After a bit of an absence, I am going to try and get back to making regular posts. And given that I recently passed 50 posts (!) this post will be about numbers. On the surface numbers are just abstract symbols that we have brought meaning to. Numbers play a huge role in our lives, from understanding prices and value ($4.99 vs. $5.00), to understanding our health (e.g., Blood pressure of 120/80, PSA Gleason Score of 7, or blood sugar of level of 5.8), and to understanding our environment (e.g., 30 degrees, feels like 35, 4 degree rise in global temperatures), but they are often misunderstood. Some numbers are lucky (here’s looking at you 8), which is good news if your address ends in a lucky number. And it turns out that some numbers are more likeable, like 10, 12, and 24. And even cooler than that, some numbers are amicable
The above video is from the excellent site Numberphile, which if you are looking to be amazed, head over there and watch some of their videos (like this one!). Numberphile does a great job of making numbers fun and easy to understand, unfortunately not all numbers are presented in cool videos by British mathematicians, and as such these numbers still remain largely misunderstood.
Perhaps some of the most misunderstood numbers are those that are presented as statistics. Not too long ago, Ed Yong had a great post titled What does it mean to say that something causes 16% of cancers? (see also: What does a 13% increased risk of death mean?). His post was in response to a study that said 16% of all cancers worldwide are caused by infections. A headline which to many would suggest these researches had examined countless medical records and counted the number of cases caused by infections, and determined the number of cancers caused by infection. This is not the case, this number is actually the population attributable fraction (PAFs), which represents the proportion of cases of a disease that could be avoided if something linked to the disease was avoided. This number is not generated from counts, but rather from statistical models based on large datasets. PAFs are frequently used by policy makers, as they are nice, solid numbers, that make good bullet or talking points, but they are blunt instruments and don’t do that great of a job actually communicating what they mean. Enter inforgraphics.
Presenting numerical information with visual representations can often increase numeracy, or the understanding of numbers. One of my favorite ways of seeing information displayed is through infographics. Despite the backlash, there are some great sites which are devoted to presenting cool infographics often with lots of complex information distilled into them (here, here, here, and here), but like the PAFs, these run the risk of over simplifying a situation.
The challenges and success of visualizing data is going to be the theme for the next couple of posts.