International Choice Modelling Conference, International Choice Modelling Conference 2015

Font Size: 
Does Travel-based Multitasking Influence Commute Mode Choice? An Investigation of Northern California Commuters
Aliaksandr Malokin, Patricia L. Mokhtarian, Giovanni Circella

Last modified: 11 May 2015

Abstract


Transportation mode choice modeling has been a well-established area of practice and research for several decades. In the current state of the art, choice is modeled as a function of objective travel characteristics (e.g. in-vehicle travel time, travel cost), individual socio-economic properties of a decision maker (e.g. age, income, car ownership) and (sometimes) personal attitudes. The common premise in these models is the universal disutility of travel time, which travelers seek to minimize, thus increasing their time spent doing activities at locations. However, the modern realization of the multitasking phenomenon, which promises to bring improved productivity and/or satisfaction through overlapping multiple activities “at the same time”, portends changing patterns of time use, and especially (in our context) of travel time use. There is a sizable and growing literature on multitasking in general (e.g. König and Waller, 2010; Circella et al., 2012), and in contexts such as the work environment (e.g. Bluedorn and Martin, 2008; Chesley, 2014) or “media multitasking” (e.g. Wallis, 2010) in particular. Still, the investigation of the activities conducted while traveling, and their impact on traveler’s choices, is a smaller but also expanding area of research (e.g. Mokhtarian and Salomon, 2001; Kenyon and Lyons, 2007; Ettema and Verschuren, 2007; Ohmori and Harata, 2008; Zhang and Timmermans, 2010; Rasouli and Timmermans, 2014).

In this study, we investigate the impact of travel-based multitasking (i.e. the engagement in additional activities while traveling) on the utility of travel, in particular during commuting trips. At the margin, individuals may choose transit (i.e. an attention-passive mode) over a shorter automobile trip, if thereby they are able to overlap travelling with other activities that increase their daily productive output and trip satisfaction. The recent advancements toward partly/fully automated vehicles are poised to further revolutionize the perceived utility of some travel options, increasing the ability to use travel time productively in cars, thus further blurring the role of travel as a crisp transition between location-based activities (Anderson et al., 2014; Wagner et al., 2014).

Accordingly, this study aims to address the following research question: how and to what extent do the ability and propensity to perform tasks “on-the-go” influence an individual’s evaluation of the utility of various travel modes? To quantify this effect, we created and administered a survey (see Neufeld and Mokhtarian, 2012, for more details) that collected information on multitasking attitudes and behavior while commuting, together with information about commute patterns, personal travel attitudes, mode-specific perceptions, and standard socioeconomic traits. The total sample size exceeded 2,000 respondents recruited among Northern California commuters.

Due to the use of various sampling strategies (choice-based sampling, random mail and e-mail blasts, and online panel) and the need to capture enough cases for travel modes that are not very frequently used, the sample is not representative of the whole population of commuters in the area of study. Rather, it underrepresents drive-alone commuters, and overrepresents users of other modes. The person with average characteristics for this sample is a female, around 45 years old, college graduate, and living in a household of 2.7 people owning 2.1 vehicles and earning $75,000 - $99,999 annually.

Multiple measures of multitasking are available in the data. For example, respondents were asked to rate each alternative mode based on the “ability to do things I need/want while traveling” (mode-specific perceptions). Also, the survey collected information about which of a number of different activities they performed on a recent commute (mode-based behavior). Since this information was only available for the chosen mode(s), it could not be used directly as explanatory variables in our mode choice model.  Instead, we (1) modeled the propensity to perform each activity (as functions of individual attributes such as socio-economic characteristics, personality traits, and attitudes) for the choosers of each mode, and then (2) used those models to predict an individual’s propensity to perform each activity for each mode, whether chosen or not.  We use the resulting predicted propensities as explanatory variables in the mode choice model, either directly or in the form of linear combinations (factor scores) obtained from a factor analysis that defined underlying latent constructs such as information and communication technology (ICT)-based productive activities,  ICT-based entertainment activities, traditional social activities, etc.

We use the revealed preference data collected in this project to estimate a multinomial logit model for mode choice that accounts for the impact of multitasking attitudes and behavior on the utility of five mode choice alternatives: driving alone, sharing a ride (either as a driver or passenger), taking public transit (including bus, subway and light rail), riding commuter rail and bicycling. The respondents have unequal choice sets defined by the availability of modes. The estimated coefficients for all core variables (travel time, travel cost, selected socio-economic properties and attitudes) have the expected signs, and are strongly significant across the various specifications that were tested. We find that mode-specific multitasking perceptions are strongly significant and positive, indicating that the perceived ability to perform activities while traveling adds to the utility of all travel modes.

We further find that the estimated propensity to perform ICT-based productive activities while traveling (i.e. electronic reading/writing and using a laptop/tablet) significantly influences utility.  To more specifically quantify its estimated impact, we create several different scenarios, which indicate that multitasking propensity accounts for a small but non-trivial portion of the current mode shares. For example, in a scenario that simulates a high interest of commuters in being productive while travelling (i.e. the highest propensity of engagement in ICT-based productive activities) given present transportation technologies, the drive-alone share substantially decreases in favor of the other modes, which would imply fewer cars on the road and increased transit reliance (however, shared ride captures most of the mode shift). Conversely, in a scenario where autonomous vehicles can offer the same multitasking experience as commuter rail does today (i.e. disengagement from the driving task, convenience and comfort of performing ICT-based productive activities), driving alone and shared ride shares increase at the expense of all other modes, suggesting a potential decline in public transportation ridership and increased traffic volumes in a future dominated by autonomous vehicles.

 

            This is still a work in progress. By April 2015 we anticipate incorporating more complex error term specifications, such as nested logit and cross-nested logit, to relax the independence of irrelevant alternatives assumption. We expect to continue testing a range of ways of including multitasking behavior indicators in the model, to evaluate their impacts more clearly. Additionally, we would like to explore taste heterogeneity through the development of deterministically-segmented and latent class models, where segment definition is based on socio-economic traits and multitasking preferences and beliefs.

References

 

1.            Anderson, James M., Kalra Nidhi, Karlyn D. Stanley, Paul Sorensen, Constantine Samaras, and Oluwatobi A. Oluwatola (2014) Autonomous Vehicle Technology: A Guide for Policymakers. Santa Monica, California: Rand Corporation. Available at http://www.rand.org/pubs/research_reports/RR443-1.html, accessed July 20, 2014.

2.            Bluedorn, Allen C., and Gwen Martin (2008) The time frames of entrepreneurs. Journal of Business Venturing 23(1), 1-20.

3.            Chesley, Noelle (2014) Information and communication technology use, work intensification and employee strain and distress. Work, Employment & Society, doi: 10.1177/0950017013500112. Available at http://wes.sagepub.com/content/early/2014/03/10/0950017013500112.abstract, accessed July 30, 2014.

4.            Circella, Giovanni, Patricia L. Mokhtarian and Laura K. Poff (2012) A conceptual typology of multitasking behavior and polychronicity preferences. electronic International Journal of Time Use Research 9(1), 59-107.

5.            Ettema, Dick and Laura Verschuren (2007) The effect of multitasking on the value of travel time savings. Transportation Research Record 2010, 19-25.

6.            Kenyon, Susan and Glenn Lyons (2007) Introducing multitasking to the study of travel and ICT: Examining its extent and assessing its potential importance. Transportation Research A 41(2), 161-175.

7.            König, Cornelius. J. and Mary J. Waller. (2010) Time for reflection: A critical examination of polychronicity. Human Performance 23(2), 173-190.

8.            Mokhtarian, Patricia L. and Ilan Salomon (2001) How derived is the demand for travel? Some conceptual and measurement considerations. Transportation Research Part A 35, 695-719.

9.            Neufeld, Amanda J. and Patricia L. Mokhtarian (2012) A Survey of Multitasking by Northern California Commuters: Description of the Data Collection Process. Institute of Transportation Studies, University of California, Davis, Research Report UCD-ITS-RR-12-32. Available at http://www.its.ucdavis.edu/?page_id=10063&pub_id=1802, accessed Aug. 20, 2013.

10.        Ohmori, Nobuaki and Noboru Harata (2008) How different are activities while commuting by train? A case in Tokyo. Tijdschrift voor Economische en Sociale Geografie 99(5), 547-561.

11.        Rasouli, Soora and Harry Timmermans (2014) Judgments of travel experiences, activity envelopes, trip features and multi-tasking: A panel effects regression model specification. Transportation Research Part A: Policy and Practice 63, 67-75.

12.        Wagner, Jason, Trey Baker, Ginger Goodin, and John Maddox (2014) Automated Vehicles: Policy Implications Scoping Study. Texas A&M Transportation Institute, Texas A&M University, Research Report SWUTC/14/600451-00029-1. Available at http://d2dtl5nnlpfr0r.cloudfront.net/swutc.tamu.edu/publications/technicalreports/600451-00029-1.pdf, accessed July 20, 2014.

13.        Wallis, Claudia (2010) The impacts of media multitasking on children’s learning and development – Report from a research seminar, The Joan Ganz Cooney Center and Stanford University, available at http://www.joanganzcooneycenter.org/Reports-22.html, accessed July 23, 2014.

 

14.        Zhang, Junyi and Harry Timmermans (2010) Scobit-based panel analysis of multitasking behavior of public transport users. Transportation Research Record 2157, 46-53.


Conference registration is required in order to view papers.