The "how, where and when" of physical activity

The "how, where and when" of physical activity

Despite their initial premise of physical-activity trackers in altering individuals’ lifestyles, recent studies have questioned their long-term efficacy. We present a study that sought to understand the walking activities individuals incorporate into their daily routines when attempting to be more physically activity – and specifically, the “why, how, when and where” of physical activity.


Ornelas, T., Caraban, A., Gouveia, R., Karapanos, E. (2015) CrowdWalk: Leveraging the Wisdom of the Crowd to Inspire Walking Activities. In Extended proceedings of Ubicomp’15. 

A total of 65 participants (mean age=37, 46% male) successfully completed an online survey. All participants owned an activity tracker such as Fitbit, Jawbone Up or Nike Fuelband. They were mostly from the USA (46%), India (25%) and Australia (11%). They were recruited through Mechanical Turk and were rewarded with 0.30 euros for their participation. Responses were screened for quality by the first two authors.


Participants were asked to identify one “walking activity” embedded in their daily routines after adopting the tracker. We asked them to narrate it in detail as to when and where it takes place, their motives for performing it and the challenges they face (if any) in keeping up with it, as well as a number of closed questions relating to the estimated distance of walking activities and their frequency.



Data was analyzed using affinity diagrams. All in all, twenty-seven distinct activities were identified. The vast majority (91%) was performed at least once per week, while 57% were performed on a daily basis.



Walking activities were mostly short in distance, with more than half (59%, N=38) reaching a maximum of 1km. The second most popular group (15%, N=9) were activities that covered approximately 5 km in length, followed by walks of 1-2km (15%, N=9), 2-3km (6%, N=3), 4-5 km (3%, N=2) and 3-4 km (2%, N=1). All in all, walking activities that were performed daily, increase walking distance on average by 2 km.


As expected, 61% of individuals responded that their primary motivation for performing walking activities was to improve their health condition over the long term, such as losing weight, counterbalancing sedentary behaviors and compensating for consumed calories. However, we found another 35% to relate to non-health motives, but rather to the situated benefits from physical activity such as boosting one’s energy (13%) and enhancing her mood (8%) as well relieve stress (8%) and taking a break from work (6%). Finally, another 4% of responses tapped to economic reasons for performing the walk activity.


Rather than adding new walking activities to their daily routines, we found individuals often to sub optimize existing routines as a means to increasing their physical activity levels, such as when doing household chores: [P2] “I always try to make several trips … when I’m doing chores. I do this several times a day, so it adds up to quite a bit of walking”, parking the car at a distance, e.g., [P7] “… I park there with the conscious intent of being able to walk further to get exercise”, going for lunch to a distant place, e.g., [P11] “… I go to the far end of the campus to get food … I also get a little exercise. I feel good about doing this”, or including walking as part of one’s commuting, e.g., [P5] “I try to walk some stations instead of using the tram”.


All in all, we found walking activities to be sustained in daily routines due to three main reasons: (1) they do not require much effort, which makes them easy to perform, e.g., [P27] “It was easy to sustain as it is not a long walk”, (2) individuals attain instant rewards from these activities, such as feeling better or healthier and achieving set goals, motivating them to continue, e.g., [P28] “This is sustained because I’m seeing results”, and (3) the variety of opportunities arising on individuals’ daily routines makes them to be easily adopted at different locations and daytimes based on individuals’ preferences and schedules, e.g., [P18] “I am a teacher so I ensure that I actively walk around the classroom monitoring students, I try not to sit down during the lesson”.


Implications for design

All in all, the study highlights the importance of task and location compatibility in building sustainable walking habits, while it highlights a breadth of motives for walking that go beyond the health and competence narratives that dominate today’s discussion on activity tracking. We believe that tying walking activities to places will provide a number of benefits. First, tying activities to places enables the creation of moments of choice [6]. Over time, users are expected to link specific parts of a routine with actions through the creation of if-then scenarios (e.g., whenever I enter the building, I take the stairs) [7], thus supporting the process of habit formation. Cues in the physical environment (e.g., the entrance of the building) should gradually become visible reminders powerful enough to activate the alternate path (i.e., taking the stairs) at a subconscious level [8]. In this sense, CrowdWalk should serve to provide only the initial triggers and inspire alternate paths, until these become habituated activities.