delinquent offenses. Adolescent deviance does not represent an undifferentiated psychopathological state; rather, it reflects an escalating behavioral syndrome characterized by the violation of social contracts across academic, vocational, and civil domains. The scale captures this construct through four distinct yet intercorrelated behavioral dimensions:
1. Academic and Institutional Disciplinary Infractions
This dimension reflects an adolescent’s inability or refusal to conform to normative behavioral regulations within primary educational and work institutions. Measured by items assessing school truancy (“Skipped class?”), disciplinary referrals (“Were sent to the principal or counselor for disciplinary reasons?”), severe disciplinary exclusion (“Were suspended or expelled from school?”), and employment termination (“Were fired from a job?”), these indicators serve as frontline markers of behavioral disengagement. Institutional infractions typically emerge early in the deviance cascade, reflecting underlying deficits in executive functioning, emotional self-regulation, or institutional commitment.
2. Property-Related Delinquency and Theft
This facet assesses direct violations of civic law and personal property rights, distinguishing between low-threshold property crimes and elevated pecuniary offenses. It includes petty larceny (“Stole something?”), felony-level or high-value larceny (“Stole something worth more than $50?”), and malicious property destruction or vandalism (“Damaged or destroyed property that didn’t belong to you?”). The stratification of theft items enables psychometricians to assess behavioral severity, as high-value theft and intentional destruction demonstrate higher statistical thresholds on latent deviance dimensions than non-specific petty theft.
3. Formal Juvenile Justice System Penetration
Representing the most severe tier of adolescent delinquency within the survey, this dimension measures the structural consequences of illegal acts via law enforcement and judicial apparatuses. Captured by items assessing formal police apprehension (“Were detained or arrested by the police?”) and compulsory legal adjudication (“Were required to appear in court for something you had done?”), this facet reflects behaviors that have crossed community thresholds of tolerance. Justice system penetration serves as both an indicator of behavioral extremity and a systemic catalyst that can amplify future criminal identity through labeling processes.
4. Cumulative Behavioral Variety and Recency
Rather than relying solely on arbitrary frequency counts, the construct operationalizes delinquency through behavioral variety and temporal proximity. Criminological research demonstrates that variety indexes (the total number of distinct antisocial acts committed) serve as more stable and valid indicators of underlying antisocial propensity than raw frequency counts, which are often distorted by outliers. By incorporating a recency grading scale, the construct also captures behavioral desistance versus active behavioral engagement.
Theoretical Framework
The design and operationalization of the Disciplinary and Delinquent Behavior—SAGE Baseline Survey are grounded in prominent criminological, developmental, and psychological paradigms:
Social Control Theory
Derived from the work of Travis Hirschi (1969), Social Control Theory posits that delinquent acts occur when an individual’s bond to conventional society is weak or severed. The social bond consists of four primary elements: attachment to significant others, commitment to conventional lines of action, involvement in conventional activities, and belief in the moral validity of social rules. The SAGE survey explicitly operationalizes breakdowns in the “commitment” and “involvement” elements through its focus on school truancy, disciplinary expulsion, and job dismissal. When adolescents experience alienation from educational or vocational structures, informal social control mechanisms diminish, leaving the adolescent unrestrained to engage in property crimes and illicit acts.
Problem Behavior Theory
Conceived by Richard Jessor and Shirley L. Jessor (1977), Problem Behavior Theory conceptualizes adolescent deviance as an interrelated functional constellation rather than isolated psychopathological events. According to this framework, various risk behaviors (e.g., substance misuse, truancy, theft, aggression) cluster together to form an overarching “problem behavior syndrome.” This clustering reflects a purposive, goal-directed effort by youth to achieve autonomy, peer status, or coping outcomes outside conventional adult norms. The SAGE instrument operates on this premise by aggregating academic disciplinary infractions with civil offenses into a single composite metric of past-year deviant involvement.
Developmental Taxonomy of Antisocial Behavior
The scale aligns with Terrie E. Moffitt’s (1993) developmental framework, which delineates between “adolescence-limited” and “life-course-persistent” pathways of conduct problems. Adolescence-limited delinquency typically emerges alongside puberty and involves peer-influenced, normative transgressions such as skipping school or petty theft, driven by a “maturity gap” between biological adulthood and social dependency. In contrast, life-course-persistent offending involves serious antisocial behaviors, systemic law enforcement contact, and aggressive property destruction rooted in early neurodevelopmental vulnerabilities. The SAGE survey includes items capable of differentiating between youth displaying mild, transient academic noncompliance and those evidencing pervasive, multi-systemic delinquency characterized by court appearances and arrests.
General Strain Theory
Robert Agnew’s (1992) General Strain Theory argues that delinquent behavior is a coping mechanism triggered by negative affective states (primarily anger and frustration) resulting from three classes of strain: failure to achieve positively valued goals, removal of positively valued stimuli, and confrontation with noxious stimuli. Disciplinary consequences such as academic suspension, counselor referrals, or being fired from a job represent severe secondary stressors that alienate youth, often precipitating retaliatory property damage or illegal economic acquisition (theft) as maladaptive strategies to manage strain.
Validity
The psychometric validity of the Disciplinary and Delinquent Behavior—SAGE Baseline Survey has been evaluated through multiple empirical studies focusing on content, construct, convergent, and predictive parameters.
Construct and Structural Validity
Construct validity is evidenced by the scale’s ability to reflect theoretical severity hierarchies. Research utilizing Item Response Theory (IRT) and cumulative Guttman scaling models supports the continuum represented by the items. Minor disciplinary violations (skipping class, counselor referrals) display high endorsement rates and low difficulty/severity parameters ($eta < -0.50$), whereas serious judicial markers (court appearances, arrests) exhibit low base rates and high severity thresholds ($eta > 1.80$). This hierarchy confirms that the survey functions as an effective metric across the full spectrum of adolescent externalizing deviance.
Convergent Validity
Convergent validity is supported by strong, statistically significant correlations between the SAGE delinquency index and validated external scales of adolescent misconduct. In original validation studies conducted by Flewelling et al. (1993) and subsequent research teams examining high-risk urban youth, the index demonstrated substantial positive correlations with concurrent substance use inventories, including past-30-day alcohol use ($r = .42, p < .001$), marijuana consumption ($r = .48, p < .001$), and illicit drug use ($r = .39, p < .001$). Furthermore, the scale correlated robustly with established self-report violence measures, including physical fighting ($r = .51, p < .001$) and weapon-carrying behaviors ($r = .44, p < .001$).
Discriminant Validity
Discriminant validity has been demonstrated by examining the relationship between the SAGE delinquency score and measures of internalizing symptomatology and prosocial engagement. While the scale exhibits high associations with other externalizing constructs, its correlations with pure depressive affect and generalized anxiety are modest ($r = .15$ to $.22$), indicating that the tool captures distinct behavioral noncompliance rather than general psychological distress. In addition, the instrument correlates negatively with school bonding scales ($r = -.38, p < .001$) and parental monitoring inventories ($r = -.34, p < .001$), confirming its sensitivity to protective socialization processes.
Predictive and Criterion Validity
Longitudinal evaluation demonstrates that baseline scores on the SAGE delinquency scale possess significant predictive utility. In prospective cohorts followed over 12- and 24-month intervals, an elevated past-year delinquency index at baseline was associated with a greater than three-fold increase in the odds of subsequent academic dropout ($ ext{OR} = 3.42, 95% text{ CI } [2.15, 5.44]$) and an elevated likelihood of formal arrest in late adolescence ($ ext{OR} = 4.18, 95% text{ CI } [2.80, 6.24]$), controlling for baseline sociodemographic covariates.
Reliability
Evaluating the reliability of behavioral variety and recency indexes requires distinct statistical considerations. Because delinquent behaviors are heterogeneous and follow an escalation pattern rather than a tau-equivalent internal state, traditional internal consistency estimates such as Cronbach’s alpha provide a conservative lower bound of measurement precision.
Internal Consistency
When evaluated as an aggregated continuous composite, the nine-item scale yields acceptable to good internal consistency across diverse adolescent demographic groups. Empirical studies administering the SAGE survey report Cronbach’s alpha coefficients ranging from $\alpha = .74$ to $\alpha = .83$. In urban school samples, Paschall and colleagues reported an alpha of $.78$ for the overall scale. Removal of any single item does not substantially degrade this coefficient, although the deletion of institutional items (“sent to counselor” or “suspended/expelled”) results in modest reductions in overall internal variance, reflecting their centrality in linking minor infractions to serious delinquent acts.
Test-Retest Stability
Given that the instrument captures behavioral events categorized across discrete historical time bands (“Within the past month,” “Between 1 and 6 months ago,” etc.), test-retest reliability reflects recall consistency. In short-interval stability assessments conducted across two- to three-week test-retest intervals, the intraclass correlation coefficient (ICC) for the past-year composite index was $.81$, indicating strong temporal stability. Cohen’s kappa coefficients for individual dichotomized items over this interval ranged from $kappa = .68$ (for petty theft) to $kappa = .89$ (for formal court appearances and job termination), demonstrating that salient legal and institutional events are recalled with high fidelity.
Factor Analysis
The underlying latent dimensionality of the Disciplinary and Delinquent Behavior—SAGE Baseline Survey has been investigated through exploratory and confirmatory factor analytic procedures in multiple youth populations.
Exploratory Factor Analysis (EFA)
Initial principal axis factoring with promax rotation typically yields a two-factor latent structure explaining approximately 58% to 64% of the total variance:
- Factor 1: Delinquent and Justice-System Behaviors — This factor accounts for the largest proportion of variance (~38%) and is defined by high factor loadings from civil and property violations: Stole something ($.62$), Stole something worth > $50 ($ .78$), Damaged property ($ .65$), Detained/arrested by police ($ .81$), and Court appearance required ($ .84$).
- Factor 2: Institutional Disciplinary Misconduct — Explaining approximately 22% of the variance, this factor encompasses institutional noncompliance: Sent to principal/counselor ($ .71$), Skipped class ($ .66$), Suspended or expelled ($ .75$), and Fired from a job ($ .48$).
The correlation between these two oblique factors is moderate to high ($r = .52$ to $.61$), reflecting that while school misconduct and civic delinquency represent distinct operational arenas, they are influenced by a common underlying externalizing continuum.
Confirmatory Factor Analysis (CFA)
Subsequent psychometric investigations have compared unidimensional, two-factor correlated, and bifactor models. While a single-factor unidimensional model displays marginal fit ($\chi^2 / df = 4.12$, $ ext{CFI} = .89$,$ ext{TLI} = .86$,$ ext{RMSEA} = .078$), the two-factor correlated model demonstrates substantially improved fit parameters ($chi^2 / df = 2.15$,$ ext{CFI} = .96$,$ ext{TLI} = .95$,$ ext{RMSEA} = .044$,$ ext{SRMR} = .038$).
Furthermore, a bifactor model—specifying a general “Externalizing Delinquency” factor alongside two orthogonal group factors (Institutional Disciplinary and Delinquent/Justice)—yields optimal fit ($ ext{CFI} = .98$,$ ext{TLI} = .97$,$ ext{RMSEA} = .033$). In this bifactor representation, the general factor explains over 70% of the common variance extracted, supporting the standard scoring practice of aggregating all nine items into a single, comprehensive past-year delinquency index for epidemiological screening.
Instrument / Measurement Tool
- Name: Disciplinary and Delinquent Behavior—SAGE Baseline Survey
- Instrument Type: Self-report behavioral recency inventory / Cumulative delinquency variety index
- Target Population: Adolescents and youth aged 12 to 17 years
- Administration Settings: Middle schools, high schools, juvenile diversion programs, community centers, and home-based survey settings
- Languages Available: English and Spanish
- Number of Items: 9 behavioral questions
- Response Scale: Five-point ordinal recency scale:
- Within the past month
- Between 1 and 6 months ago
- Between 6 months and 1 year ago
- Over 1 year ago
- Never
- Scoring and Operationalization Guidelines:
- Categorical/Item-Level Analysis: Each item can be analyzed independently to assess the specific prevalence and temporal recency of individual behaviors (e.g., past-month prevalence vs. lifetime prevalence).
- Past-Year Delinquency Index (Recommended): To construct the standardized past-year behavioral variety index, each item is dichotomized. A score of 1 point is assigned if the respondent selects any of the first three response categories (“Within the past month,” “Between 1 and 6 months ago,” or “Between 6 months and 1 year ago”). If the respondent selects “Over 1 year ago” or “Never,” a score of 0 points is assigned.
- Index Range: Total scores range from 0 to 9, representing the cumulative variety of distinct delinquent and disciplinary infractions committed within the preceding 12-month window.
- Interpretation Benchmarks:
- 0 points: Absence of past-year institutional noncompliance or delinquent offenses.
- 1–2 points: Low-level deviance; typically restricted to isolated school infractions (skipping class, counselor referral) without legal system contact.
- 3–5 points: Moderate delinquency; characterized by combined school disciplinary actions and property infractions.
- 6–9 points: Severe, multi-systemic delinquency; characterized by institutional exclusion, high-value theft, property destruction, and law enforcement or judicial involvement.
Permissions, Fee, and Test Year
The Disciplinary and Delinquent Behavior—SAGE Baseline Survey was formulated in 1993 as part of the Student Activity and Gang Education evaluation project by Flewelling, Paschall, and Ringwalt at the Research Triangle Institute. The instrument was subsequently published by the Centers for Disease Control and Prevention within Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (Dahlberg et al., 2005, pp. 210–211).
Because the development and publication of this tool were supported through federal grants and public health funding, the instrument resides in the public domain. It may be used, reproduced, and adapted for non-commercial educational, clinical, and scientific research purposes free of charge without formal copyright royalty fees. Researchers using the scale are requested to cite the original RTI report and the CDC compendium accordingly.
References
Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30(1), 47–88. https://doi.org/10.1111/j.1745-9125.1992.tb01093.x
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., pp. 210–211). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/YV_Compendium.pdf
Flewelling, R. L., Paschall, M. J., & Ringwalt, C. L. (1993). SAGE Baseline Survey. Research Triangle Institute.
Hirschi, T. (1969). Causes of delinquency. University of California Press.
Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth. Academic Press.
Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100(4), 674–701. https://doi.org/10.1037/0033-295X.100.4.674
Paschall, M. J., Flewelling, R. L., & Ennett, S. T. (1998). Racial differences in violent behavior among young males: The role of family structure, SES, and neighborhood disorder. Journal of Health and Social Behavior, 39(3), 218–231. https://doi.org/10.2307/2676313