Fear and Risk

Just before Christmas I posted two blogs that referred to the Reassurance Gap (RG). They can be found here and here.

The RG is the difference between police recorded crime figures which were going down and what people perceived to be the case, which was crime was a bad as ever, in fact it may even be going up.

The Reassurance Policing Programme (RPP), started in earnest in 2004 was set up to address this problem. Since then the emphasis in government and policing policy, and academic literature has been on reducing the fear of crime. It seems to be accepted now without any real thought or discussion that the fear of crime is a bad thing that we should be getting rid of it.

Taking a step back though we can see that in every other sphere of life fear has a positive aspect. It is what keeps us out of dangerous situations, it what helps us make sensible decisions, its what keeps us alive.

One of the interesting aspects of the fear of crime is that it seems that the people that are apparently most at risk, young males (see my street robbery analysis posts) are the least fearful and those least at risk, the elderly, are most the fearful.

Risk has two aspects though, probability and consequence. This is where lifestyle and vulnerability come in. Young males to a large extent are victims because of their risky lifestyle which is a result of a feeling of a lack of vulnerability. Most take being a victim of crime in their stride as long it is not too serious. Elderly people on the other hand tend to cultivate a risk averse lifestyle due to their feeling of vulnerability because the consequences of being a victim of crime, even relatively minor, tends to be far more serious.

That is the reason why so called distraction burglaries (or burglary artifice to use the policing term) is such a nasty, despicable and serious crime. Its criminals target elderly people in their homes and rely on faulty memory and sight, and engendering a feeling of guilt in their victims that they have been stupid, to get away with it.

Crimestoppers have quite rightly made this type of crime a priority and published a list of the 10 most wanted. Have a look here you may be able to help.

Perception is in the eye of the beholder

Perception and knowledge is a curious thing. The top picture, when I first saw it and I found out that it was by an artist working in New York, I could see a 3D map of a cross road in that city with four skyscrapers looming out towards me at each corners. The Avenue at the crossing was obviously called Delaware.

The knowledge there is an iconic painting called Washington Crossing the Delaware by Emanuel Leutze hanging in the New York Metropolitan Museum made me see Frank Stella's picture differently. The composition of Leutze's picture is designed to draw your eye to the image of George Washington standing defiantly in the rowing boat on Christmas Day 1776 at a pivotal moment in the American War of Independence. Now Stella's image was all about drawing the eye to the point of the cross at the centre of his composition. Both images are about a single moment in time and a single person that defined the making of USA. To many New York's skyscrapers are a symbol of USA's power and greatness (that's what 9/11 /2001 was about) so perhaps my first thought was also correct in a way (excuse the pun).

Of course if you are Dan Brown you will also see a pyramid symbolising Washington's Mason membership!

If you want to know more about the paintings go here for Stella and here for Leutze.

What has all this got to do with the theme of this blog? Well the next few days I am planning to embark on the complex and confusing subject that is the fear of crime. Perception is at the centre of it. This is a reminder that the same image or event can be perceived differently by different people. Even the the same person can have different views when different knowledge is brought to mind.

Also once you are able to read the language of minimalist symbolism it is a far more direct, specific and powerful means of communication than realistic imagery. The contrast I am making is between a Google hybrid maps and grid squares, for instance.

Crime data like caviar please

One of the things that annoys me is the fact that crime data are normally presented divided by the resident population. I understand that this is meant to aid comparison of administrative units, such as boroughs, with different population sizes. But it only rarely makes sense because it seems to assume that the only people who are victims of crime or commit crime are those that reside within the boundaries. In London with its huge commuter, business, student, tourist, sporting and entertainment seeking transient population this is clearly not the case.

I tend to like my crime data how I like my caviar, raw and with good provenance. The new data store referred to in my previous post has raw as well as mucked about data. And the provenance is not always clear.

I present the three maps above more to make a point than to illustrate facts because I do not know how the numbers of alcohol attributed violent crimes are calculated. I assume they are based on police recorded crime but because the figures are not integers I assume some sort of sampling has taken place.

The point I wish to make is using the raw data is valid in my view so the top map gets a tick. Alcohol attributed violent crime must be hugely influenced by non-resident populations so in my view the second map is meaningless. The third map is a result of me putting together two datasets from the London Data Store. I am using the number of bar employees as a proxy indicator of the number of alcohol outlets there are in each borough because I think it is reasonable to assume that the alcohol attributed violence is connected with people who have frequented such places. Therefore I think my map is better for comparison purpose.

Conclusion: only muck around with crime data if it is based on a well thought out hypothesis and state what that is.

PS I don't eat caviar... but the point about crime data remains.

A quick map from the exciting new data source

I have been feeling a bit guilty about not having posted anything over the last few days. The reason is I have been trying to finish off a paper that has been hanging around a bit long and it required my full concentration.

What I have here is a map visualisation of an exciting new data source that Alex brought to my attention. It can be found here. I have not had a chance to have a look through it all but this one looked relevant to crime. So I quickly mapped it using ARCGIS.

Does snow influence crime?

As I write this I am looking out onto a winter wonderland. There is a thick carpet of snow over everything. Valiant attempts have been made to clear the snow off the road to allow vehicles up the hill but the snow is winning the battle.

I hear on the news that police in affected areas are trying to hire 4x4 vehicles and are only attending life and death type emergencies.

The question is does snow influence crime. Logically it must do, peoples' behaviour changes, they stay at home. It seems though that not all of London is equally affected. Fabian has reported that London this morning was a ghost town, not because it was white with snow but because there was hardly anyone around.

There is an interesting paper in the September 2009 issue of the academic journal Environment and Planning B entitled "The influence of weather on local geographical patterns of police calls for service." This did not specifically investigate the influence of snow but does show evidence of that high temperatures increasing violent crime and high temperatures and humidity affecting the geography of disorder.

The problem with this type of study is firstly obtaining detailed temporal and geographical details of the weather and secondly determining the extent of geographical and temporal influence of the weather. For instance, as we have seen, snow influences areas where it has not snowed, so comparing the part of London where it has snowed to where it has not snowed will give only a partial answer. Equally the affect of snow lasts till it melts rather than just when it falls.

It would interesting to do a brief analysis on these few days to see how crime/police incident patterns change. Hopefully this will not be considered pointless research like that conducted on penguins presumably in the snow!

This is just a quick post today to answer my question yesterday - "Are the locations of victim addresses random if population density is taken into account?" I have not bothered with legends because if you have been following my discussion (if not look back) a quick sentence or two will explain what is being represented.

By the way I did mention that I am experimenting with this form of representation and now realise that showing about the top twenty grid squares for each category and using only three shades seems to provide the simplest and therefore the least confusing map.

The red shades show the known home locations of suspects for street robberies in the London borough and the blue shades victims living in the borough who were victims of street robberies in the borough.

The two maps look very similar because only four of the squares a have both colours. This shows that suspects and victims though living close to each other live in distinctly different locations.

If victims and suspect home locations are random taking population density into account then it would be logical to expect a greater overlap between the two. The lack of overlap also suggests differences in geodemographic characteristics of the two groups.

Are the locations where victims of street robberies live random?

Returning to the analysis of street robberies in a London borough. I have established that street robberies occurred more in certain locations and times than others; that certain people are more likely to be victims and/or offenders than others. Even that times and locations and types of victims are linked. I have also shown that even though these elements of the crime commission process are non random the actual occurrence of crime at a specific location within a broad time frame (that eliminates to a large extent cyclic processes) is random.

Today I am going to show you some unsophisticated analysis of where the victims of street robberies live in relationship to where the crimes occur. The first thing to say is that about a third of all victims live outside the borough. The two thirds that live in the borough homes have been plotted again within the 250m by 250m grid squares. The results are shown above. Unsurprisingly there appears not to be the same degree of clustering as there is with the locations of where the crimes occur. Whether this is a random distribution if population density is taken into account is an interesting question.