Exposure to Violence, Gender Norms, and Youth Victimization In Post-Conflict Colombia: A Mediation Analysis

Background and Purpose

The 2015 Peace Agreement in Colombia raised expectations that a decrease of violence would follow the end of armed conflict. However, GBV has continued to exist in both household and community realms, with hierarchical gender roles continuing to be deeply rooted. Based on the continuum of violence framework and social norms theory, this study investigates whether gender violence-supportive norms –(GVSN) mediate the association between exposure to violence and risk for intimate partner violence (IPV) and sexual violence among Colombian youth.

Methods

Data were obtained from the Violence Against Children and Youth Survey (VACS), performed during 2018 by the Ministry of Health, Center for Disease Control and Prevention (CDC) and international partners (N=2631). Three investigator-generated indices were based on VACS items: (1) exposure to household violence (defined as witnessing physical violence against the mother, stepmother, or sibling), (2) external exposure of violence (operationalized as witnessed physical attacks in the community or conflict-related) and (3) GVSN (a standardized continuous index measuring attitudes that justified or tolerated GBV ; (0.70). Binary outcomes were constructed IPV (romantic partner or husband/committed physical acts of violence) and sexual violence (unwanted sexual touching, attempted rape, sexual assault by partner or another person). A Baron and Kenny mediation framework guided this analysis; estimates of total and direct effects on binary outcomes were obtained using logistic regression models (reported as odds ratios with 95% CIs). OLS regression model treated exposure as a predictor of norms, and the Sobel test indicated that the indirect effect was statistically significant. All models were adjusted for sex, age, and school attendance (p <. 05, two-tailed).

Results

Approximately one in four respondents reported experiencing IPV, and one in five reported experiencing sexual violence, with higher prevalence among female respondents affected. Both household exposure (OR = 1.846, p <.001) and external exposure (OR = 1.699, p<.001) were significant and positively associated with increased odds, of IPV. Comparable associations were observed for sexual violence (household (OR =1.1730, p<.001 external: OR = 1.685, p<.001). Both exposure indices also significantly predicted higher levels of GVSN (household: b=0.180, p<.001; external b=0.140, p<.001 = .065). When GVSN were included as a mediator, their effect was significant for both IPV (OR=1.252, p<.001), and sexual violence (OR=1.150, p=.003), while exposure effects were mitigated but remained statistically significant. Sobel test confirmed that the indirect effect of GVSN were statically significant across all four pathways; household exposure to IPV ( z=3.48, p<.001), external exposure to IPV (z=2.93, p<.01), household exposure to sexual violence (z=2.96, p<.01), and external exposure to sexual violence (z=2.38, p<.05), indicating partial mediation.

Conclusion and Implications

These findings highlight that ending armed conflict is insufficient to eliminate GBV. Violence exposure witnessed within the household remained a significant and independent predictor of IPV and sexual violence even after accounting for gender normative mechanisms. Transitional justice policies often fail to address the domestic sphere where violence is equally and pervasive. When policy responses prioritize community-level, or legal reforms while overlooking what happens inside the home, they leave a critical gap.