I have a love/hate relationship with pie charts – when they are used well, they are a brilliant way of showing proportions (x is bigger than y, which is bigger than z) and seeing where a particular slice fits in as part of the whole (mmm pie). I’m certainly not the first person to wax lyrical about pie charts, I’m aware that Nathan Yau has demonstrated a good use of pie charts and I totally agree with him.
However, as I was browsing the web this morning, a story caught my eye in The Mail (don’t judge me) about Britain’s crime hotspots and how Stratford in East London is awarded the country’s worst crime hotspot (I was a Games Maker at the 2012 Olympics and Stratford is very much of interest to me). And then I saw THIS MONSTROSITY…
Daily Mail Pie Fail
Source: Daily Mail
WHOAAAHHHHH THERE – I’ll give you a couple of minutes to digest that beauty.
This is a perfect example of why I also hate pie charts…I think it should be reported to the data viz police.
Here’s a few of the problems:
- It contains far too many slices – it’s information overload and it’s really hard to compare categories, it’s just a sea of labels.
- It’s 3D which really isn’t the best way to project proportions as they are liable to misinterpretation; I can’t really say it any better than Drew Skau on the wonderful Visual.ly blog.
- It’s not in any order, again making it harder to read (I personally prefer a pie chart in descending order with highest proportion first).
In fact, I think it breaks every rule in this Eager Eyes piece.
So I thought I’d see if I could improve it by turning it in to a bar chart instead…
Gosh, that looks much better! It’s in descending order and so is easier to compare categories and to see the categories containing the most/least offences. It’s probably not the prettiest chart but it’s what I could muster up in Excel in 10 mins.
It’s really interesting data and it’s definitely a set I’d like to explore more – for example what would fall under “Other theft”? I’d also be interested to compare these stats to a time when an average 130,000 people weren’t visiting the area every day for four weeks (the amount of people visiting for the Olympics & Paralympics have to skew the figures right?). But that’s for another time. My goal was to make the Mail’s pie chart easier to read and to allow proper dissemination of the data and I think my solution is definitely a step in the right direction!
AdeSquared 
[…] the “worst way to convey information ever developed in the history of data visualization”. I found writers wanting to report poor uses of pie charts within the UK national media to “the data viz […]