论文部分内容阅读
Experiments using smartphones to influence behavior have been growing rapidly in many fields, especially in health and fitness research, and studies on eco-feedback technologies. In these studies, users are first tracked to understand their baseline behaviors, then measured continuously while they receive feedback about their actions. In transportation, studies using smartphones to change behavior have been limited due to the difficulty in even tracking users in the first place. Collecting data from smartphones in a battery efficient manner is a large research problem, and behavior change studies depend on being able to track travel behaviors. We developed an automated travel diary system which eciently and unobtrusively collected travel data using smartphones and ran an experiment to evaluate how peoples awareness of their transportation behavior, attitudes towards sustainable transportation, intentions to change behavior, and measured travel behavior changed. For three weeks, 135 participants used an application on their iPhone or Android smartphone which unobstrusively tracked their location and sent data to a server which processed their data into trips and attributes related to their trips, such as time spent traveling, amount of money spent for transportation, amount of CO2 emitted, and calories burned during travel. Learning from prior work in eco-feedback studies and behavior change studies about health and fitness, a webpage was designed in which participants received feedback on their travel data along with trends and comparisons with various peer groups.Using surveys administered before and after the experiment, we measured a statistically significant change in partcipants awareness of statistics related to their travel behavior, and an intention to drive less and walk more amongst the "mainly-driving" group of the study population. In addition, a significant decrease in the amount of driving and increase in the amount of walking was measured. However, in a regression analysis, we were not able to find statistically significant covariates explaining what types of people and travelers were more likely to shift.