The trouble with averages - Anti-social behaviour

Last week saw the publication of at least two news stories on anti-social behaviour (ASB) in local news papers (The Wharf and East London Advertiser) based on a Tower Hamlets Labour Party press release (here) and even saw the creation of an infographic!

Jim Fitzpatrick used FOI requests and parliamentary questions to obtain the data. I on the other hand just used the freely available data published by the police nationally and didn't trouble anyone. I'm using data down at a ward/neighbourhood level, though more granular data is available.

The Wharf interpreted the data (ie press release) by stating that:

A total of 17,287 incidents were recorded by the police in the east London borough last year, putting it second, behind only Westminster City Council, home to the West End

The ELA went with:

Only Westminster had a higher rate of ASB than Tower Hamlets, with nightlife hotspots such as Soho accounting for many reports.

So far, so unoriginal. Looking at the data at a borough level is interesting, but hides massive levels of variation within each borough. Within Westminster, the level of ASB is significantly higher in areas around the West End than compared to say Bayswater, Belgravia, Marylebone and Regent's Park. If you were to strip out the figures for the West End, the ASB rate for Westminster would be 33 per cent lower.

Even in Tower Hamlets there are reasonably significant variations in ASB, with Bethnal Green South, Spitalfields, Weavers and Whitechapel having ASB rates that are markedly higher than other areas in the Borough, albeit to a less striking extent than in Westminster. These to my mind correlate broadly to key thoroughfares and areas of entertainment. However, I cannot conclude this without drilling down into the data, but the presence of high levels of ASB around Whitechapel Road, Commercial Road, Bethnal Green Road, Brick Lane and Spitalfields seems quite intuitive.

Contextualising by looking at a ward-by-ward level for both LBTH and Westminster, we can see that actually, ASB in the West End dwarfs anything in the East End. Other than being annoyed by drunks, I have never felt unsafe in Central London at night or in the day, but, ASB is clearly reported at a much greater rate than in Tower Hamlets, where I do sometimes feel uncomfortable, so we also need to consider the nature of ASB and how each individual instance affects people.

So what should we conclude? Jim Fitzpatrick is quoted in the press release as saying:

"These figures show there's a lot more work to do to ensure local people feel safe in their neighbourhood."

 From my perspective, the data doesn't show this. What it shows is that the level of reported ASB in LBTH is relatively high, but doesn't necessarily feed into making a statement about safety, or how perceptions of ASB varies across the borough. If there are key hotspots where ASB is focussed, individuals living near these hotspots may feel unsafe, but we cannot extrapolate this across a population of 200,000 people.

Both LBTH and Westminster have a similar number of reported incidents of ASB, yet Westminster has two and a bit times more reported crime as shown below. ASB may be annoying, and it may make people people unsafe, but we shouldn't overplay the issue when we can't understand what the data says, particularly when we think of the myriad offences that can be classed as ASB.

There used to be data on this issue, though alas the Government ended collection of National Indicators after 2010. What the last dataset to be published showed was that LBTH had the second highest perceived level of ASB, with 46 per cent of survey respondents having a strong perception of ASB, compared to 20 per cent in Westminster - yet both have identical levels of reported ASB, suggesting, that reported ASB is not the metric to measure people's perceived safety on.

One final note - from my previous research, we do actually know that crime decreased significantly over most wards in Tower Hamlets in the 11 years to 2012.