A while ago I stumbled upon this website with executions in the state of Texas. Then you start thinking about all that happened and other intimidating stuff. After that it took me a while – after bookmarking the source – when I found it again in one of my bookmark collections.
This started me to think, after also doing some text analysis with the Twitter Analytics App, that I could create some intriguing insights when loading this data into a dashboard. So I played around with it and created a sample app for myself to just see what would catch my attention.
Here’s the preview:
A few things popup:
- Most spoken words: Love, Family, You
- Interesting words: Mom, God, Innocent, Victims, Forgive, Apologize, Ready
The rest speaks for itself I think.
After I posted an update to Twitter, one of my co-workers – Murray Grigo-Mcmahon – posted some comments on designing data with this sensitivity, to which I completely agree. My challenge for you is to create, if you’re up for it, a new version of these insights with Murray’s comments. I will make sure your submissions will be showcased on this post.
If you wanna be able to have a more thorough look at this dashboard or want to participate in the challenge, you can download it here.
See u next time!
After the comments of a few highly respected Qlik fans I decided to add a new version that might be more aligned with the topic.
The thing that I wanted to know that isn’t visible from the screengrab, was the reasons behind the spikes and drops in the number of words spoken. A chart beneath it showing average last words per person would immediately show if spikes were due to a increase in death sentences or whether the amount each person had to (or was allowed to) say spiked in the same places.
An interesting challenge to make the visualisation reflect the gravity of the information being displayed. The website that the data comes from doesn’t really achieve that at all.
In my response to Patrick on twitter I mentioned the need for ‘design sensitivity’ in dealing with this subject matter. It’s a similar idea to Steve Dark’s comment about the visualisation reflecting the gravity of the information.
Within graphic design this has been often debated and tends to fall into 2 camps. One says treat the information with absolute neutrality and present it in the most legible and concise way, the other says reflect and draw out the humanity in the information and bring meaning to it. If you decide to take the first route then look to Stephen Few for the handling the visualisations and reduce any ornament or frivolity, and that includes colour, font, size and style variants. For instance the word cloud shown above. Visually it’s ‘playful’, due to the colours and type sizes but most of all due to the ‘jaunty’ text angles. This may not have been the intension but that’s the resulting visual experience. When you take the neutral route the design may well end up being a very plain or even ‘boring’ (in some peoples eyes) set of visualisations. And that’s when it gets tricky again, didn’t you want people to engage with the subject? Wasn’t the point for them to look at it and engage, explore and consider what these numbers really mean?’
The second design camp says, you can not be neutral! It says that whatever choices you make are the design decisions that shape the experience for others coming to this data, not just from a legibility perspective but also for how they experience the data and even how the meaning of the data is framed. Katherine McCoy said “the challenge is for the graphic designer to turn data into information and information into messages of meaning”. But the biggest challenge is to realise that we can’t avoid creating meaning, whether intensional or not.
In a way Patrick took the first route as he treated the data in the same way as he treated any other data (it’s based on his twitter analytics app), he took a neutral stance. But as you can see, sometimes other things conspire against that and end up ‘colouring’ the design unintentionally. Personally I sit in the second camp as I believe even when we strive to be neutral we never are. That means that it ends up that the designer has to take responsibility and be mindful off and sensitive to the information they are working with. It becomes about being very conscious of how your design choices will effect the people who come to your design.
For me this challenge is an extremely difficult one. One that raises many issues beyond the data, that I can’t avoid bringing to this work. My biggest challenge is deciding whether this is a data set I feel happy ‘playing’ with at all.
This isn’t intended to be a dig at Patrick who’s work and thinking I admire. It more that it’s the perfect example of this type of challenge and it’s an issue we rarely speak about in regards to data visualisation. All too often we are simply caught in the data visualisation v the infographic debate, but I feel it’s much more sophisticated than that.