Case Study

Time-Use in Commuting


Much research on the environmental impacts of driving habits in the US has led behavioral scientists to ideate on the best ways to encourage use of more sustainable transportation modes. Misconceived perceptions of travel time for options other than driving can dissuade people from exploring alternatives. We developed and launched an online survey to better understand people’s perceptions of various transportation methods and the time associated with using each.


The study, which ran in spring 2021, aimed to assess participants’ estimates of the travel time and potential productivity associated with various transportation modes. For the study, 650 participants were sourced from the Amazon Mechanical Turk platform to participate in a survey about commuting perceptions. 

The first part of the survey randomized participants into the following two conditions:

  • Single estimate- asked participants “Imagine you [drive, cycle, or ride the bus] to work. How many minutes would your commute take?” Participants were asked for a numeric response. 
  • Split time estimate- asked participants “Imagine you are going to [drive, cycle or ride the bus] to work. How many minutes would it take to complete each of the steps below?” After this question, a step-by-step checklist was listed for the specific transportation method outlined and participants were asked for a numeric estimate for each step. 

The second part of the survey asked all participants how “productive” they would feel during a commute using various transportation modes (productivity not pre-defined). Additionally, participants were asked which mode they would choose if they wanted to achieve productivity socially, in terms of work, and in terms of leisure. 


We found that split time estimate survey participants responded with significantly longer commute time estimates for driving and riding the bus compared to responses from single estimate survey participants. Participants who were asked to consider each individual step of driving averaged an estimate of 37 minutes of commuting time compared to participants in the single estimate condition whose responses averaged an estimate of 24 minutes.  For riding the bus, split time condition participants responded with an average estimate of 67 minutes while single estimate condition participants responded with only an estimate of 45 minutes.  

Interestingly, the differential between driving framed in terms of individual steps and riding the bus framed as a single step is only 8 minutes.  For riding a bike, the single and split time measurements only differed by 5 minutes. This suggests that to encourage sustainable commuting, we should get people to think about time use in a single measure when riding the bus or a bike and to split out the steps when thinking about driving.

Figure 1. Comparison of estimated minutes commuting from the single and split condition groups. 

In the second part of the survey, participants felt that riding a train was the most productive commute option in comparison to other forms of transportation. Driving was deemed to be the least productive commute mode. 

Figure 2. Survey participants’ feelings of productivity during a commute by various transportation methods. 

Zooming in on specified forms of productivity, participants most frequently rated driving as the transportation mode that made them feel most socially productive.  This was surprising given the individual nature of driving compared to using public transportation or carpooling.  In terms of work or leisure-related productivity, participants most often cited riding a train as the best option. 

Figure 3. Survey participants’ feelings on the best transportation method for each “type” (social, work, and leisure) of productivity.

Why It Matters

Applying these findings can help achieve behavior change as we prioritize encouraging use of sustainable transportation modes. Framing of the time required may have a significant impact on people’s feelings toward not driving as length of commute is an often-cited reason for not using sustainable options.  Respondents were fairly consistent in estimating their bicycle commute times, but underestimated their driving commute length when framed as a single-point estimate.  

In places where public transit resources are limited, it may be especially beneficial to explore why people feel that riding a train yields high productivity in comparison to riding a bus.  These findings also demonstrate the benefit of framing sustainable commuting options as productive in order to increase the likelihood that people will select these options for travelling.