Today saw the UK Police map service launched which brings local policing to the desktop of the population through a range of maps and simple data summaries. In terms of a step towards more open data it is certainly to be applauded but in terms of the mapping it adds yet another woeful example to the plague of poor online web maps.
Users type in a place or a unit postcode and get data on the location showing handy point symbols identifying locations of crimes disaggregated by broad types...or do they? No! The points are some sort of generic locator placed on each road segment that is used as the point to symbolise crimes for that road. So what does Mr and Mrs average do with the site?..of course, they type in their postcode and are delighted to find out that all crime happened some distance away. Unless, that is, their property is located near the geometric centre of the road in which case they are now horrified.
I tried this for my own road and found that a well known crime that occured during the census period (I know because the dawn raid was fairly obvious) was missing. So either the data is incomplete (which doesn't bode well) or it has been positioned elsewhere to 'protect privacy'. Err, surely the location of the new point is now inaccurate and while the original crime is not properly reported it is, worse, attributed to another location altogether.
This sort of utterly inept mapping is precisely what enrages cartographers. Representing data effectively to report geography properly and communicate reality is what we do. It is also what any basic GIS course or cartography course would teach. The annoyance is borne of the frustration of seeing a site that will now be viewed widely be completely misinterpreted because of erroneous mapping techniques. Not everyone is cognisant of cartographic theory but the least anyone should do when preparing a map is keep this fact in mind and design the map/map service so that it will be read and understood as accurately as possible by the uninitiated. The basic tenet of good design is to ensure you take account of the potential user and design accordingly to remove uncertainty and the potential for mis-communication.
The use of totals and a single point is poor to say the least. If the desire is to build in fuzziness to deal with the privacy issue then use rates or proportions; use areas instead of points; or use some form of road-based linear symbol that varies by thickness or colour. Instead, the UK Police have been lazy and taken a nice standard base map service and contrived to make up points where crimes didn;t exist, then sum the crimes locally to give a value to that location. Utterly meaningless in cartographic terms so pointless and dangerous to show it as such.
Unfortunately the new Police maps' use of points, when far better representational techniques are available, is such a fundamental error it beggars belief. It is indeed a crime in its own right. X does not mark the spot as any reader of Treasure Island knows. Now, the national map of crime perpetuates this myth by illustrating a nation of criminal locations that never actually existed.
Review: Mapmatics
-
The first intimation I had that maps involved mathematics was when I looked
up a map projection and came face to face with the equation that generated
it. ...
1 day ago
Enjoyed reading this, agree totally with the style of the site and generalisation of data, but with regard to the dawn raid issue isn't it more likely the raid led to no charges hence no crime?
ReplyDeleteQuite. There seems to be a complete lack of understanding on this project that it needed a robust cartographic approach. This is particularly damaging because, as you say, it's aimed at a general audience who won't be attuned to look for the flaws in the mapping but will simply take it as face value -- although it's heartening that so many people have noticed the gap between the map and their local knowledge and called foul.
ReplyDeleteMy own thoughts: police.uk official crime maps — there should be a law against it
JamesDavidSmith, you may very well be right about the dawn raid of course...though I doubt it given the nature of the crime and local reporting. Sorry - I know I'm skirting around but I don't really wish to go into the detail.
ReplyDeleteContext is so important in these maps.
ReplyDeleteI work for Surrey Police which has "the second most burgled road in England in December" according to the data in the map (as reported by Daily Mail).
The road marked has had no crimes in the last year (as far as I can tell), however the nearby university hall of residence
had one incident, in which ten rooms were broken into. This one incident is recorded as ten crimes...and the fuzzy plotting then places them in a completely different road. If you wanted to walk from the halls to the road named, it is about a mile (three sides of a square).
I welcome (in a personal capacity) open data about policing and crime - but it needs context, and it needs to be appropriately mapped!
> and it needs to be appropriately mapped!
ReplyDeleteHere, Here
Ken
ReplyDeleteGood post
Should we be conflating of cartography and GIS in this discussion?
Aggregating crime by street and then snapping to a pretty arbitrary point at the street centroid is just plain wrong, wrong, wrong. This is GI and Stats 101 stuff. I know people within the Police Service were very concerned about this but they were unable to get a change in methodology. If we are going to do this at street level and anonymise the data we need to apply crimes to road segments and then theme with some kind of population normalisation. Now that is not easy, it takes a good understanding of the data and techniques, old fashioned paleogeo. Once that analysis is done we then have the challenge of representing it visually so that it conveys info to the users, that's cartography?
In fairness to the guys who built this, the representation of the points and the interface is not bad. It's just a shame that the points are so misleading.