First Data Mural Pilot | MIT Center for Civic Media
Rahul Bhargava is a researcher and technologist specializing in civic technology and data literacy. He creates interactive websites used by hundreds of thousands, playful educational experiences across the globe, and award-winning visualizations for museum settings. As a Research Scientist at the MIT Center for Civic Media, Rahul leads technical development on projects ranging from interfaces for quantitative news analysis, to platforms for crowd-sourced sensing. He has a special interest in how new technologies are introduced to people in settings focused on learning. Rahul is a drummer and father based in Somerville, MA.
First Data Mural Pilot
As you might have seen in my short talk earlier this year, I'm excited about taking Data Therapy in a more artistic direction. Towards this goal, my collaborator/wife Emily and I just facilitated a tiny Data Mural pilot at the Doctors for Global Health General Assembly. The idea of a Data Mural is to work with a community group to collaboratively tell a data story in visual form. Our primary goal for this pilot was to test out some of our facilitation techniques. I'm writing this blog post to share lessons from this pilot, and to get you excited about the idea! Email me if you have a Data Mural project in mind.
I'll start off by sharing the final mural. The participants liked it so much that they are taking it to display at a protest in New York City next month! Here are some more pictures.
Doctors for Global Health (DGH) is a small, all-volunteer organization that cares deeply about health disparities and works to rectify them around the globe. Full disclosure - I've been a member of DGH for years and my wife Emily is a previous board member. I guess that means the crowd was predisposed to be forgiving, but that's a great setup for a pilot workshop! Their annual general assembly took place in Boston this year, so it was a great chance for us to try to some stuff out with a crowd.
After prying our way onto the agenda, we decided to host a Data Mural lunch session with anyone interested. Our goal was to try out some of our facilitation techniques, despite the super limited 1.5 hour time slot. DGH is primarily made up of doctors, but also involves many public health professionals, community health workers, and artists. They're earnest about their their work, but committed to having fun while doing it.
Our whole planning was all built around the limited timespan. Below is an intertwined summary of the agenda and reflections/feedback about how each step went. A participant also posted a short live blog of the workshop.
Overview, introduction and goals
We started off by making sure to describe the desired outcome - namely a sketch that we could paint together later in the evening at an already-scheduled fundraiser party. This meant we'd have a good public opportunity to engage people in painting the sketch we came up with, and it provided an opportunity for participants to show off their involvement.
Present datasets and review them together
Based on DGH's focus and interests, I pulled together some sets of health data about the countries DGH works in. I created simple spreadsheets focusing on education, general health, and maternal health. This is more generic than we would do with a community partner focused on just one topic, but it seemed to fit the audience. Our idea was that people would pick one area to focus in on, and then review that data to find stories.
It turned out that DGH's focus is so broad that participants had a hard time deciding on one topic area - they linked education and health so closely that they weren't willing to examine just one topic area. I completely agree in the abstract, but that richer definition of health certainly made this step harder. After a bit of discussion we decided to split into two groups and looks for stories in both areas. We didn't have time to engage some sort of consensus or voting process to decide.
Share and evaluate stories found in the data
This step was designed to have the group to work together to review the raw data sheets and find any interesting and powerful stories. To facilitate this, we wanted them to summarize the stories in this format:
This only kind of worked. We shared the stories between the two groups and highlighted the ones that people found most powerful. Many of the participants were unwilling to give up the nuances of the data in favor of a simple story (this is a classic dilemma, of course). Someone would make a suggestion, then another participant would reply that the story left out some critical subtlety of the issue. At a certain point one participant suggested that maybe we step back and just tackle the underlying question of "Why do we live in a world where..." People liked this, so we decided to go with it, despite the departure from our pre-designed story format.
A comment left on a feedback form the next day was insightful. It suggested that we change the agency of that sentence from the data to the person saying the story. Perhaps something like "The data makes me think _______. That is interesting because _______." The comment suggested that phrasing would make people less likely to talk about the lack-of-richness in simple datasets, because it is about the interpretation of the data, not the data itself. We're going to have to ruminate on this one, because it may depend totally on the participants.
Turn the stories into visuals
To start turning the stories into visuals, we decided to use a collaborative sketching exercise that Emily learned from some muralists she has worked with before. We set up the chairs in a big circle around a table, gave each person a uniquely-colored pen, and gave each a white piece of paper. They had two minutes to start sketching a visual of the story we had just identified. After those two minutes we told them to stop. Then we told them to pass the paper to the right, and with the paper they received they had two minutes to build on the sketch they received (no questions about what it means). We continued this for about 6 or 7 passes.
This totally worked. It is a super fun game of "visual telephone". People were hesitant to start sketching, but once they saw what others were creating, their nervousness disappeared. I think folks were worried about their lack of drawing skills, but once they realized no one is a 2-minute Picasso, they settled down. We may need to think about this more in groups with participants that don't already have a safe and trusted history of collaborating. Then again, drawing is fun, so it might not be a problem at all!
Identify powerful visual elements
Our next step was to share the collaborative sketches and identify powerful elements to use in the final mural.
After some individual sharing, we realized it was smarter to spread out all the papers and look at them together. Luckily a participant had some stickies, so we used those to mark drawings and indicate the elements that resonated with folks. We found a number of visual elements showed up in multiple drawings... a good sign that they were powerful and resonant for the group.
Combine the visuals into a mural sketch
Emily then started sketching a design that brought together many of the elements people identified as powerful. This ended out workshop session, and we told the participants that we'd see them later and wanted their help painting the mural at the evening celebration.
This step suffered from the time constraint, like others. There were a few voices dominating the discussion, but we didn't have time to create a process that would address that.
While drawing a participant asked a great question about how much this result was related to the data discussion and stories we had identified earlier. This gets at one of the core hard questions of this Data Mural idea - how much data needs to be in the result to count as "data driven"? The group certainly took ownership of the data, and the resulting story was backed by the data. That said, the idea of global health disparities could be seen as so obvious that it isn't a "data story". I wonder if we'd run into this with a more focused community partner in a specific geographic area. There are no right answers to this question of "data driven-ness" - it is something for us to continue to think about.
Paint the mural and share it
Emily then took the sketch and turned it into a mural design on canvas. In the evening we had a whole bunch of folks help us paint it!
This worked really well - we had people from age 2 to 90 help us out! We set it all up on canvas with watered down acrylics.
As a pilot to test out our techniques and process, this was a great success! In fact, the mural is heading right now to New York City to be presented at a protest march in September, so clearly DGH found the experience fun and the outcome useful. Drop me a line if you want to hear more.