The Birmingham Civic Dashboard has been live for a little less than two months now. We received quite a bit of publicity when we released it, and
in the first few weeks had quite a few people commenting on it, using the Disqus commenting system on the site.
We had an early blog post from Dave Harte on the Bournville Village blog. This is gratifying, although I should also probably declare that Dave worked in Digital Birmingham on our open data stuff before returning to be course director at BCU.
So, I’ve been wondering why we haven’t seen more use of the Civic Dashboard by local bloggers. When myself and Matt, from Mudlark, were first putting the bid together for Nesta we thought that they would be the most likely people to talk about it and use it.
Looking at the site I realised that I didn’t want to just grab a screenshot or image from any of the visualisations that we provide on the site. I was much more interested in what a month’s worth of data about our local area might tell me.
So, I went to the page for Hall Green constituency and I downloaded a csv of the data for each day. Once I’d done that I merged that data all into one spreadsheet. Now I had something that I could start playing with.
I’ve used Tableau a bit before. It’s a piece of data visualisation software that has a free, public edition and so I ran that, watched the instruction video, and imported the data. I watched the instruction video again and started playing around with the data and separating it out so that I could look at the different wards within Hall Green separately.
Once I’d got a line graph that I was happy with I published the work I”d done to the web and downloaded the graph as a graphics file that I could use in the post. this is it below.The obvious thing to note there is that there are a significantly greater number of requests from Sparkbrook ward than the others. I thought this was because compared with other wards in the constituency, Sparkbrook is not so well off. So, I filtered the data so that it only showed requests related to benefit claims and social housing. Here is the graph for that.
And so we can see that the difference in the number of requests is explained by the benefit related calls.
The description above doesn’t convey a full sense of how many times I watched the instruction video for Tableau, nor of the frustrations I had when trying to create the graphs and put them online. Many thanks are due toCaroline Beavon and Neil Houston for their kind help.
I wonder how many other people would want to create something different to the visualisations we provide. If they do, I can see that the process of downloading the data and then manipulating it in a separate piece of software is going to put some of them off.
The use of constituencies and wards wasn’t hugely helpful for my purposes. Balsall Heath isn’t a ward on it’s own; it sits within Sparkbrook ward. It has similar characteristics and so I’ve fudged things a bit by talking about Sparkbrook as though it has the same qualities and characteristics as Balsall Heath.
Hyperlocal blogs don’t necessarily exist to serve a whole ward. At our recent Brewcamp event Nicky Getgood told us how difficult she found it talking to the council. Part of that discussion centred on how the residents of Digbeth, part of Nechells ward, often feel they aren’t considered when plans are made about that area.
Nicky is coming to our event about the Civic Dashboard tonight, where we will be asking people what they find useful about the dashboard and how it might be improved.. I’ll be interested to hear if she thinks the dashboard tells her anything about Digbeth when she can only look at it in the context of the wider Nechells ward.
There are still a few tickets left for this evening for anybody who is interested. Both myself and Matt Watkins, from the developers Mudlark, will be there. We’re at Fazeley Studios between 4 and 7pm and there will be a buffet.