Duration: Febr – July 2013
During the past five years people all over the world have been recording data about themselves and their surroundings (self-tracking), producing a flux of information with accelerometers. Technology, such as accelerometers, can be supportive in this process and very useful in assisting people with increasing their physical activity. However, with the increasing amount of generated and collected data available, the efficiency of their use becomes more challenging.
This study aims to evaluate the effectiveness of using a visual analytics tool to facilitate insight into how this daily physical activity data yields to reflective learning, also known as learning by returning to and evaluating past work performances and personal experiences in order to promote continuous learning and improve future experiences (Boud, 1985).
To find out how to present analysis of Fitbit data for its user to gain optimal insight, both quantitative and qualitative research were conducted. Based on gathering data from a 4-week field trial in which 24 participants used a Fitbit and a GPS tracker, design and interaction recommendations were made by taking the Fitbit dashboard as a starting point.
A dashboard was developed according to the visual analytics model proposed by Keim, Mansmann, Schneidewind, Thomas, and Ziegler (2008). This model describes the iterative process of data gathering and exploration, hypothesis building, visualization design and analysis, and the gaining of insight by directly interacting with the gathered data.
Based on the gathered Fitbit and GPS data and an iterative user feedback process, a redesign was made for the Fitbit dashboard. The study is described in a thesis for the Master Information Studies, Human Centered Multimedia.