It's worth repeating what the @ukhomeoffice research on #stopsearch looked at, and what it found. @HackneyAbbott https://www.gov.uk/government/publications/do-initiatives-involving-substantial-increases-in-stop-and-search-reduce-crime-assessing-the-impact-of-operation-blunt-2 #stopandsearch

1/ https://twitter.com/HackneyAbbott/status/1105508168979959808

First, it was looking at whether any impact on 11 crime rates could be attributed to the impact of the *additional volumes* of #stopsearch *weapons searches* resulting fr Operation Blunt 2, at a borough level. Ten 'Tier 1' boroughs received most of the additional SS volumes.

2/

It did *not* assess:
(1) The impact of overall #stopsearch volumes (just those involving weapons searches)
(2) Whether #stopsearch has an impact on crime rates at all - ie comparing no SS with an amount of SS

3/

Nor did it assess:
(3) The relationship between volumes of #stopsearch & crime - ie how much impact on crime do you get for x more SSs - other than in relation to the Operation Blunt 2 'additionality'.

4/

FWIW, I'd suggest we might assume the relationship to be *curvilinear* - ie the strongest effect is from 'none to some', & at some point you get into diminishing returns (or even negative returns, eg if legitimacy is undermined & public co-opn suffers as a result) #stopsearch

5/

The Home Off evaluation found: "no statistically significant crime-reducing effect from the large increase in weapons searches during the course of Op BLUNT 2. This suggests that the greater use of weapons searches was not effective at the borough level for reducing crime."

6/

Further: "ambulance call-outs [for weapons-related injuries] actually fell faster in those boroughs that had smaller increases in weapons searches"

7/

Main concl: "Overall, analysis shows that there was no discernible crime-reducing effects from a large surge in #stopsearch activity at the borough level during the operation. However, it does not necessarily follow that stop and search activity does not reduce crime."

8/

It's worth saying that overall volumes of #stopsearch have fallen around 3/4 since the days of Op Blunt 2, so we are not in the same territory of 'additionality', which is to say that Blunt 2 was built on top of much higher pre-existing rates of SS. https://twitter.com/gmhales/status/970956520946061312

9/

It should also be said that the Home Office research isn't the only project to have looked at the relationship between #stopsearch and crime in London - there is also this by @MatteoTiratelli @pkquinton and @ben1971b looking at 10yrs of data http://whatworks.college.police.uk/Research/Documents/SS_and_crime_report.pdf

10/

They found "higher rates of #stopsearch were occasionally followed by very slightly lower rates of crime. The inconsistent nature and low strength of these associations, however, provide only limited evidence of stop and search having had a meaningful deterrent effect"

11/

They further found that "The fact that any associations with broad categories of crime at a borough level were identified suggests that stop and search might have more of an impact on more specific crime types and at a more local level." #stopsearch

12/

At the heart of the debate about #stopsearch is a paradox: how do we reconcile the fact #stopsearch clearly does detect offences, including knives being illegally possessed in public, when there is almost no evidence that large increases in SS reduce (borough-level) crime?

13/

I have previously framed this as tactical wins but strategic failure. I'm not sure there are any good answers.

We do know, from 2015/16 MPS data, that on avg 300 #stopsearch-es resulted in only 2 arrests for points/blades & 5 for offensive weapons http://www.police-foundation.org.uk/2016/12/stop-and-search-and-knife-crime-revisited-a-small-piece-of-the-bigger-picture/

14/

Maybe the harms caused by #stopsearch, eg to police legitimacy, counteract any crime reduction effects? Maybe the impact of #stopsearch is more local than research has been able to detect? Maybe most of the knives recovered would never be used in crime?

15/

My personal take is #stopsearch is an important power, but probably a poor strategy, and the real goal for policing is good intelligence allowing officers to deploy their powers in the most specific & targeted way possible.

16/

The temptation with the £100m of funding that has been announced must be to splash it on overtime and #stopsearch. While part of the mix, I hope policing can be more imaginative, including focusing on police-community relationships and improving intelligence.

17/ ends for now

To add to this thread, this very useful summary and discussion of the evidence about the links between #stopsearch and crime by @ben1971b and @MatteoTiratelli for @CrimeandJustice (which I've only just seen) https://www.crimeandjustice.org.uk/publications/does-stop-and-search-reduce-crime

18/