Work With Data Rather Than Making It To Work For You
Everyone knows the stereotypical kid with food allergies. The one that can’t go to restaurants and sits at a special table in the cafeteria. I was not this child and found that independence at a young age helped me learn self-advocacy when it came to my food allergies. For my Information Design assignment, I wanted to use data to prove that independence for kids with allergies is crucial. So, I took what I thought was “raw data” at face value to encourage parents of children with food allergies that their child needs to be independent. However, data is never neutral, as Danah Boyd and Kate Crawford explain in Critical Questions for Big Data, and access to contextless data enables people to use the data to prove their specific beliefs.
Catherine D’Ignazio and Lauren Klein emphasize in their book, Data Feminism, that not only does context matter with data collection, but “context also comes into play in the framing and communication of results.”
After starting my research, I soon realized there is not much data that proves children with food allergies should be independent. So, though I know the potential harm of this now, I made the data I found work for me rather than trying to work with the data and consider its context.
I used food allergy hospitalization data overtime to support my argument, even though one would typically interpret an increasing trend in hospitalizations as evidence to keep children sheltered. Unknowingly, I failed to “frame context in [my] data visualization”, and this could have been harmful.
Luckily, this graphic was only for a school project. When I enter the real world, it is my responsibility and duty as a designer to be conscientious with the data I visualize by examining the context of how the data was gathered, what biases exist, and how my conflicts of interest affect the communication of the data. Although I have a newfound awareness, the potential for harm from manipulating data still exists.
To prevent this from happening, there needs to be an expectation for providing context with data collection and visualization so people can have a clear picture of what they are being told and how it is being told to them.
