Ö÷²¥´óÐã

The Ö÷²¥´óÐã Office says today's launch of six-month "trials" in eight police forces in England and Wales will help them decide "what works best" when it comes to logging calls and responding to complaints about anti-social behaviour.

But this afternoon I learned that the official who might normally oversee the methodology of such research, the Ö÷²¥´óÐã Office's Chief Scientific Advisor Professor Bernard Silverman, was not consulted.

The government claims to believe in the principle of evidence-based policy, but with no money to spend and a philosophy of "not being prescriptive", gathering robust data from local areas becomes difficult.

that the "new approach to tackling ASB is seeing responsibility and control move from Whitehall to local agencies and neighbourhoods".

"Building on that principle, the trials are bottom-up, with each volunteer area deciding how to implement the five principles (see below). At the end of the trial the Ö÷²¥´óÐã Office will assess each area's approach and publish details about what worked best across the eight forces and what other areas should be looking to copy."

The five principles are these:
• creating an effective call-handling system where each individual has a log of complaints created from the very first call
• introducing risk assessment tools to quickly identify the most vulnerable victims
• installing off-the-shelf IT systems to share information on cases between agencies, removing the need for meetings
• agreeing a protocol across all local agencies setting out how they will manage cases
• engaging with the community to clearly set out the issues which are causing the most harm to individuals and neighbourhoods, and setting out how the police, other local agencies and the public can work together to address them.

When I asked the respected Cambridge statistician Professor Sheila Bird for her assessment of the trials, she expressed concern that different police forces will define the process in different ways, making accurate comparison impossible. "What constitutes a call? What constitutes an incident? Are there agreed definitions?"

One significant problem is that there is no agreed definition of anti-social behaviour itself. made the point that "anti-social behaviour (ASB) is a confusing term which has been variously applied to a wide spectrum of activity, from serious criminal violence and persistent ongoing intimidation and harassment at one end of the spectrum, to subjective feelings of unease caused by relatively minor and perhaps occasional environmental disturbances, such as litter, at the other."

"Policy should try to move away from subjective interpretations of what constitutes ASB (as enshrined in the 1998 Crime and Disorder Act) towards a greater prescription and definition."

There's nothing wrong with encouraging innovation and a range of approaches to problems, but statisticians will be wary of how "non-prescriptive" trials without formal evaluation might be used as the evidence-base for policy.

Professor Mike Hough, president of the British Society of Criminology and formerly deputy director of the Ö÷²¥´óÐã Office's Research and Planning Unit, warns that "if the informal approach to piloting turns out to be a low-cost substitute for formal evaluation, then that's obviously a cause for concern".

For the narrow and technical protocols under consideration in the ASB "trials" it probably doesn't matter hugely whether the evaluation is strictly scientific. Police forces will do their own assessment of how the pilots worked.

But if one of the consequences of localism is a non-prescriptive model for policy pilots, then it signals a move away from the principle of evidence-based government.

• Bookmark with:
• |
• |
• - What's this?