Deadly Gun Violence Exposure and Harsh Parenting: Evidence from a Propensity Score Analysis

Background and Purpose:
While harsh parenting has traditionally been attributed to individual or family-level factors, emerging evidence highlights the role of environmental conditions on caregiver stress and behavioral responses. Exposure to community violence may heighten perceived threats to child safety, prompting increased parental control and assertive discipline. This study examines the causal effect of exposure to nearby deadly gun violence on harsh parenting behaviors. We hypothesize that exposure to nearby gun violence is associated with increased harsh parenting, independent of individual and household characteristics.

Methods:
Data were obtained from the Future of Families and Child Wellbeing Study (FFCWS), a longitudinal birth cohort of 4,898 children followed from birth through age 22. Exposure to deadly gun violence at Year 15 was measured using restricted-use Gun Violence Archive data linked to geocoded caregiver and youth locations, capturing proximity and timing of incidents. The treatment was operationalized as a caregiver self-reported binary indicator of exposure to at least one deadly gun violence event in the past year within one mile of the home. The primary outcome was harsh parenting, measured as a summed indicator of caregiver-reported disciplinary behaviors, including 2 items (psychological aggression and physical assault), at Year 15, with a similar youth-reported measure used as a sensitivity analysis.

Given the non-random assignment of community violence exposure, propensity score analysis was employed within the Rubin Causal Model framework. Propensity scores were estimated using logistic regression and incorporated pre-exposure covariates, including baseline harsh parenting behaviors, socioeconomic indicators (household income, poverty, education), caregiver mental health (depression and anxiety), and demographic characteristics (caregiver and child’s age, race/ethnicity, and sex), and household composition. Nearest neighbor matching with three matches per treated unit was used to create a balanced analytic sample and to estimate the average treatment effect on the treated (ATET). Covariate balance was assessed using standardized mean differences, and overlap between treatment groups was evaluated to ensure common support.

Results:
Of 3,220 participants, 2,657 had complete data for propensity score estimation; 2,654 were retained in the final matched sample (1,397 exposed; 1,257 controls) with 566 excluded due to missing data or lack of common support. Propensity score matching substantially improved covariate balance, with standardized mean differences below 0.1 for nearly all variables. Exposure to nearby deadly gun violence was associated with increased harsh parenting among caregivers (ATET = 0.12, 95% CI: 0.05–0.20, p = 0.002). Results were consistent in sensitivity analyses using youth-reported parenting behaviors (ATET = 0.10, 95% CI: 0.02–0.18, p = 0.016).

Conclusions and Implications:
Findings provide support that exposure to community violence contributes to increases in harsh parenting behaviors, supporting ecosyndemics frameworks that emphasize the interaction of environmental and interpersonal stressors. By applying a causal inference approach to longitudinal data, this study highlights community violence exposure as an important contextual factor shaping family functioning. Results have implications for violence prevention efforts, trauma-informed interventions, and policies aimed at reducing exposure to neighborhood violence.