Police Now – Practising Evidence Based Policing


As the Police Now 2015 participants prepare for the Police Now Summer Academy by studying Evidence Based Policing, we met with Inspector Ben Linton from the Society of Evidence Based Policing (www.sebp.police.uk) to learn from his extensive experience in this area:

Does lots of walking the beat mean that you know what works in policing? Do officers with a certain level of service somehow know what works in policing? Or put another way, if experience is not founded on a proper basis of what works, is it useful? If you went to your doctor with a bleeding mole, would you want the diagnosis to be based on scientific evidence of what works to treat bleeding moles, or would you be happy that the doctor had set up a surgery twenty years ago with no training in treating bleeding moles, but now considered himself very experienced at treating them?

Policing should be no less based on scientific evidence of what works than medicine.

Does walking the beat reduce crime? When you get to your wards, is general patrol an effective way of reducing crime.

No. It isn’t, but we still do it. That’s not to say that targeted patrol doesn’t work, but it has to be properly targeted at where crime historically concentrates – hotspot policing or “cops on dots” as it is sometimes called. The evidence? 31 well conducted randomised control trials that prove it works.

Is hotspot policing practical? Yes.

Take Police Sergeant Renee Mitchell from Sacramento, California who implemented hotspot policing on her patch, comparing the effect on the hotspots getting more patrol with some other hotspots she left alone. She found a 25% reduction in crime c/f control hotspots, calls for service down 7.7% c/f 10.9% increase in the control, and crime reduction amounting to a cost saving of $290k. In three months.

Hot Spot policing locations in Sacramento

Do we do things as police officers that actually increase harm?

Scared Straight was a widespread initiative in America whereby young offenders were taken to prisons and shouted at by the inmates in an attempt to scare them out of committing further crime. Good idea everyone thought, sounds intuitively like a good idea. Seven randomised control trials later and we know it increases offending. It actually causes the kids you do it to to offend more.

Are there things that work but which aren’t done by us?

Restorative Justice Conferences where the victim, offender and a trained facilitator talk about the offence in addition to any punishment mandated by the court are hugely effective. They reduce reoffending by 27%, reduce post traumatic stress symptoms in the victims by 40% and save £8 for every £1 spent. We knew this in 2003, but we still don’t do them.

Police Officers need to have the courage to find out what might actually work in their areas, based on good evidence, and do that which works, rather than that which is just accepted good practice. And don’t be bamboozled by bad statistics that tell you that a particular operation reduced crime when actually there was only a simple correlation between the operation and crime going down. The thing best correlated to reductions in US Highway fatalities over five years? Tons of limes imported from Mexico. Really, its an almost perfect correlation. Did crime go down because of the robbery operation, or because it rained? How are you sure? Remember, correlation does not equal causation.

So, how do we find out what works?

You can prove prevention, though everyone will tell you that you can’t. One way of doing it is the Randomised Control Trial. For instance, you could select the 100 highest crime bus stops in London and randomly selected 50 of them to get increased police patrol. When we did this, we found that calls for service from bus drivers went down by 52% where we did the patrol, compared to the control groups. With a control group you can quantify the preventative effect and prove that your tactic works. You should do stuff which has this level of evidence behind it, not just the say so of an experienced officer.

Policing must be increasingly based on proper evidence of what works, and that responsibility lies with all of us.