Note: Images may be inserted into your messages by uploading a file attachment (see "Manage Attachments"). Even though it doesn't appear when previewed, the image will appear at the end of your message once it is posted.
Register Latest Topics

  Author   Comment  

Posts: 18
Reply with quote  #1 

This is actually one of the better ones from David McCandless. However, I still put it here due to the y-axis: It's partially logarithmic scale; partially linear with no visual cues what so ever.

His grouping of contagious into "not very", "quite contagious", "very", "highly" and "vaccinate now" seems a bit arbitrary. Especially since vaccination strategy is not only dependent on how contagious the disease is, but also how sever the complications are. This is why we see several vaccine-preventable diseases in the left side of the chart.

Also, if this chart is made to compare Ebola to other diseases, Ebola is not in any way emphasised to stand out. I knew approximately where Ebola should be, and I still had to look for it. 


Posts: 194
Reply with quote  #2 
Not surprising that the lease offensive chart I've ever seen from McCandless still needs work.

Posts: 812
Reply with quote  #3 
Oh my, this y-axis scale is indeed misleading. According to the data, Typhoid is 20 times as deadly as Salmonella, but the chart suggests that it is only twice as deadly. It's hard to believe that McCandless started out as a journalist, isn't it? I just returned from Miami where I spent two days with Alberto Cairo, author of The Functional Art. Alberto has been invited by McCandless to serve as a judge in the next Information is Beautiful competition. Perhaps Alberto will have an opportunity to remind McCandless that journalists are supposed to tell the truth.
Stephen Few

Posts: 8
Reply with quote  #4 
I assume the decision to use a partial logarithmic axis was to prevent all the diseases w fatality rates around 1% from being scrunched together. What would be the suggestions to improve this chart? Simply remove the logarithmic portion? Or perhaps use two separate charts - one w 1-100% and the other w 1-5%?

Posts: 194
Reply with quote  #5 
@beecom99 - I would probably approach this by making one chart with all of the data, at normal scale, and a second chart that zooms in on just the data that would end up scrunched together at the bottom.

I wouldn't make two charts showing two different sets of data.

Previous Topic | Next Topic

Quick Navigation: