The Home Office’s recent £300K attempt to make crime data easily accessible through their new crime mapping service has failed rather spectacularly.
But as an example of how not to go about data visualisation it can actually prove quite useful.
Back in 2008 the Labour government promised to produce easy to use crime maps of the entire country to provide all residents with greater insight of their area within a year.
On the 1 February 2011 the Home Office and Police.uk have produced these crime maps but have been incredibly conservative over the level of detail given due to fears of privacy invasion coming from the statistics they give.
According to the Information Commissioner’s Office while there is “no general consensus as to the degree of granularity that strikes the right balance between social transparency in respect of criminality and the privacy of those associated with a crime” we should for now err on the side of caution.
The Home office has decided to muddy the information available and only give crime statistics per street (or equally large vague area) and define crimes only as one of six blanket categories (including the especially unhelpful ‘Other Crimes’).
But by being rather conservative with the data used the Home Office has shot itself in the foot.
The maps could have potentially led to greater transparency – allowing all manner of regular citizen to panic about the level of muggers, rapist and paedophiles (Oh my!)
What we are left with is some large, pretty and oh-so-nearly useful map that doesn’t allow anyone to do anything of use – such as investigate problems for themselves.
Nothing like people seeing “what’s happening with crime in their area, not just on their street but in their neighbourhood,” that Theresa May had promised.
The Home Office’s rather opaque way of defining areas has also thrown up a few secondary problems – Britain’s most crime ridden street, home to over 150 crimes in December, had in reality only seen three crimes committed.
There are many other problems with the map (as succinctly summed up by Conrad Quality-Harper) including one or two technical issues – but the key problem is the clarity of information.
We at The Data Day cannot help but think simply releasing the data itself would have been far more productive.
While this experiment in data visualisation shows the Government is moving in the right direction for now it simply acts as an example of what not to do.
The Data Day 
As with so many data releases, the impression is that the government wants to tell you as little as possible – certainly nothing useful – while appearing to tell you everything. Or am I just being cynical?
What a good looking point there Alice.
I’d agree that there is definitely something a bit sneaky about what they’re doing, but If you look at the ICO’s report on Crime Mapping ( http://www.dpr.gov.uk/~/media/documents/library/Data_Protection/Detailed_specialist_guides/crime_mapping_advice.ashx ) they point out their reasons for it.
Say if you had a family on your road who had a bad kid and you were able to find out more specific information about the locations of crimes on your street then there’s a pretty good chance you could put two and two together and figure out exactly what this kid had done even if he was underage.
When publishing large sets of data there’s always a very real risk of jigsaw identification of confidential information. Keeping on the right side of privacy legislation is one of the most difficult aspects of dealing with Data Journalism (look out fort my next blog post).
In this case though claims for privacy are somewhat negligible, If you’ve committed a crime you forfeit your right to anonymity. These maps were promised to give us a greater understanding of what’s happening in the street around us but in reality they’ve missed this target by some way.
They do however look very nice.
Once again, a very sharply dressed point there Alice.