Advancing Personal Informatics through Semi-Automated Tracking
The research objective of this proposal is to investigate semi-automated tracking as a self-monitoring approach to balance information needs and data capture burden while enhancing user engagement. Self-monitoring offers many benefits including increased awareness and reduced negative behaviors, and serves as a building block for other behavioral change interventions such as goal setting. But it also imposes a high data capture burden, preventing broad adoption and long-term engagement by users. As a result, many self-monitoring applications use automated sensing for data collection to reduce the awareness, accountability, and involvement achieved when a person actively engages in manual tracking. Contrary to this conventional approach, this research proposes a hybrid self-monitoring approach called semi-automated tracking, which combines manual and automated tracking methods to capture diverse data types in varying levels of data granularity. I hypothesize that semi-automated tracking can enhance user engagement while reducing data capture burden, thereby positively affect the target behavior. Through formative studies, intervention design, and deployment studies with older adults and surgical patients, who have extra challenges with self-monitoring, this research examines semi-automated tracking’s efficacy in supporting individuals’ diverse self-monitoring needs, adherence to tracking, and adherence to health behaviors.
In particular, this research examines the following research questions:
What challenges do older adults and surgical patients have in practicing self-monitoring?
How does semi-automated tracking affect older adults’ and surgical patients’ self-monitoring practice and their engagement in health behaviors?
How can semi-automated tracking facilitate collaborative self-monitoring?
This research will achieve three research outcomes:
(1) advancing the theoretical and empirical understanding of older adults’ and surgical patients’ self-monitoring practices;
(2) providing a novel approach to facilitate self-monitoring that balances information needs and data capture burden; and
(3) developing a research platform that allows flexible configuration of semi-automated tracking and collaborative self-monitoring that embodies the findings of the investigations.
June 2017 - January 2022
- Accessibility and Inclusive Design
- Data Science, Analytics, and Visualization
- Health Informatics
- Human-Computer Interaction
Total Award Amount: $546,348