Mapping Crime in Oxford Over Time

Aug 12, 2009

Mentorn Media and Cimex Media, on behalf of BBC, explore crime patterns in Oxford over time. In a map, that I am happy to see is not a Google mashup, select different kinds of crime (e.g. violent crime, burglary & theft), or if you live in the area, compare different neighborhoods by postcode. The interactive also provides three animations for a week in crime – street violence, street robbery, and rowdy behavior – complemented by narration and explanation.

One thing I’m not so sure about is the color scale. I think I would have gone with a yellow to red progression and left out the green since green usually means something positive. I’m also not sure what ‘high’ and ‘low’ levels of crime actually means in numbers. What do you think?

[Thanks, Jack]


  • I’ve gotten maps like this from my local police force. The problem is that they tend to actually be maps of population density. With a comparable crime rate a higher density area – with apartment buildings – always looks like a it has a greater crime problem.

  • The trouble is there is so much crime there that the whole map is covered. Lol! Must remember that when I go there next.

  • I don’t know how heat maps are generally constructed, but suspect it is based on kernel density techniques. This technique would require geocoding all the crimes, and choosing a bandwidth that would determine how many of those points would influence the resulting density value. The bandwidth can really toy with how the overall density surface looks. As has been explained to me, data like this would probably benefit more from a variable bandwidth technique which might help with the underlying population density (as mentioned by srboisvert).

    This book explains it better than I could:

  • As with so much of the current NeoGeography we are seeing these days this is an example of people who don’t know what they are doing not bothering to read the literature and use existing techniques that the PaleoGeographers solved in the early 1990’s.

    Check out for example.


  • Thanks for the interest in the crime map – I led the project at Cimex. Re: “this is an example of people who don’t know what they are doing not bothering to read the literature”:

    We worked closely with Spencer Chainey on this project (Chapter 5 of the book you recommend – Director of Geographical Information Science at the Jill Dando Institute of Crime Science, University College London) and the various local services in Oxford. All our data processing and map based presentation techniques (using geocoded source data) were approved by University College. The important thing here was to make the content accessible to broader public audience and to compliment the stories on the associated TV show and the feedback we received form user testing was this did succeed.

  • So did you allow for the background population or not? unless you link to your methods it looks like a pretty coloured map with no scientific value like all the other crime maps out there. Have a look at chapter 2 (the one I wrote with Stan Openshaw) and then decide if your map mean anything.


Think Like a Statistician – Without the Math

I call myself a statistician, because, well, I’m a statistics graduate student. However, the most important things I’ve learned are less formal, but have proven extremely useful when working/playing with data.

Years You Have Left to Live, Probably

The individual data points of life are much less predictable than the average. Here’s a simulation that shows you how much time is left on the clock.

Divorce and Occupation

Some jobs tend towards higher divorce rates. Some towards lower. Salary also probably plays a role.

Graphical perception – learn the fundamentals first

Before you dive into the advanced stuff – like just about everything in your life – you have to learn the fundamentals before you know when you can break the rules.