Over the year I’ve been to three so-called ‘hack days’, one of which me and my team accidentally won (Hack the Government). The last of which my team just as accidentally
lost didn’t win (Girl Geek Hack Day). The one in between however was the most intriguing, perhaps because it wasn’t a competition at all.
Held in Whitehall on a blustery Autumn day, it was run by a bunch of people called Datakind, and some sponsors who I’ve already forgotten the name of (sorry). Datakind used to be called Data Without Borders but changed their name a few years ago – I’m not sure why – and this was their first British event. They’d invited stats bods from around the country to come help some charities get the most out of their data, and I thought I’d see whether I could help, despite not being a stats bod.
You can read all about it in the Guardian. Pay special attention to the bits about the girl who arrived early and then left to go clubbing, because that was me… ahem.
My first, bleary eyed, work-fatigued impression was that they were Very Definitely American. They were so damn cheerful and sincere it was infectious. Even to a cynical and occasionally slightly suspicious Brit like me. At this point (inbetween pastries) I got talking to some people who actually knew about statistical analysis, and got a little downhearted – I had nothing like their expertise! I’m just some oddball information designer in need of a bit of practice.
I plumped for Place2Be as my project. They were all very worthy but this one had the most potential for impact that I saw – the data was all about treatments for kids with behavioural problems and our mission was to find out what worked, and what didn’t. There was a lot of Heavy Discussion about data cleansing and appropriate formats and software which was way beyond me, so I did some extremely rudimentary pivot tabling to see if I could generate some questions we might want to answer – and came up with what I thought were some intriguing curiosities. Did girls really do better in the programme if they had single dads? etc, etc.
Most of which turned out to be nonsense, because I simply don’t know enough about statistics. But I felt helpful. Sometimes.
In the end it turned out we had the Wrong Sort of Data anyway – which is much like the Wrong Sort of Snow and about as helpful, so our findings didn’t end up being spectacular or transformative.
But it piqued my curiousity, so now I’ve signed up for a Coursera data … erm… course. Lets hope it helps.