Pitfalls to Avoid in Communicating Science Through Mainstream Media
The communication of science used to be limited to scholarly collaborations through restricted (not open source) media. While that had previously been the case and scientists were considered beckons of information, recently the public audiences have begun turning online for information about science. Rightfully, the public should be allowed to fill the void of their curiosity through their own means. In return, the scientific body should be allowed to review and collaborate with the media to help forward this information to the public. When citizens lack opportunities for a broader discussion with scientists and interaction on the scientific matter, polarization becomes a concern. In simple terms, there is a disconnect where scientists cannot communicate with the public. So where did I come to these philosophical conclusions? This all stems from reading a popular news article published in my technical wheelhouse that exhibits poor data visualization, communication about the numbers (concentration units) used and has a focus on a specific narrative instead of presenting the analytical findings from reputable scientific researchers.
This article from the CBC addresses per- and polyfluoroalkyl substances – known as PFAS - from an environmental and biomonitoring standpoint within Canada. PFAS, more popularly known as a class of ‘forever chemicals’, have reached the attention of the public recently and sparked the need for more regulation in Canada. This topic has even been adapted into a very interesting movie!
Units of concentration and what are PFAS?
Not to bore you with details, but the first issue of the article is that they do not address what PFAS are. These are a class of synthetic organofluorine chemical compounds that have multiple fluorine atoms attached to an alkyl chain. So why do I think the understanding of the chemical structure is important to the general audience? For one thing, we are looking at a class of compounds that contains a large amount of different chemical structures and not a single individual compound. For instance, EPA lists over 10,000 different PFAS compounds in their internal toxicity database. So when the article begins comparing the concentration of a residential groundwater site, they have zero specification on which compounds are being looked at. Additionally, it is unclear the methodology that the groundwater assessment used to quantify, “PFAS”, within their sample.
Geographic information and environmental assessment
When the article proceeds to visualize the PFAS data that they collected in environmental samples, this is where I start to cringe. What I got from the number is that they did a poor job at visualizing the environmental monitoring data and just put numbers on. They had no reference to gauge if these were high concentrations or an easy-to-read data visualization. They used bar charts on a 2D map where you can’t pinpoint the location. There is an oversampling leading to overlap in Ontario (Mimico where I grew up isn’t in Georgian Bay …).
Instead of just critiquing the data, I took it a step further and created how I would interpret the data. The limitation is the spatial awareness of these sampling points. Where are they exactly located, how big are they in relation to each other, are these hotspots linked to industrial activity, and are these supposed to be background samples? My simple solution was to run this same map but in 3D space. Now we can see the hotspots clearly and their relationship in size to the other sampling regions! In this day and age adding these types of interactive digital maps should be an essential component to let the audience play with the data. Try it out for yourself. Use your mouse to spin the map, tilt the angle to see the size of the 3D bar graph and zoom in and out to see all the sampling points.
Figure 1 - This is the overlapping data visualization used by the CBC article which does not allow to pinpoint the location or easily compare magnitudes between concentrations.
What can be next is to add additional samples to this to see how other environmental sites compare to these. There are likely background concentrations that may be below some of these samples, and it would be important to showcase the magnitude of a known area of PFAS contamination (Borden airport in Ontario for example?). What I’m getting at is a number without much context isn’t an effective narrative to present to the public and a more critical evaluation should be conducted in collaboration with active experts in the field.
If you have any data that you’d want to visualize using these techniques, I’m always happy to work on a new data visualization project!
