How small modifications can change the perspective of a story
Keeping charts and graphs simple is essential to give the correct perspective to the wider audience.
I’d like to kick this one off with a quick introduction as this is my first post here on Medium. I’m Renee and I’ve been working with data for over 11 years. My main focus has been on storytelling and visualizing data, so if I see something that doesn’t get the story out, then I like to either ignore it or improve it. As we know, data is all powerful and if you misrepresent it, you can create a whole new narrative that just doesn’t tell the correct story.
Yesterday I was reading an article that visualized data about the news headlining coronavirus 2019-nCoV. What captured my eyes were the following two graphs from the article:
Graph 1 — Infected around the world (cumulative)
Graph 2 — Death toll (cumulative)
While in their essence, they look ‘okayish’. they tend to tell an incorrect story. First of all axis’s. On Graph 1 the axis runs from 0–40000 while on Graph 2 it runs from 0–800. That is a 50x difference. Hence if we are comparing such numbers, we should always use the same axis values.
What can we improve or do differently in order to make this chart work and give even more information to the readers? In essence, there are multiple ways to approach this.
Option 1 — Use the same axis values for both charts
Option 2 — Combine those charts into one. You wouldn’t waste valuable space and you would get a better idea of what is actually going on.
Option 3— Mortality Rate. While virus infection spread and its death toll is valuable, their ratio has even higher impact. This showcases if the threat is increasing or decreasing and if something has actually changed in its behaviour. In this case, the most powerful chart to spread the message would contain 3 indicators: Infected, Death toll & Mortality Rate.
So let’s create Option 3. Do note that I’m using the same data as was used to create the charts above and the data is until the 6th of February.
What did we achieve?
If we look at the story it isn’t as bad as the journalist initially showcased in their graphs. While we do see an increase in Infected, specially since Feb 1st, the Mortality Rate has actually decreased since the tracking began, decreasing from 2.8% to 2.0%. As always we have to take those numbers with a grain of salt as the numbers will continue to increase and things will change over time. For reference 2019-nCoV mortality rate is lower than the rate for SARS or MERS (SARS 9.6%; MERS 34.5%).
While the analysis part does play a role, the main focus of this article was on the visual perspective. I took the view of the journalist and created a chart that shows more information and where everyone can base their own assumptions on. Previously if you didn’t pay any attention to the axis’s, you got a completely different story and you were missing a key component — Mortality Rate.
So we cleaned up the chart and data, added additional information and used just one graph to tell the complete story.
Dear journalists, please look and use your data carefully as you are communicating your message to the wider public. The charts might look cool and blingy, but those charts can easily mislead your readers and not all readers are data literate.
Sources:
Data & Graphs — Delfi Forte — https://forte.delfi.ee/news/varia/uuenev-kaart-ja-graafik-mitu-haigusjuhtu-lahtub-uuest-murettekitavast-allikast?id=88770709
SARS — World Health Organization (WHO) — https://www.who.int/csr/sars/country/table2004_04_21/en/
MERS — World Health Organization (WHO) — http://applications.emro.who.int/docs/EMROPub-MERS-SEP-2019-EN.pdf?ua=1&ua=1
