DataSelfie: Empowering People to Design Personalized Visuals to Represent Their Data
This article summarises a paper authored by Nam Wook Kim, Hyejin Im, Nathalie Henry Riche, Alicia Wang, Krzysztof Gajos and Hanspeter Pfister. The paper will be presented at CHI 2019, on Thursday, May 9th 2019 at 9:00 AM, in the Communication through Visualization session, in the Hall 1 room.
Quick Takeaway
We investigate the question of how to enable people to determine the representation of their data. DataSelfie allows any individuals to create a visual vocabulary to represent their personal data.
Personal Informatics Systems
More Convenient but Less Engaged
Many personal informatics systems allow people to collect and manage personal data and reflect more deeply about themselves. The manual approaches for data collection have high data capture burdens, often hindering people from sustaining long-term practices. Automated tracking technology attempts to address this issue by leveraging personal devices embedded with various sensors to remove the need for manual data inputs. However, complete automation of data collection often eliminates additional opportunities for engagement and reflection with personal data.
Lack of Support for Customizing Visualization
Existing self-tracking tools focus on data collection and management, not necessarily on how to display the data. As a result, the design of the tools dictates the presentation of the data, showing data summaries in the form of standard charts or tables. People rarely have a chance to customize their data and have passive ownership of the visuals.
Advantages of Expressive Visuals Using Pen & Paper
On the other hand, a lot of people still use pen & paper to create much more expressive representations that can communicate their emotion and personality, as well as the context of the data much better than standard charts.
Open Coding of 104 Postcards
To understand what types of data can capture personal lives and what kinds of visuals can be used to represent the data, we analyzed visualizations and retrospective texts of 104 postcards from the Dear Data project in which two professional infographic designers, Lupi and Posavec, collected data about themselves and drew custom visualizations with the data every week for a year.
Design Elements to Represent Personal Data
While visual postcards revealed common design elements to represent personal data, retrospective texts allowed us to gain insights into the underlying rationale for the design choices and identify different stages of the process.
Visual Postcard Design Process
Implications for Personal Informatics
- Capture qualitative aspects of self
- Reveal missing and uncertain data
- Provide different modes of data collection
- Support data exploration for story harvesting
- Use visual annotations to encode moments
- Enable designing personalized visuals
- Choose visual variables for personalization
- Leverage visuals as personal documentaries
- Address challenges with qualitative processes
- Support conversations through data
DataSelfie
Our design goal is to provide structured support for novices to crisply articulate the data they want to collect and visualize. We developed DataSelfie, a novel system that allows individuals to create a questionnaire to ask questions about qualitative and nuanced aspects of ourselves and to design a personalized visual vocabulary to represent the collected data, and share with others.
To assist users with less experience in drawing, especially with a mouse, we support auto-drawing, duplicating a shape across all options in a question, along with the automatic size and color encoding.
User Study
We conducted a user study with 14 participants to evaluate the usability of DataSelfie as well as its potential for individual and collaborative sensemaking of the data.
It allows you to visualize goals and impacts in advance — P9
I like visuals as they invoke more thoughts. — P7
Freedom to draw anything to represent my data makes it fun and lively — P13
It would be good for emotion and well-being that lack clear forms — P8
Actionable Conclusion
We believe that personal informatics systems that allow custom visualization benefit people to determine the representation of their data.