Increasing evidence suggests that neighborhood-based measures of socioeconomic status are correlated with traffic injury. The main objective of this study is to determine the differences in associations between predictive variables and injury crashes (i.e. including injury and fatal crashes). To this end, crash data, socio-demographic, socioeconomic characteristics and road network variables are collected at the neighborhood-level and categorized by different genders and transport mode; “car driver”, “car passenger” and “vulnerable road users” (i.e. pedestrians and cyclists). In this study an activity-based transportation model called FEATHERS (Forecasting Evolutionary Activity-Travel of Households and their Environmental RepercussionS) is utilized to produce exposure measures. Exposure measures are in the form of production/attraction trips for several traffic analysis zones (TAZ) in Flanders, Belgium. Analyzing crashes at a neighborhood- level provides important information that enables us to compare traffic safety of different neighborhoods. This information is used to identify safety problems in specific zones and consequently, implementing safety interventions to improve the traffic safety condition. This can be carried out by associating casualty counts with a number of factors (i.e. developing crash prediction models) which have macro-level characteristics, such as socio-demographic and network level exposure. The results indicate that socioeconomic variables are differently associated with casualties of different travel modes and genders. For instance, income level of residence of a TAZ is a significant predictor of male car driver injury crashes while it doesn’t significantly contribute to the prediction of female car driver injury crashes.