Note – special thanks to Adena Bowden for co-authoring this post.
On Friday, December 9, 2022, we met for the seventh installment of our virtual “Data Literacy for Data Stewards” workshop series to build empathy into our data visualization practices.
Before the workshop, we asked participants to create and re-make a data visualization after reading the “Do No Harm Guide: Applying Equity Awareness in Data Visualization” by Jonathan Schwabish and Alice Feng, published by the Urban Institute. By refining their initial visualization, participants were able to immediately adopt principles from the guide into their practices. We reviewed a few examples at the beginning of the workshop.
We then directed participants to review and discuss a summary of recommendations from page 41 of the “Do No Harm Guide.”
Participants expected to use the following recommendations most often:
- Ordering labels purposefully;
- Communicating with people and communities in focus;
- Critically examining the data used in the visualization;
- Using person-first and inclusive language, labels, icons, and shapes;
- Reflect lived experience in the data visualization;
- Considering missing groups;
- Considering alternatives to labeling the “other” catch-all category;
- Considering the needs of the audience.
Attendees planned to include these recommendations into their own work by:
- Centering lived experiences;
- Critically examining all features of data visualizations;
- Ensuring that the labels reflect how people identify themselves;
- Ensuring communities are directly involved in creating the data visualization.
Our group would add the following advice to the guide’s recommendations:
- Making data visualizations more engaging;
- Being mindful of the audience, purpose, and intentions up front;
- Creating data visualizations in a collaborative, iterative process;
- Adopting accessible design, including choice of colors and alt. text;
- Using less-stigmatizing language (e.g. “income qualifying” rather than “low income”);
- Providing multiple ways to interact with the data.
We then critiqued a visualization produced by KFF together through the lens of the “Racial Equity in Data Visualization Checklist” on page 42 of the “Do No Harm Guide.”
Some of our observations and notes were that:
- Subgroup labels are not person-centered;
- Using “better,” “worse,” and other comparative language assigns value and can cause harm by implying some people are “less than;”
- While the colors didn’t stereotype, the color choices could have been more-accessible;
- Racial subcategories could be ordered alphabetically rather than ordered from the “worst” to “best” performing groups;
- Each individual racial group could have been presented as separate charts;
- The chart centers whiteness, since all other races listed are compared to the white population;
- More context about structural causes of health disparity could be presented.
For more information on creating and consuming data visualizations in a diverse, equitable, and inclusive way, we encourage you to check out the following resources:
- Do No Harm Guide: Applying Equity Awareness in Data Visualization by Jonathan Schwabish and Alice Feng, published by the Urban Institute
- Dear Data, a year-long, analog data drawing project by Giorgia Lupi and Stefanie Posavec
- Storytelling With Data: A Data Visualization Guide for Business Professionals
- ColorBrewer: Color Advice for Maps
This Friday (December 16, 2022), we will discuss data privacy. We will talk about the risks and benefits that can come from data sharing, become familiar with frameworks for balancing harms and positive outcomes, and learn about methods to protect people included in data.
If you are interested in participating in the next cohort of our Data Literacy for Data Stewards peer learning series starting in the first quarter of 2023, email us at email@example.com and we will let you know when registration is open.