This study aims to examine child pedestrian safety around parks by considering four rule-compliance measures: temporal, spatial, velocity and visual search compliance. In this regard, street crossing observations of 731 children were recorded at 17 crosswalks around four parks in Montreal, Canada. Information on child behaviors, road features, and pedestrian–vehicle interactions were gathered in three separate forms. Chi-square tests were used to highlight the individual, situational, behavioral and road environmental characteristics that are associated with pedestrian rule compliance. About half of our sampled children started crossing at the same time as the adults who accompanied them, but more rule violations were observed when the adult initiated the crossing. The child’s gender did not have a significant impact on rule compliance. Several variables were positively associated with rule compliance: stopping at the curb before crossing, close parental supervision, and pedestrian countdown signals. Pedestrian–car interaction had a mixed impact on rule compliance. Overall, rule compliance among children was high for each of our indicators, but about two-thirds failed to comply with all four indicators. A few measures, such as longer crossing signals and pedestrian countdown displays at traffic lights, may help to increase rule compliance and, ultimately, provide safer access to parks.

Much attention in the scientific literature has been paid lately to the road around schools when studying child pedestrian injury risk. However, little research and much less effort to prevent injuries has been done regarding other frequent children’s destinations such as parks. Road insecurity might be a good reason to avoid them or drive instead of walk. The goals of this study are 1) to assess the occurrence of interactions between child pedestrians and vehicles based on observations of individual and crossing characteristics and 2) to compare the prevalence and the characteristics of the interactions near schools and parks.Observations of children’s behaviors while crossing around schools (n=869) and parks (n=731) were recorded at 11 schools and 4 parks in Montréal, Canada. Environmental characteristics were recorded for 14 school and 17 park crosswalks with a minimum of 40 observations for each crosswalk. For recorded interactions, information was also collected to characterize the behaviors of involved parties. First, chi-square tests reveal the differences of interaction characteristics at school and park crossings. A mixed-effect logit regression model was then performed to identify factors associated with interactions while ensuring that the fixed effects of the crosswalks characteristics are treated as such.Much fewer interactions were recorded around schools (7.13%) than parks (18.33%) and these interactions were posing less risk to the children safety (more careful behaviors from drivers/cyclists). Our results point out the need to adapt the crossing environment to children’s capacity since most of them already respect the rules. Providing safety near parks, combined with safe routes to other destinations, can promote active transportation and livable neighborhoods for all.