Last year, Western Australia became the first jurisdiction in the country to establish an evidence-based policing unit to test the assumptions behind traditional police operations and bring in a more scientific approach.
One reason is a familiar demand to achieve better outcomes with limited resources. WA is experiencing a population boom but can’t rely on a continued mining boom to help support it. But equally, the EBP unit came about because a group of senior officers have seen the value of augmenting their years of experience and knowledge with empirical evidence.
In 2013, a group of senior police formed the Australia and New Zealand Society for Evidence-Based Policing in Brisbane, modelled on and formally linked to a similar academic group in the United Kingdom. West Midlands Police has been a particularly enthusiastic British adopter of the EBP doctrine, and has built up some ties to WA Police through the growing worldwide movement.
Evidence-based policy often seems like a no-brainer to people with an appreciation of rigorous academic research, but it’s generally not what citizens demand of their elected representatives. At the same time, the nature of policing means there is a lot of faith put in the way things have always been done, and a reluctance to question its value.
Experimental trials are perhaps seeing increasing use as a tool in policy and program design, and credible academic research has always played a big role in the confidential advice public servants provide to government.
But voters still prefer ministers who project certainty and self-confidence, as though their political ideology held the answer to any question. This is particularly apparent in policing or anything else to do with public safety.
Highly visible solutions like putting more cops on random patrol where the public can see them or public CCTV cameras are often found to be overrated, when evaluated. Nonetheless they can still become useful (and expensive) props for political leaders at all levels of government, who stand to gain support from appearing tough on crime.
As Western Australian deputy police commissioner Stephen Brown wryly comments, everyone supports a tough response to crime until it’s their son or daughter in the lock-up.
He also points out the founder of England’s first professional police force, Sir Robert Peel, laid down some important guiding principles in the 1820s that still apply. Peel’s nine points promote prevention over repression of crime and disorder, the idea of “policing by consent” and conclude with the wise counsel:
“To recognise always that the test of police efficiency is the absence of crime and disorder, and not the visible evidence of police action in dealing with them.”
Brown sits on the board of the ANZSEBP and the WA Police EBP unit was set up after he returned from Cambridge University, where he studied under the pioneering professor who coined the term, Lawrence Sherman. Since then, several more WA officers have taken the same Master of Applied Criminology course designed for police executives, as have others from interstate. In Queensland, a scholarship was established last February to send one senior officer to Sherman’s course each year.
“I found myself in March 2014 sitting in the halls of Cambridge University hearing about things that were working to drive down crime and antisocial behaviour, from around the world, not only the UK, that I’d never heard of before,” deputy commissioner Brown told The Mandarin. “And I just found that absolutely confronting.”
Out with the old
WA’s chief EBP proponent points out a lot of police work is based on “some supposed knowledge, some experience over time, but not real evidence — not evidence that’s been tested to actually prove specifically whether the interventions that we’re applying on the ground actually work or not”.
The traditional approach can be summed up with three Rs: rapid response, random patrol and reactive investigations.“What better gift can senior police give to their officers than the knowledge that what they are doing is actually making a difference?”
People want the police to come quickly when they call, but most of the time this makes little difference as most crimes are unfortunately reported a while after they’ve been committed. The research evidence shows “getting there a minute or two earlier” wouldn’t help most of the time, says Brown, and that most people take a surprisingly long time to report crimes even after they become aware of them. Nonetheless there are KPIs based on the speed of response, reflecting public expectations.
Secondly, reactive investigations unfortunately can’t solve all the cases like on television. Around the world the best average clearance rates for burglaries are around 14%, according to Brown, who spent over a decade as a detective and was senior investigator for the 2001 Kennedy Royal Commission into police corruption.
That doesn’t mean they should stop doing investigations, he says, but it does suggest other most cost-effective ways of lowering crime rates are needed too.
Finally, despite the reassurance it might give the good burghers of our towns and cities to see a police car every now and then, the deputy commissioner says random patrolling “doesn’t actually deter much crime” on the whole.
“Now I know there’s a lot of public debate around that, but what we say in evidence-based policing is that the patrol work itself needs to be focused on crime hotspots — and pretty much nowhere else,” Brown explains.
Sherman came to prominence when he popularised this kind of thinking in the 1980s by showing that crime is surprisingly predictable as it clusters around certain nodes, geographical or otherwise.“When you can display to people that some things that are somewhat counter-intuitive do work, then the light bulb comes on.”
Brown says a common finding across the world, particularly in the United States, was that about 50% of calls for police assistance come from as few as 5% of addresses. Initial research in WA suggests a “much higher” spread than that, but his point is that all police can work smarter by trying to figure out what is going on at certain addresses with repeated calls for assistance.
“Is it an offender? Is it a repeat victim? Is it something you can design out, something in the physical environment you can change to reduce crime?”
Similarly, he says the bulk of complaints about police typically refer to only about 1% of officers, and the same general patterns are found in most data related to policing all over the world. “We know that some women in our communities, as an example, are repeatedly subject to domestic violence, so that’s where the science says you should target your resource,” said Brown.
In with the new
In 2013, Sherman came up with a triple-T model to replace traditional approaches like the three Rs: targeting finite resources where they will have most effect, testing to find out what works best, and tracking outcomes over time.
The small WA police unit has about 21 staff led by a superintendent, and the deputy commissioner says several senior officers are also running various experiments in EBP themselves.
“We’ve got analysts in there, researchers, geospatial people who are looking at our crime data and helping us to evaluate the different things that we’re experimenting on at the moment, and we’re just about to employ a director of criminology,” he explained.
The police hope to attract a leading criminologist by offering a competitive annual salary to maintain academic rigour of the EBP unit’s work and provide expert internal advice to guide the evolving project. According to Brown, “What we’ve seen over the last, almost two years now … is the language in the agency is changing and people, officers in particular, are starting to get a better understanding about what evidence-based work is really about.”
Having evidence about what works and what doesn’t, and tracking it over time, could also help police and government sell new and more effective policies to sceptical officers, members of the public and the media. Brown believes a lot of police activities that sound like a good idea could actually be wasting a lot of money, or even doing more harm than good, but he doesn’t expect to convert everyone in the community to EBP overnight.
“You can’t win all of these arguments every day but when you can display to people that some things that are somewhat counter-intuitive do work, then the light bulb comes on,” said Brown. “And whilst you’ve got permission from a government or a commissioner of police to experiment, you’d be mad not to.”
On the other hand, he can’t think of much worse than supposed solutions that don’t really work. EBP is also about evaluating existing practices, and Brown hopes it can improve morale and motivation by giving officers more assurance that their work is having a positive effect:
“What better gift can senior police give to their officers than the knowledge that what they are doing is actually making a difference?”
Part of the EBP unit’s job is to produce a learning, training and capability framework to embed the new theory and practice into the WA Police at all levels, and assist frontline officers all over the state in putting the triple-T approach into action in local situations.
Three examples of EBP in Western Australia
Operation Communique combines EBP with behavioural economics, another field of academic research that advocates greater use of experimental trials in the public service.
Letters and text messages are sent to a portion of 14,000 repeat traffic offenders, identified from infringement data, whose likelihood of causing or suffering road trauma is elevated 14 times. A control group is left alone while other experimental groups receive two types of messages. One is a typically blunt police missive (of the sort that hardened road warriors laugh at, according to the local tabloid press) and the other a “very well crafted emotive letter” designed with advice from psychologists and lawyers, according to the deputy commissioner.
The results will take a while to come in but at a cost of only $10,000, Brown says even a small decrease in road trauma, as little as 2%, would mean it’s probably worth continuing. A negative result is also valuable in such trials.
The Turning Point program has been imported from the UK’s West Midlands Police and links with principles from another field of criminology, restorative justice, and the related diversionary programs that seek to give young first-time offenders a second chance to make amends and avoid going into the system.
WA’s first early results are expected in 2017, but Brown has some expectations based on the UK program, which has demonstrated ways to significantly improve “victim satisfaction” after a few years. Recidivism dropped by only around 5%, but there was a much more significant reduction in violent reoffending.
Normally, victims have little to do with criminal cases in court, unless there’s a not-guilty plea, in which case they’re a basic witness. But in restorative justice, they get involved, explain the impact of the crime to the offender and maybe get some kind of apology. Research by academics like the Australian National University’s John Braithwaite suggests efforts to reintegrate offenders into the community do more to reduce crime than the tough approaches that voters prefer, but many police believe “arrest equals success” and are often disappointed when offenders seem to get off lightly.
“Secondly, and importantly for policing into the future, the Turning Point program as opposed to arrest, prosecution and going through to the criminal justice system, runs at a third of the cost, typically,” explained Brown.
Operation Crackdown in early 2015 took the classic EBP approach of targeting crime hotspots — in this case, entertainment precincts — and compared the effect of a heavy police presence every weekend to less regular crackdowns.
Having more cops on the beat outside the nightclubs every week resulted in the biggest drop in crime. “But you know what? The one-week-on, one-week-off and two-weeks-on, two-weeks-off models were very close to the same sort of reductions in crime, but at about half the cost,” said Brown.
“So that’s a massive learning for police and something which I think needs replicating around the world, to see how those sort of numbers pan out in other entertainment precincts.”
He adds that WA Police won’t follow such obvious patterns when they implement the findings, to make it hard for people to guess when they will be out in force in a particular area.