The Subtle Ways Data Can Undermine our Advocacy
Co-written by Sylvina Poole of the VEJC and Isa Gaillard of the Greenlining Institute
In recent years, people have started paying more attention to the role that data -- and how it’s framed -- plays in driving policy. This plays out in a number of ways.
For example, advocates have highlighted the need for maps that visualize how socioeconomic barriers and environmental risks associated with pollution and climate change combine to produce cumulative impacts on communities. The Virginia Environmental Justice Collaborative and Mapping for Environmental Justice took this conversation one level deeper by incorporating narratives from community organizing groups across the Commonwealth into a map that visualizes cumulative environmental justice impact across the state. The key here being that these communities have shared ownership of the data because they were able to tell their own stories rather than have the narrative written for them.
Another way to ensure data is used to uplift communities is to acknowledge the risks associated with how we frame the narrative when using data to make a case.
For example, the TEEM Community of Practice collaborated with VEJC to organize a webinar in which Lakeshia Wright of Guided by Community led a presentation and discussion on data equity. Lakeshia’s presentation, “Is That Datapoint Racist?” called out five key issues where the framing of data could lead to the disempowerment and even harm of Black and Brown communities. The five issues include: aggregation, burden, interpretation, weaponization and exploitation, and implied comparison. Here, we are going to focus on two of these issues: aggregation and burden, because these two are arguably the most subtle and often perpetuated, even by those who mean well.
Aggregation refers to the grouping of multiple pieces of information into a single datapoint. When we group different categories of people together and present a data point about this group we risk telling a story that does not capture key underlying differences in how factors
such as a person’s race, gender, age, or income lead to different outcomes related to that data point. Here’s an example: “The number of electric vehicle charging stations in Virginia increased from 995 to 1,601 between 2016 to 2018, a 62% increase over two years.” While interesting and broadly useful, this statistic aggregates data from across the entire state and can’t tell us where specifically these charging stations were installed. Disaggregating this data by city, or better yet, by neighborhood would help us understand who is benefiting from the increase in charging stations and who is not. In some places, EV charging infrastructure has been concentrated in wealthy neighborhoods -- something a single statewide total can’t tell us.
The second issue stems from how the data is framed, and on whom that framing places the burden. Framing often places the burden of change on the population that is most negatively affected. For example, “Research has shown significant racial/ ethnic and income
disparities in asthma rates, with populations of color and people in poverty experiencing consistently higher asthma prevalence rates and poor asthma outcomes compared with the general population.” This analysis does not include any exploration of why this disparity in asthma rates occurs across socioeconomic and racial groups, leaving it up to the reader to draw their own conclusions --which could be based on incorrect assumptions.
This analysis should be reframed to place the burden on the root causes of asthma disparities, such as disproportionate sources of pollution and traffic in Black and Brown communities. A more equitable way to phrase this would be, “The air quality in low income communities of color tends to be worse than in more affluent or majority white communities due to higher amounts of toxic waste sites, vehicle traffic, and lack of air pollution mitigation measures such as the planting of trees. These discrepancies lead low income communities of color to have higher asthma rates than white communities”.
Both of these issues, aggregation and burden demonstrate the complicated nature of data and how it’s presented. They also point to the need for us as advocates and policymakers to dig beneath the surface of the data and information we are presented with. This requires us to spend more time considering the meaning behind the statistics and data trends we read and see graphed in the daily news. It also challenges us to consider the narratives being told by the data we use, and the unspoken assumptions that may underlie how it’s framed. When the narrative reinforces unjust histories and systems of oppression, it’s up to us to rethink and reframe.