Children’s Exposure to Community Violence

Abstract

The Children’s Exposure to Community Violence (CECV) scale, adapted from the foundational structured interview protocol Things I Have Seen and Heard developed by John E. Richters and Pedro E. Martinez (1990) at the National Institute of Mental Health (NIMH), is a standardized self-report psychometric instrument designed to quantify the frequency and severity of violent events witnessed or directly experienced by school-aged children and adolescents. Comprising 12 operationalized items, the instrument captures a broad spectrum of ambient neighborhood threats, ranging from sensory awareness of firearm discharge and overt illicit drug transactions to home invasions, direct physical assaults, weapon brandishing, and domestic or neighborhood homicides. Items are rated on a 4-point Likert-type frequency metric ranging from 1 (Never) to 4 (Many times), yielding a continuous composite mean exposure index ranging from 1.00 to 4.00. Psychometric evaluations across diverse pediatric, clinical, and urban community samples demonstrate robust internal consistency (Cronbach’s alpha typically ranging from .78 to .88), coherent factor structures reflecting direct victimization and ambient neighborhood violence witnessing, and pronounced convergent validity with measures of post-traumatic stress disorder (PTSD), internalizing distress, depressive symptomatology, externalizing conduct problems, and neurocognitive dysregulation. Documented prominently in the Centers for Disease Control and Prevention (CDC) compendium of youth violence assessment tools, the CECV remains an indispensable instrument in developmental psychopathology, urban public health surveillance, school-based mental health triage, and pediatric trauma research.

Keywords

Children’s Exposure to Community Violence, CECV, Things I Have Seen and Heard, John E. Richters, Pedro E. Martinez, pediatric trauma, neighborhood violence, witnessing violence, direct victimization, post-traumatic stress disorder, developmental psychopathology, psychometrics, trauma screening

Authors

The instrument was formulated by John E. Richters, Ph.D., and Pedro E. Martinez, Ph.D., during their tenure at the Child and Adolescent Disorders Research Branch of the National Institute of Mental Health (NIMH), an agency of the United States Department of Health and Human Services located in Rockville, Maryland. Dr. Richters, a distinguished developmental psychopathologist and research methodologist, focused extensively on the developmental epidemiology of childhood psychiatric morbidity, the etiology of conduct problems, and the cumulative psychological toll of chronic ambient adversity. Dr. Martinez collaborated closely on epidemiological field studies evaluating inner-city youth facing pervasive socio-environmental stressors. Their pioneering collaborative project in the early 1990s culminated in the structured assessment system Things I Have Seen and Heard: A Structured Interview for Assessing Young Children’s Violence Exposure, which was later adapted into standardized brief questionnaires, including the 12-item CECV format documented in federal assessment compendia.

Purpose

The primary clinical, epidemiological, and psychometric objective of the Children’s Exposure to Community Violence (CECV) scale is to provide a reliable, developmentally sensitive, and standardized metric for measuring the frequency of traumatic and violent occurrences within a child’s immediate socio-ecological environment. Originally designed during a period characterized by escalating urban firearm violence and drug-epidemic-related community destabilization, the CECV was engineered to bridge a critical diagnostic and measurement void: the empirical documentation of ambient trauma in pediatric populations who were frequently overlooked unless they presented with catastrophic physical injuries.

From an applied perspective, the instrument serves multiple interrelated purposes across clinical, educational, and research infrastructures:

• Epidemiological Surveillance and Triage: The CECV provides public health authorities and community mental health programs with standardized epidemiological prevalence estimates of violent events (e.g., witnessing beatings, stabbings, firearm discharge, drug sales, and home invasions) across specific geographic and socio-demographic sectors.
• Clinical Trauma Screening: In pediatric primary care, community mental health clinics, and inpatient psychiatric facilities, the scale functions as an initial screener to detect latent traumatic stress exposures that might otherwise be mischaracterized as attention-deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), or unprovoked conduct problems.
• Disentangling Direct vs. Vicarious Victimization: The tool allows clinicians and researchers to parse differences between ambient sensory exposure (e.g., hearing gunshots, seeing drug deals), vicarious or witnessed violence (e.g., seeing someone stabbed or shot), and immediate household or personal threats (e.g., home break-ins, seeing a family member shot or stabbed).
• Empirical Research in Developmental Psychopathology: By quantifying the ecological dose of environmental violence, the CECV facilitates testing moderation, mediation, and cumulative risk hypotheses regarding allostatic load, hypothalamic-pituitary-adrenal (HPA) axis dysregulation, emotional numbing, cognitive appraisal distortions, and developmental cascade models of academic and psychological decline.

Psychological Construct

The psychological construct evaluated by the CECV is Childhood Community Violence Exposure (CVE), conceptualized as a multidimensional, ecologically situated stressor encompassing both indirect (witnessed or ambient) and direct (proximal or personal) encounters with intentional acts of interpersonal aggression, crime, and weapons-related events within the domestic and neighborhood ecologies. Rather than treating violence exposure as a uniform, binary phenomenon (present vs. absent), modern developmental psychopathology conceptualizes it as a complex gradient of chronic ecological toxicity characterized by varying levels of proximity, personal threat, sensory immediacy, and relational salience.

The construct measured by the 12 CECV items can be disaggregated into distinct conceptual facets:

• Ambient Neighborhood Disorder and Sensory Clues: Represented by items such as “I have heard guns being shot” (Item 1), “I have seen drug deals” (Item 3), and “I have seen gangs in my neighborhood” (Item 10). These items measure the chronic, low-to-moderate-level ambient backdrop of community dysfunction. Chronic exposure to these cues signals continuous environmental unpredictability and danger, prompting persistent physiological hyperarousal, hypervigilance, and cognitive anticipation of imminent catastrophe.
• Witnessing Severe and Lethal Interpersonal Violence: Represented by items such as “I have seen someone being beaten up” (Item 4), “I have seen somebody get stabbed” (Item 6), “I have seen somebody get shot” (Item 7), and “I have seen somebody pull a gun on another person” (Item 11). These items evaluate direct visual witnessing of life-threatening physical aggression. Visual confrontation with real-world bodily mutilation or homicide violates the core developmental expectation of physical safety, generating intense acute horror, acute intrusive memories, and psychological shattered assumptions.
• Sanctuary Invalidation and Household Compromise: Represented by items such as “My house has been broken into” (Item 5), “I have seen a gun in my home” (Item 8), “I have seen alcohol such as beer, wine, or hard liquor in my home” (Item 9), and “I have seen someone in my home get shot or stabbed” (Item 12). The psychological sanctuary hypothesis posits that the developmental impact of community violence is mitigated when the home functions as an emotionally and physically secure haven. When violence penetrates the household perimeter (e.g., through burglary or household firearms) or when domestic violence culminates in severe injury (Item 12), the protective buffer is dismantled, exacerbating the risk of severe post-traumatic syndromes, dissociation, and attachment insecurity.
• Institutional and Legal Encounters: Operationalized by “I have seen somebody arrested” (Item 2). Witnessing law enforcement apprehension of community members or relatives introduces complex socio-emotional constructs involving perceived institutional injustice, fear of authority, the sudden disruption of familial networks, and the visual reality of coercive state force.

Theoretical Framework

The conceptual architecture of the CECV is grounded in four major theoretical traditions within developmental psychopathology, ecological systems theory, and cognitive-behavioral trauma paradigms:

1. Bronfenbrenner’s Social-Ecological Systems Model

Urie Bronfenbrenner’s ecological systems theory conceptualizes child development as nested within concentric environmental systems: the microsystem (family, home, classroom), the mesosystem (interactions between home and school), the exosystem (neighborhood safety, community infrastructure, policing), and the macrosystem (cultural beliefs, systemic socioeconomic disparities). The CECV explicitly evaluates pathogenic events occurring across the boundary between the microsystem and the exosystem. Richters and Martinez recognized that when the neighborhood exosystem becomes violent, toxic influences breach microsystemic boundaries, undermining parental protective capacity and disrupting normal developmental transitions.

2. Developmental Psychopathology and the Sanctuary Hypothesis

Grounded in the work of Dante Cicchetti and Norman Garmezy, developmental psychopathology conceptualizes psychological disorder as an emergent property of dynamic interactions between ecological vulnerabilities, ontogenetic limitations, and environmental assaults. Garbarino’s conceptualization of “environmental toxicity” directly influenced Richters and Martinez’s model. Under this framework, children possess biological and emotional homeostatic buffers. However, chronic community violence acts as an ongoing environmental pathogen that depletes coping mechanisms. Garbarino’s Sanctuary Hypothesis posits that children can withstand considerable external adversity if their domestic microsystem remains intact and protective. The inclusion of household-specific items (e.g., Items 5, 8, 9, 12) reflects the critical theoretical distinction between external neighborhood hazards and the collapse of the primary caregiving sanctuary.

3. Allostatic Load and Neurodevelopmental Stress Paradigms

From a neurobiological standpoint, the CECV’s measurement of cumulative frequency aligns with Bruce McEwen’s theory of allostatic load. Repeated sensory and visual exposure to life-threatening danger triggers recurrent surges of catecholamines and glucocorticoids via the hypothalamic-pituitary-adrenal (HPA) axis and the autonomic nervous system. Over time, persistent hyperactivation precipitates structural and functional alterations in the amygdala, hippocampus, and prefrontal cortex. This neurobiological adaptation promotes chronic threat-detection biases, emotional dysregulation, and impaired executive functioning, translating ecologically into academic underachievement and interpersonal hostility.

4. Cognitive Processing and Social Learning Theories

According to Ronnie Janoff-Bulman’s Shattered Assumptions Theory, traumatic exposure fundamentally challenges fundamental worldviews: the belief in personal invulnerability, the belief that the world is meaningful and just, and positive self-worth. Witnessing stabbings, shootings, and weapon brandishing forces rapid cognitive reorganization, often resulting in fatalistic outlooks, hopelessness, and foreshortened future expectations. Concurrently, under Albert Bandura’s Social Learning Theory, observed community violence models aggressive behavioral scripts as standard or efficacious methods for conflict resolution, thereby accelerating the transgenerational transmission of aggressive behavioral repertoires.

Validity

The Children’s Exposure to Community Violence scale and its parent interview protocol (Things I Have Seen and Heard) have undergone extensive psychometric validation across three decades of developmental and clinical research. Psychometric evaluations provide compelling empirical evidence for construct, convergent, discriminant, and predictive validity.

Construct Validity

Construct validity is substantiated by significant linear trends observed between CECV exposure scores and known sociodemographic indices of community risk. In initial validation studies conducted by Richters and Martinez (1990, 1993) among elementary school children in high-crime urban areas, CECV mean scores reliably differentiated between distinct neighborhood blocks categorized independently by municipal police department crime statistics. Furthermore, cross-informant construct validation studies (comparing child self-reports on the CECV with maternal reports using the Parent Report of Children’s Exposure to Community Violence) demonstrate convergent construct validity, while highlighting the well-documented clinical phenomenon of parental under-reporting: children routinely report significantly higher exposure frequencies to ambient and outdoor violence than their primary caregivers are aware of, confirming the construct validity of child self-report methodologies.

Convergent Validity

Convergent validity is established through significant positive correlations with well-validated psychological and psychiatric symptom batteries:

• Post-Traumatic Stress Symptoms: CECV composite scores consistently exhibit robust positive correlations ($r = .35$ to $.55, p < .001$) with standardized measures of post-traumatic stress, such as the Child Post-Traumatic Stress Disorder Reaction Index (CPTSD-RI) and the UCLA PTSD Reaction Index. Specific subscale analyses confirm that items measuring direct witnessing of weapon violence and personal household victimization account for the largest proportion of variance in intrusive re-experiencing, persistent avoidance, and physiological hyperarousal.
• Internalizing Distress: Significant positive associations are consistently documented between CECV scores and child-reported anxiety and depression indices, including the Children’s Depression Inventory (CDI; $r = .28$ to $.42$) and the Revised Children’s Manifest Anxiety Scale (RCMAS; $r = .30$ to $.45$).
• Externalizing and Aggressive Behaviors: Substantial correlations have been demonstrated with externalizing indices on the Child Behavior Checklist (CBCL) and the Teacher Report Form (TRF), with correlation coefficients typically ranging between $.25$ and $.40$. Exposure to physical fights and weapon use correlates with peer aggression, proactive hostility, and disruptive classroom conduct.

Discriminant Validity

Discriminant validity is supported by analyses demonstrating that the CECV does not merely capture general negative emotionality, response acquiescence, or somatic complaints unrelated to trauma. When controlling for general socioeconomic status, household income, and generalized trait negative affectivity, CECV scores retain their unique predictive power regarding trauma-specific intrusions, avoidance behaviors, and hypervigilance. Moreover, factor-analytic studies show clear empirical separation between CECV items and scales assessing general family conflict or non-violent domestic chaos, confirming that the tool specifically measures violent, threatening events rather than generalized family dysfunction.

Predictive and Longitudinal Validity

Longitudinal investigations employing the CECV confirm its capacity to predict downstream psychological and behavioral trajectories over extended developmental intervals. Elevated baseline CECV scores predict prospective increases in adolescent substance misuse, academic attrition, delinquency, and elevated biological markers of chronic stress (such as blunted morning cortisol awakening responses and elevated systemic inflammatory markers like C-reactive protein) evaluated years post-baseline assessment.

Reliability

The psychometric reliability of the CECV has been thoroughly vetted across diverse community, school-based, and clinical samples of children and adolescents (typically spanning ages 6 to 18 years, with verbal administration protocols utilized for children below age 9).

Internal Consistency

Across empirical studies, the 12-item CECV demonstrates acceptable to high internal consistency:

• Overall Composite Scale: Cronbach’s alpha ($lpha$) coefficients for the total 12-item scale typically range from $.78$ to $.88$ in urban school-aged and adolescent cohorts. In high-exposure community samples, alpha coefficients frequently reach or exceed $.85$, indicating high interrelatedness among items while retaining sufficient breadth across distinct trauma typologies.
• Subscale Consistency: When partitioned into empirical subscales (e.g., Ambient/Witnessed Violence vs. Direct/Household Victimization), internal consistency values reflect the distinct nature of low-frequency, severe events. Subscales capturing frequent ambient exposures (hearing gunshots, seeing drug deals, seeing gangs) exhibit alpha coefficients between $.75$ and $.84$. Conversely, subscales comprising rare, high-severity events (seeing someone stabbed or shot, home invasion) occasionally show lower alpha coefficients ($lpha = .62$ to $.72$), which psychometricians recognize as an expected mathematical consequence of severe positive skewness and low base-rate endorsement rather than scale invalidity.

Test-Retest Stability

Test-retest reliability has been investigated across short-to-moderate temporal intervals. Over a 2- to 4-week retest window, Pearson correlation coefficients ($r$) and Intraclass Correlation Coefficients (ICC) generally range between $.72$ and $.84$, indicating substantial temporal stability of child recall. However, over extended intervals exceeding 6 to 12 months, test-retest coefficients naturally moderate ($r pprox .50$ to $.65$), reflecting both genuine fluctuations in community violence rates across time and developmental changes in children’s retrospective recall, emotional processing, and coping strategies.

Factor Analysis

Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) across various validation cohorts have examined the underlying latent dimensionality of the 12-item CECV. Although the instrument is frequently scored as a single unidimensional composite index representing total violence exposure, structural equation modeling and factor-analytic studies indicate that a multidimensional latent structure best fits the observed data.

Exploratory Factor Analytic Solutions

Early principal axis factoring and principal component analyses (utilizing Promax or Varimax rotations) consistently yielded a two- or three-factor solution, accounting for approximately 48% to 62% of the total shared variance:

• Factor 1: Witnessing Severe Community Violence (Ambient Interpersonal Aggression): This factor accounts for the largest proportion of common variance (often 30%–38%). Salient item loadings ($lambda > .50$) include Item 1 (Hearing guns being shot), Item 4 (Seeing someone beaten up), Item 6 (Seeing somebody get stabbed), Item 7 (Seeing somebody get shot), and Item 11 (Seeing somebody pull a gun on another person).
• Factor 2: Ambient Neighborhood Disorder and Illicit Activity: This factor captures community environmental instability. High-loading items ($lambda > .55$) comprise Item 2 (Seeing somebody arrested), Item 3 (Seeing drug deals), and Item 10 (Seeing gangs in the neighborhood).
• Factor 3: Household Compromise and Direct/Proximal Victimization: This factor isolates events occurring within the primary domestic microsystem. Primary loadings encompass Item 5 (House broken into), Item 8 (Gun in home), Item 9 (Alcohol/liquor in home), and Item 12 (Someone in home shot or stabbed). Item 12 cross-loads moderately onto Severe Violence witnessing, but uniquely characterizes catastrophic family-level disruption.

Confirmatory Factor Analysis (CFA) Fit Indices

In modern psychometric structural evaluations comparing a single-factor unidimensional model against a correlated two-factor (Witnessed Community Violence vs. Household/Direct Victimization) and a hierarchical bi-factor model, the correlated two-factor and bi-factor models demonstrate superior goodness-of-fit across adolescent populations:

• Root Mean Square Error of Approximation (RMSEA): Values for the two-factor model typically range from $.042$ to $.058$ ($90%\text{ CI } [.031, .068]$), demonstrating excellent approximation in the population, whereas a strictly unidimensional model often yields marginal fit (RMSEA $> .085$).
• Comparative Fit Index (CFI): Values routinely exceed $.94$, frequently reaching $.97$ in bi-factor specifications.
• Tucker-Lewis Index (TLI): Typically ranges from $.93$ to $.96$.
• Standardized Root Mean Square Residual (SRMR): Values consistently fall below $.05$, indicating minimal residual variance between the observed covariance matrix and the implied model.

These findings substantiate that while a single summary score is clinically convenient and reflects cumulative stress exposure, research evaluating differential etiologies of PTSD versus conduct problems benefits from examining separate subscale scores representing direct household victimization versus ambient community witnessing.

Instrument / Measurement Tool

The operational specifications, administration protocols, and scoring algorithms for the 12-item Children’s Exposure to Community Violence instrument are summarized below:

• Instrument Name: Children’s Exposure to Community Violence (CECV)
• Original Protocol: Adapted from Things I Have Seen and Heard: A Structured Interview for Assessing Young Children’s Violence Exposure (Richters & Martinez, 1990)
• Primary Reference Document: Published in Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (Dahlberg, Toal, Swahn, & Behrens, 2005, pp. 321–332; Centers for Disease Control and Prevention)
• Target Population: Children and adolescents aged approximately 6 to 18 years. (For children aged 6 to 9 years, structured one-on-one verbal interview administration with visual response cards is recommended; children aged 10 and older may complete the questionnaire via self-report).
• Administration Time: Approximately 5 to 10 minutes for self-administration; 10 to 15 minutes for clinician-led interview administration.
• Item Count: 12 standardized items.
• Response Scale: 4-point ordinal frequency metric:
  • Never = 1
  • Once or twice = 2
  • A few times = 3
  • Many times = 4
• Scoring Methodology:
  • Total Cumulative Score: Point values assigned to each of the 12 items are summed together (theoretical raw sum range: 12 to 48) and then divided by the total number of items completed (12), yielding an overall mean exposure index.
  • Intended Score Range: 1.00 to 4.00.
  • Interpretation: Higher numerical scores denote greater frequency and severity of exposure to ambient, witnessed, and domestic acts of crime and violence. Scores approaching 1.00 signify absent or negligible exposure, while scores exceeding 2.50 to 3.00 denote pervasive, chronic exposure across multiple trauma modalities.
• Clinical Consideration: Endorsement of critical sentinel items (e.g., Item 6: stabbing, Item 7: shooting, Item 11: gun pulled, Item 12: family member shot/stabbed) at any frequency other than Never warrants immediate follow-up trauma screening and clinical safety planning.

Permissions & Fee and Test Year

The initial structured assessment system, Things I Have Seen and Heard, was developed in 1990 by John E. Richters and Pedro E. Martinez at the National Institute of Mental Health (NIMH), a federal research institute within the United States Government. Subsequently, the 12-item adaptation was incorporated into the Centers for Disease Control and Prevention (CDC) public domain compendium titled Measuring Violence-Related Attitudes, Behaviors, and Influences Among Youths: A Compendium of Assessment Tools (published in 2005).

Because this work was funded and produced by federal agencies of the United States Government, the instrument resides in the public domain. No licensing fees, commercial purchase charges, or formal copyright royalty payments are required for its administration in clinical practice, public health programs, or academic research. Investigators and clinicians are permitted to utilize, reproduce, and adapt the scale provided appropriate academic attribution and bibliographic citation are accorded to Richters and Martinez (1990) and the CDC compendium editors. Researchers planning digital adaptations or translations into other languages are strongly advised to adhere strictly to standardized cross-cultural translation back-translation protocols to preserve linguistic and psychometric equivalence.

References

• 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. 321–332). Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/YV_Compendium.pdf
• Margolin, G., & Gordis, E. B. (2000). The effects of family and community violence on children. Annual Review of Psychology, 51(1), 445–479. https://doi.org/10.1146/annurev.psych.51.1.445
• Martinez, P. E., & Richters, J. E. (1993). The NIMH Community Violence Project: II. Children’s distress symptoms associated with violence exposure. Psychiatry, 56(1), 22–35. https://doi.org/10.1080/00332747.1993.11024618
• Richters, J. E., & Martinez, P. E. (1990). Things I have seen and heard: A structured interview for assessing young children’s violence exposure. National Institute of Mental Health.
• Richters, J. E., & Martinez, P. E. (1993). The NIMH Community Violence Project: I. Children as victims of and witnesses to violence. Psychiatry, 56(1), 7–21. https://doi.org/10.1080/00332747.1993.11024617
• Richters, J. E., & Saltzman, W. (1990). Survey of Children’s Exposure to Community Violence: Self-report version. National Institute of Mental Health.
• Scarpa, A. (2001). Community violence exposure in a young adult sample: Lifetime prevalence and socioemotional functioning. Journal of Interpersonal Violence, 16(1), 36–53. https://doi.org/10.1177/088626001016001003

13. Items of the Scale (Questionnaire)