Is green space associated with reduced crime? A national-scale study from the Global South.

Assumptions about the link between green space and crime mitigation are informed by literature that overwhelmingly originates in the Global North. Little is known about the association between green spaces and crime in the Global South. We utilized 10 years of precinct-level crime statistics (n = 1152) over South Africa, a global crime hotspot, to test the hypothesis that green space is associated with reduced crime rates. We found that, after controlling for a number of socio-demographic confounders (unemployment, income, age, education, land use and population density), for every 1% increase in total green space there is a 1.2% (0.7 to 1.7%; 95% confidence interval) decrease in violent crime, and 1.3% (0.8 to 1.8%) decrease in property crime, with no effect on sexual crimes. However, the direction of the association changed for property crimes when exploring the effect of green space characteristics including tree cover and park accessibility. Property crimes increase by 0.4% (0.1 to 0.7%) with a percentage increase in tree cover, and by 0.9% (0.5 to 1.3%) with every kilometer increase in proximity to a public park. Further research, including experimental studies, is needed to better isolate causal mechanisms behind crime-green space associations, especially considering that green space may map to race and income inequality and that there may be more crime reporting in affluent areas. Nevertheless, our results provide a complementary contribution to the evidence from the Global North, highlighting the need for more nuanced definitions of green space and its characteristics when considering links to crime. When viewed in light of the broader suite of ecosystem services provided by green space, our results support urban greening as a major strategy towards achieving just and sustainable cities and towns.

Development and Validation of the Youth Violence Potential Scale.

The assessment of violence risk among youth remains a priority for researchers and practitioners globally. The absence of scales designed or validated in developing countries drives the need for a psychometrically sound, alternative measure. The purpose of this article is to validate the Youth Violence Potential Scale (YVPS). The YVPS was administered twice over 12 months to 318 South African males, aged 12-24 years. Exploratory and confirmatory factor analysis of separate samples confirmed a 19-item, three-factor solution, comprised of a Deviant Peers subscale, a Pro-Gangs Attitude subscale, and a Pro-Violence Attitude subscale. The YVPS exhibited internal reliability (α = .91) and discriminant and convergent validity among its subscales. Correlation with self-reported problem behavior/offending (r = .48, p < .001) and risk assessment from the maternal caregiver (r = .39, p < .001) evidenced concurrent and external validity.