| description abstract | An increasing number of smartphone apps with driving safety diagnosis functions have been developed in the market; however, empirical evidence on whether these apps are actually effective in promoting road safety or not is limited. This study examines how the use of a global positioning system (GPS)-enabled smartphone app with driving safety related diagnosis affects changes in driving behaviors on expressways. The app was developed by the authors as a simplified diagnosis tool using only GPS information. The app can measure three types of driving safety related indicators second by second, i.e., speed limit compliance, acceleration/deceleration, and driving stability, in addition to a set of subjective driving safety related indicators. This study used 187,549 epochs (2 s/epoch) as the analysis unit, collected from 15 drivers who made 21 trips during a three-month driving experiment on expressways in the western part of Japan in 214. To account for the influence of drivers’ unobserved heterogeneity in driving speed control behavior, this study built a multilevel ordered probit model. Model estimation results confirmed that, during a three-month period, using the app contributes to less dangerous driving as measured by compliance level with the speed limit; however, its effects on acceleration/deceleration and driving stability were limited. The effects of some additional functions of the app on improvement in driving safety related performance were mixed. Finally, it was revealed that drivers’ heterogeneous driving propensity is significantly influential to actual driving safety related performance. | |
