Our results show that neither the household nor the school ambient conditions were strongly correlated to the exposure during trips of either active transportation mode, which indicates our approach is able to provide additional exposure information. When we examined if the combination of both of these locations’ ambient concentrations could account for the variation in exposure, all the models performed poorly. The multiple linear regression model with the highest explanatory power could only account for six percent of the variability in exposure. The mode of active trans- portation can significantly affect the resulting exposure during a trip, and we find that cycling results in a significantly lower dose than walking. Our results confirm the need to produce activity based exposure models. Without incorporating the activity patterns of the students, we would have only been able to estimate student exposure at their household or school location, which may have excluded peak exposures that occur near traffic.