Considerations of whether a travel alternative (e.g., mode, destination) affords a traveler an acceptable level of safety from traffic collisions and personal security from crime may be especially important for active modes of walking and bicycling. Although theory suggests that safety and security influence nonmotorized travel decision making, empirical evidence to support these hypotheses is inconclusive. In addition to measurement limitations and data constraints, safety and security concerns are often inadequately distinguished. This study advances knowledge in this area by analyzing objective safety and security measures within a mode–destination choice context. Discretionary trips from a Portland, Oregon, household travel survey were analyzed with a joint mode–destination choice multinomial logit model. Some measures of traffic safety—more comfortable facilities, sidewalks, traffic signals, and traffic calming installations—were positively associated with walking and bicycling. Measures of a lack of personal security (higher levels of crime) were negatively associated with walking, but this effect was countered by a positive interaction of origin–destination crime on walking. Age and gender rarely moderated these safety and security relationships. Although results provide tentative support of hypotheses on the relationship between safety and security and travel mode choice, more work is needed in this area. Measuring safety and security perceptions and applying more flexible structures for discrete choice modeling might be the most promising avenues for future research on the effects of safety and security in discretionary decision making of active travel. This analysis informs interventions to promote active travel and provides one way to include safety and security concerns in travel demand models.