Crime mapping

The 1st of February saw the release of a new website Police.UK, the latest website that allows the general public access to data that they have never really had free access to. The site was released early on the 1st and was given second top billing on the BBC News home page which was great publicity.  I tried to get on the site at about 0730 but it was already having problems, in fact it is reported to have had 5 million hits an hour during its first morning.  Incredible traffic.

Quality Mapping?

But what of the maps?  Well, as the dust settles it appears that there is growing criticism over the way that the data has been represented and this has raised fears that maps may mislead users. Most of the critisism centres around the use and placement of points on the map which represent crimes.  Police.UK explain that the position of crimes have been approximated to protect victims of crime.  This seems to be a very good idea but if the position has been approximated why continue to use a point to represent the crime?

By retaining a “point” after it has been approximated the crime may well be wrongly associated with another location on the map.  Perhaps a better method for representing the data, while ensuring that positions are approximated, would be to use Postcode Districts. This would then show what crimes are occurring in an area and the area would be small enough to see spatial patterns across towns and cities while being large enough to protect the victims of crime.

Complete Datasets?

Another criticism has been that some users have not been able to find crimes that they know have occurred. What has happened to these crimes?  Were they incorrectly reported or has the dataset been filtered before it has been published?  It is clear that some police forces have not uploaded their data yet, Scotland’s forces are absent and it is not clear if they will be added.

A good start but could do better?

Police.UK is a big step towards making data available to the public in a format that they can understand.  Credit should be given to the team that brought everything together from all the police forces, this would not have been a simple task.  However, when making data available in a useable format, such as maps, it is vital that the map doesn’t mis-represent the data.  The fear with the maps on Police.UK is that they could easily be misinterpreted.  Given the nature of the data, this could result in some serious problems.

Jonathan Raper summarised the problems with site as:

  • Locations used are approximate (so that “top 10 crime streets” may actually be places which have had no crime).
  • Data for streets with fewer than 12 postal addresses is apparently not recorded “to protect privacy”.
  • Some data such as sexual offences and murder is removed – even though it would be easy to discover and locate from other police reports.
  • Data covers reported crimes rather than convictions, so some of it may turn out not to be crime.
  • The levels of policing are not provided, so that two areas with the “same” crime levels may in fact have “radically different” experiences of crime and policing.

If you want to read about the criticism levelled at Police.UK, please read the following blogs:

Five reasons to be cautious about street level crime data by Jonathon Raper

UK Police maps: X does not mark the spot by Kenneth Field official crime maps — there should be a law against it. by Andy Powell

About Addy Pope

Addy is a member of the GeoData team at EDINA and work on services such as GoGeo, ShareGeo and the FieldtripGB app. Addy has over 10 years experience as a geospatial analyst. Addy tweets as @go_geo
This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *