Eliciting perceptions of travel time risk and exploring its impact on value of time

Abstract

The value of time is crucial to transport policy as it drives the majority of benefits in appraisals. However, individuals can perceive time differently to reality, and this can result in misleading estimates of the value of time in stated preference surveys. Furthermore, there is a lack of methods to determine the distribution of travel time perceptions. This study proposes a model of perceived travel time distribution, including an experimental mechanism to elicit perceived travel time distributions from stated travel time ranges. Two theorems are derived that relate the stated travel time range to the perceived travel time distribution and incentive structure to report accurately. The theorems are validated from a field driving experiment with taxi drivers in Dhaka, Bangladesh, which is also used to study whether stated perceptions elicited with and without incentives are different from actual travel time distributions. The results show that individuals’ perceived travel time distributions are significantly different from actual travel times, and that this can result in statistically significant differences in values of time of up to 17% due to perceptions of lower travel times leading to higher values of time. The experiment also validated the assumptions underlying the theorems. Finally, the study shows that ranges stated under clear monetary incentives follow the theoretical predictions, and that Dhaka taxi drivers on average convey a risk of 77% when stating travel time ranges. The theoretical results presented in this study can be applied more broadly in other domains to study perceptions.