Neighbourhood watch schemes aim to reduce crime by directly involving the community in activities that promote safety or assist with the detection of crime. They actively seek the greater involvement of local people in crime prevention activities. The first schemes were seen in the US in the late 1960s and they are now very common in both the US and the UK. They are known under a variety of names including home watch, block watch and community watch (Review 1).
This narrative is primarily based on one systematic review covering 18 studies.
Overall, the evidence suggests that neighbourhood watch can reduce crime. There are however some important considerations, such as variation in practices across schemes, and large differences between countries in terms of observed reductions.
The review estimated that overall, for every 100 crimes, an average of 26 crimes were prevented with Neighbourhood Watch (based on 18 studies). There was no evidence of a backfire effect (where crime increases) across the evaluations reviewed.
The review was sufficiently systematic that many forms of bias that could influence the study conclusions can be ruled out.
This evidence is taken from a systematic review covering 18 studies. The review demonstrated high quality in terms of having a transparent and well-designed search strategy, featured a valid statistical analysis and considered the validity of the way outcomes are measured and/or combined. The review did not quantify an overall effect for unanticipated outcomes such as displacement caused by the intervention and, whilst it considered publication bias, it did not consider bias introduced by coder subjectivity or statistical outliers.
The authors note that:
"it is … difficult to determine from current research how neighbourhood watch works". However, they go on to suggest that Neighbourhood Watch might reduce crime by:
(a) Deterring burglary offenders through increasing their awareness of a great propensity of residents to look for and report suspicious activity; (b) reducing perceived opportunity via increasing signs of occupancy in vacant homes (moving bins, removing newspapers); (c) enhancing community cohesion and increasing the effectiveness of informal social control and; (d) facilitating detection via an increased flow of intelligence between public and police.
The authors raise the possibility that reductions might be a result of programme components other than the ‘watch’ element (e.g. the effect of property marking on theft of stolen goods; the effect of target hardening as a result of security surveys). Since Neighbourhood Watch is generally implemented together with these other components, it is difficult to rule this out. In fact, the watch element is often implemented with both property marking and security surveys - collectively referred to as the big three. The schemes reviewed below, unless stated otherwise, include those with such combinations.
None of these potential mechanisms are empirically tested and the authors note that this should be done.
There is evidence that Neighbourhood Watch effectiveness varies considerably by context.
In the reviewed studies for every 100 crimes:
An average of 47 crimes were prevented for schemes in the USA and Canada (based on 4 studies)An average of 15 crimes were prevented for schemes in the UK (based on 14 studies)
An average of 47 crimes were prevented for schemes in the USA and Canada (based on 4 studies)
An average of 15 crimes were prevented for schemes in the UK (based on 14 studies)
Evaluations of neighbourhood watch conducted in the US and Canada were significantly more likely to show a reduction in crime than studies conducted in the UK.
The authors tested for differences in effect by size of scheme. They reasoned that large schemes (7 studies) might be more effective because more neighbours would look out for suspicious behaviour; or alternatively that smaller schemes (11 studies) might increase the interactions between neighbours who knew each other. There were no significant differences in the effects of larger and smaller schemes.
The reviewers found no significant differences in the outcomes of schemes implemented earlier (up to 1988; 11 studies) compared with later schemes (1989 onwards; 7 studies). There was a large drop-off in frequency of evaluations in the latter period.
The authors note the very large variation in size of the schemes. For example, the 'cocooning' schemes in Rochdale, UK, involved 3 households whereas one large scale programme in the USA covered 30,000 residents. Such contrasting schemes are likely to be very different in nature.
There is very little information on the implementation of Neighbourhood Watch.
A notable complication in determining effectiveness is that the 'watch' component of Neighbourhood Watch is often implemented along with other-elements - most commonly property marking and/or security surveys. Interactions between these elements are not considered in detail. For example, it might be that a scheme was focused on property marking and had Neighbourhood Watch as an 'add-on', or the opposite could be the case. Further, those where all three were implemented might differ from those with different combinations of two of them.
The analysis undertaken demonstrated that the mean difference in effect was not significant for those with property marking and security surveys (10 studies) and those without (8 studies). Hence, the authors concluded that the presence or absence of 'extras' did not affect the outcome. This counter-intuitive conclusion was discussed with one of the authors of the review and the revised conclusion is that there is some evidence that NW plus might be effective in reducing crime but no evidence that NW alone is. In this case, there is 'added value' to implementing the fuller versions of the scheme.
Schemes differed as to whether they were police or public initiated. Some schemes were initiated by police in areas where it was unlikely that the public would make a request. There is also variation in terms of who runs the scheme - in general, this is likely to be a block/street captain who reports to a block/area co-ordinator. The latter then liaises with the police. There is little available information on number/ type of meetings or whether meetings involve the general public or just the organisers.
There is no monetary information on the costs of Neighbourhood Watch. In terms of coverage, the authors note the large-scale implementation of this type of scheme in both the UK and the US. The 2000 sweep of the British Crime Survey estimated that over a quarter (27%) of all households (approximately six million households) in England and Wales were members of a neighbourhood watch scheme (Sims, 2001). The 2000 National Crime Prevention Survey (US National Crime Prevention Council, 2001) estimated that 41% of the American population lived in communities covered by Neighbourhood Watch. These calculations do not account for differences in levels of engagement or activity when signed-up to the scheme - many households are inactive members.
In terms of funding, the authors note that this is nearly always jointly supported by the police and fund-raising activities of scheme members (such as voluntary contributions or raffles). The balance of the two contributions varies considerably; sometimes the police provide just an information pack, other times they offer premises for meetings or support newsletter production.
It is unclear how Neighbourhood Watch schemes achieve reductions in crime. Specifically, it is unclear what the independent contribution of the 'watch' part is over and above activity that can be categorised as publicity, property marking, or target hardening. The best advice at present is to include the 'big three' (neighbourhood watch plus property marking and security surveys) in any planned NW scheme.
This review has highlighted some gaps in the research base – specifically more is needed to test directly the mechanisms underlying Neighbourhood Watch – i.e. how does it work. This might assist in isolating the elements that are particularly effective.
Displacement of crime or diffusion of benefits to neighbouring areas are possibilities and should be considered.
There is evidence that Neighbourhood Watch, as typically defined and implemented, modestly reduces crime overall. It appears that it is more effective in the USA and Canada than it is in the UK. There is little evidence on how it works in practice and schemes vary considerably in terms of coverage, management, funding and initiation.
Bennett, Trevor, Farrington, David and Holloway, Katy (2008) The Effectiveness of Neighbourhood Watch: A Systematic Review. Campbell Systematic Reviews.
This narrative was prepared by UCL Jill Dando Institute and was co-funded by the College of Policing and the Economic and Social Research Council (ESRC). ESRC Grant title: 'University Consortium for Evidence-Based Crime Reduction'. Grant Ref: ES/L007223/1.
Uploaded on 19/02/15