'Working together' : Neighbourhood Watch, reassurance policing and the potential of partnerships

Foreword | Few well designed evaluations have found strong support for neighbourhood watch (NW) schemes; in fact there have been no formal, published, peer reviewed evaluations of NW in Australia. This paper argues for a change in focus in what is examined to determine success. Overseas evaluations suggest such schemes are ineffective because they looked at whether NW prevented and reduced the fear of crime, and improved information flows between the community and the police. The author proposes that a better way to assess the efficacy of NW is to view it as a vehicle to enhance partnerships between police, other agencies and the community and that these partnerships can effectively improve police/community relations, improve perceptions of safety and security and enhance community involvement in wider crime prevention initiatives. Such a change in focus is consistent with the emergence of 'reassurance policing' that targets street crime and disorder and has been a key component of recent reforms in the UK police force. This paper challenges practitioners, policy makers and communities involved with NW schemes to rethink the outcomes they are seeking to achieve and provides a rationale for community police partnerships that have the potential to improve feelings of safety and security in local communities.

Toni Makkai
Director

Neighbourhood Watch (NW) began as a response to shifting attitudes about the role of police in the community, and as a movement to have more community involvement in crime prevention². Similar schemes in the United States are referred to as 'Block Watch', 'Apartment Watch', 'Home Watch' and 'Community Watch'. Each is a collective attempt by the police and community to reduce crime. Through these schemes (and others like them) police have encouraged communities to take responsibility for crime prevention and controlling social/physical disorder in their neighbourhoods.

Despite widespread enthusiasm and support for NW by large numbers of people in Australia and overseas, there is significant evidence suggesting that it is ineffective at preventing crime (Sherman & Eck 2002: 315; Bright 1991: 78; Laycock & Tilley 1995), has a displacement effect (Barr & Pease 1990 cited in Bright 1991; Mukherjee & Wilson 1987: 3-4) and is only marginally successful in white middle-class areas where crime rates are already low (Bright 1991: 78; Skogan 1990; Husain & Bright 1990). Nor are the assumed subsidiary benefits, such as a reduced fear of crime and the increased flow of information between police and the community, substantiated by available evidence.

This paper reflects on NW in Australia, and considers the assumptions associated with NW schemes. Evidence-based research on NW schemes is discussed in this context. The report suggests that NW is found to be ineffective when evaluated in terms of minimising the incidence of preventable crime, reducing fear of crime and increasing the flow of information between the community and the police. The paper is not intended as a critical analysis of NW or of existing research. Rather, it suggests that NW schemes should not be assessed in terms of these factors, but by three alternative criteria:

• the capacity to enhance the relationship between police and the community;
• the ability to improve feelings of safety and security; and
• the ability to expand community involvement in wider safety and crime prevention initiatives.

Forming partnerships with police, other agencies and the community would enhance NW 's ability to achieve this. Recent literature on reassurance policing would seem to support this view (Dalgleish & Myhill 2004; Singer 2004; Povey 2001; Millie & Herrington 2005).

NW elements and structure

The first visible measure dealing with social disorder and rising residential burglary in Australia was the implementation of the NW program in Kannanook near Frankston, Victoria in June 1983. Other state and territory jurisdictions subsequently implemented various schemes with a view to reducing the opportunity for and vulnerability to crime. The various schemes across Australia, as elsewhere, differ in emphasis on crime prevention activities and organisational context. The extent of police involvement in these programs varies considerably.

The 'Working Together' sign that announces the presence of a NW program is a familiar sight in many parts of Australia. NW is a community crime prevention scheme in which police and community volunteers work together. It is entirely voluntary. The local community holds a public meeting to elect a coordinator and small committee. Typically, the coordinator liaises with local police and passes on information to the committee and the area's residents through public meetings and newsletters. Newsletters contain information on local police issues, safety and crime statistics. The meetings are usually informal with guest speakers involved in community interests in attendance, or there may be reports from the local police. The aims of the organisation are largely focused on: minimising preventable crime; increasing reporting of crime to the police; improving personal and household security levels; and encouraging people to engrave their property for ease of identification, should it be stolen.

The main elements of NW schemes are crime reporting, property identification, home security surveys and crime awareness. They can be implemented as a complete package (encompassing all elements) or as discrete programs (individual and non-conventional elements) that are designed for a specific problem or area. These programs can be based on a diversity of populations and mechanisms whereby programs seek to achieve their objectives. The discrete programs tend to be implemented as pilots in the first instance. For example, South Australia has recently established three pilot NW programs in suburbs with heavy Vietnamese, Greek and Arabic populations.

There are some structural differences between the NW schemes in Australia. Some schemes incorporate large residential areas of 300 to 3000 households. In North America, much smaller numbers (20-30) of households are involved. In Britain, the Metropolitan Police and other police services are moving towards smaller NW schemes, some as small as a single street. In some Australian states, NW programs are being established in single institutions, such as hospitals, high rise residential developments and tertiary institutions.

The general organisational structure of most NW programs consists of a board of management headed by a president, and other designated positions³. Most programs have a series of individual division/district, area and zone coordinators which allows for the demarcation of roles at all levels of NW . Some state-wide NW programs also hold a state forum to encourage the exchange of information between the divisions and the board. Aside from the practical aspect of NW initiatives, the divisions and the board are usually assisted by liaison officers. These positions often contain a sworn officer from the state police.

The level of police involvement in NW schemes differs between countries, and schemes within countries. Schemes in many countries are joint ventures between the local police and community members. In times of public sector fiscal restraint, funding has increasingly become the responsibility of the various NW programs. This is particularly true in Australia where state and territory organisations rely heavily on sponsorship and specific government grants. The police contribution is often 'in kind' support, for example, attendance at meetings, providing information and providing criminal checks for NW volunteers. Despite protestations of independence and a general view that police do not see NW as a priority in community policing, most NW organisations would welcome a more positive and 'hands on' approach from public police. The police resources required by each NW scheme vary depending on a community's individual needs, active membership levels, public liability insurance requirements and sponsorship availability.

NW National Secretariat

In recent years, state and territory NW bodies have established a national secretariat, whose duties are taken on by a different state or territory each year. Whoever holds the secretariat also hosts the national conference. In 2004 the secretariat was held by Queensland and the annual conference was held in Brisbane. In 2005, Victoria will be the host. The secretariat sees itself as a collaborative body that assists with the sharing of information between the various programs on issues such as information technology, facilitating national sponsorship and uniform marketing strategies.

Evidence-based research on NW: assumptions and reality

The main impetus for NW came from a shift in attitudes about methods of urban policing to more community-style policing (Weatherburn 2004: 108-109). Police have traditionally considered their role to be one of crime control and law enforcement. In the second half of the twentieth century, police services around the world sought to develop a closer relationship with local communities. Led by the US, experiments were conducted that drew on community resources to address rising crime rates. These changes have been termed 'community policing' and 'community-oriented policing'. In the US and UK, specific legislation and government funding explicitly promote the concept of partnership policing and community involvement. In Australia, while state and territory police organisations and all levels of government support the concept of partnership and 'working together to fight crime', significant structural, financial and legislative support is largely absent for such programs and schemes.

Community policing arose from criticisms that crime control methods previously adopted by police services had failed to address crime adequately. Rising crime rates, increased fear of crime and low clear up rates were held as evidence of this failure. In short, it was argued that the changes in the urban environment had not been accommodated by police practice. After recognising that crime control could not occur without the assistance of the community, police departments worked at developing closer relationships with members of their community and harnessing the resources these communities had to offer. As Weatherburn has pointed out, the 'best known community-policing strategy [was] Neighbourhood Watch' (2004: 108). The organisation seemed to be an ideal vehicle for the community to work more closely with police and assist in the fight against crime. The ideal was based on a number of assumptions.

Assumptions/benefits of NW schemes

• The willingness of the community to actively watch and report anything suspicious increases the chance of an offender being caught, that is, through reduced opportunity. This scrutiny deters potential offenders.
• The distribution of NW schemes across diverse geographical areas allows for homogeneous membership levels.
• As the 'eyes and ears' of the police, communities' levels of crime reporting and intelligence provided to the police have increased. This information leads to more arrests and convictions, resulting in a reduction of crime.
• Police share information with citizens, such as crime statistics and crime education material, with a view to reducing crime and victimisation.
• NW reduces crime through decreasing opportunities to offend by creating signs of occupancy, property marking, security surveys and greater security awareness.
• Active NW schemes reduce fear of crime.
• Residents and communities are generally willing to participate in NW schemes.
• Formal meetings strengthen neighbourhood dynamics.
• NW activity consolidates a strong and meaningful partnership between law enforcement officers and the community.

Evidence-based research on NW

Despite the benefits and potential advances heralded by the promoters of NW, the available evidence does not wholly substantiate the benefits assumed from NW activities. (This is true even when we concede that NW was not established with a view to reducing 'all crime'. Its original brief was to reduce burglary and property theft.) Disappointingly there are no formal evidence-based evaluations of NW schemes in Australia⁴. This section looks at some of the available research evaluations⁵ available from the US and Britain⁶. Unfortunately most of this research is over 10 years old.

In their discussion of evidence-based crime prevention, Sherman and Eck (2002) canvass the findings on NW schemes⁷. The evidence for the findings is pronounced by the authors as being 'moderately strong'. The researchers found the following.

• People in high crime areas were the most reluctant to organise themselves in terms of crime prevention, and less likely to be involved in organisations like NW. Distrust of neighbours was a factor in reluctance and refusal to attend or host community meetings. Areas where that trust is higher (e.g. middle-class areas) had little crime to begin with and the effects of NW were difficult, if not impossible, to measure (Hope & Trickett 1995).
• Fear of crime was seen to rise in areas where a NW scheme was operating (Skogan 1990). In his evaluation of a Chicago 'Block Watch' program, Rosenbaum (1987) found that there were significant increases in the fear of personal crime and property crime.

NW meetings intended to integrate the community, increase community contact with police and allow police to pass on information may fulfil these objectives. However, various studies found 'no reduction in victimization' (Wycoff & Skogan 1993).

• Public meetings, rather than meetings held in neighbours' homes, 'show more promise' with wide attendance found in the districts with high crime rates and a more targeted discussion about police/community strategies (Skogan 1996).
• Meetings where police have regularly attended and provided consistent information to community groups (18 monthly meetings in five districts) suggested reductions in some crime and victimisation measures (Skogan et al. 1995).
• Rosenbaum et al. (1986) suggests meetings 'reduc[e] feelings of efficacy and social cohesion'. There is no evidence that NW can be 'justified on the basis of reducing middle-class fear of crime ... since no such effects have been found' (Sherman & Eck 2002: 317). Indeed Skogan (1990) finds that NW increases fear of crime.
• An increased flow of information from police to the community through police newsletters was tested in Newark and Houston in the US. This communication failed to reduce victimisation and violence (Davis & Taylor 1997; Pate et al. 1986).
• An assessment of a Seattle NW program in 1986 concluded that burglary was reduced, but only for an 18-month period (Lindsay & McGillis 1986).
• On evaluating NW schemes generally, researchers found that 'Block Watch' had no effect on crime and that there was less surveillance in poorer areas (Bennett 1990; Pate et al. 1987; Rosenbaum et al. 1986).
• An additional problem is the displacement effect. Most NW schemes focus on residential burglary. A good crime awareness campaign can change criminal behaviour. If no opportunities for burglary exist, offenders may change their target and method. Instead of reducing crime, NW schemes may displace those rates into other types of crimes (Mukherjee & Wilson 1987).
• In his evaluation of two UK suburbs running comprehensive NW programs Bennett (1990) found some improvement in levels of satisfaction over both suburbs. The level of satisfaction with the area and a sense of social cohesion was found in one suburb, and satisfaction with the area and home protection behaviour was found in the other. The same research evaluation found:
  • no measured impact on the crime rate;
  • no improvements in reporting rates;
  • no improvements in clear up rates;
  • no changes in public calls to police;
  • no decrease in the fear of personal victimisation or probable victimisation;
  • no change in the way people viewed the police; and
  • no discernible improvements in home protection behaviour.
• In one suburb there was no improvement in fear of household crime or a sense of social cohesion. These disappointing results were despite the fact that:
  • police met regularly and enthusiastically with coordinators;
  • coordinators held regular meetings with volunteers;
  • coordinators regularly invited outside speakers;
  • newsletters were distributed regularly; and
  • police worked hard to make the project a success.
• Bennett also found that there was little evidence to show that NW schemes encouraged police reporting. Nor was there any indication that police responded more favourably to calls for service from NW areas.
• Husain and Bright's research into NW and the police in Britain (1990) suggests that in high crime, multi-racial areas (particularly in public housing) where community integration and trust levels of neighbours and police are low, NW membership is minimal. These community members often lack the organisational skills and resources available to those in high status neighbourhoods (cited in Bright 1991: 78).

This limited, but consistent and moderately strong evidence supports the view that when assessed in the context of reducing crime and alleviating anxiety about crime rates, conventional NW programs are 'ineffective'. The evidence that such programs increase the information flow from the community to the police is 'at best only promising' (Sherman & Eck 2002: 315; Bennett 1990). In evaluations that have been conducted using Maryland Scientific Methods Scale quality assessment scores (see n2), researchers found that people are reluctant to organise in high crime areas (Hope 1995). Where people are prepared to organise, crime rates are lower to begin with (thus making it more difficult to measure effects on crime). Membership/support of NW shows a general bias towards 'middle to better off sectors and communities' (Sims 2001). Expenditure on NW programs cannot be justified on the basis of reducing fear of crime since no robust evidence to support this claim can be found.

Overall, these results are disappointing in terms of crime reduction and lowering fear of crime. However, as Sherman and Eck (2002: 316) note in their discussion of community policing, we do know that 'increasing the quantity and quality of police citizen-contact reduces crime'. Improving police visibility, accessibility and familiarity improves police effectiveness and reassures the public, improving feelings and perceptions of safety (Dalgleish & Myhill 2004: vii-viii). In areas where police are responsive to community concerns, demonstrate a willingness to listen and interact positively with the community there is a greater willingness to obey the law, a reduction in serious crime and, in some instances (for example, domestic violence), a lower recidivism rate (Tyler 1990; Sunshine & Tyler 2003; Paternoster et al. 1997; Skogan 1990; Skogan et al. 1995; see also Skogan & Frydl 2004: Chapter 8). Given these positive outcomes, there is the potential to use the existing NW structure as a vehicle to build partnerships with police and the community, increase the quality and quantity of police citizen contact (and thus legitimacy) and at the same time build community capacity. It may, as Bright (1991: 78) has suggested, 'help to strengthen communities and lead to benefits that will only be apparent in the longer term'. Recent research in the UK on reassurance policing suggests that this approach offers promising opportunities.

Reassurance policing

Increasing anxiety about crime and disorder problems in the suburbs can reduce confidence in police and exacerbate feelings of insecurity. In recent years, raising confidence in police and increasing feelings of safety in the community has been a central part of police reform in Australia and the UK (see for example, Dalgleish & Myhill 2004; Singer 2004; Povey 2001; HM Inspectorate of Constabulary (Scotland) 2002; Millie & Herrington 2005). At the same time, there has been widespread recognition that no single agency, and certainly not the police, commands the resources necessary to control crime in contemporary Australia. Effective crime control requires community coordination and support, involving individuals and institutions outside law enforcement and beyond the public sector. Indeed, Loader (2000: 330) refers to the existence today of 'dispersed, interorganizational policing networks'.

The idea that communities need reassurance from police has become a key component of the UK police reform agenda. It is incorporated in the Police Reform Act 2002. In August 2004, the British Home Office released Reassuring the public - a review of international policing interventions. In that document 'reassurance' is defined as:

the intended outcome(s) of actions taken by the police and other agencies to improve perceived police effectiveness (mainly confidence in, and satisfaction with, the police), and to increase feelings and perceptions of safety (including reducing the fear of crime) (Dalgleish & Myhill 2004:9).

Home Office researchers evaluated programs on the basis of factors which improved feelings and perceptions of safety and police effectiveness. In the case of both concepts, NW programs were classified as unknown. This appraisal was allocated on the basis that there was insufficient methodological information available to the researchers for assessment. Where the program yielded 'positive reassurance outcomes' but were statistically insignificant, the authors conceded that in a different context these programs may be classified as 'promising' and that community engagement and community policing are 'promising' in terms of perceptions of police effectiveness and feelings of perceptions of safety (Dalgleish & Myhill 2004: 52-55).

Conclusion

The notion that NW programs can promote neighbourliness and social cohesion is not a new one. Early NW programs in Britain emphasised these aspects of the schemes and promoted an idealised notion of 'the community' and community spirit (King 1991: 98). In Victoria, where NW celebrated its 21st birthday in 2004, the organisation promotes its central objective as the 'facilitation of communication between police at all levels and members of local communities' (Victoria Sentinel 2004: 9). As Bright has noted, 'many people involved with NW are convinced of its value and gain reassurance from its existence' (1991: 78). Despite this, NW programs are not ordinarily assessed formally on these attributes⁸. Brown, in her extensive literature review of NW, sees better community-relations and an improvement in community spirit 'more appropriately ... perceived as benefits resulting from the programme rather than aims in themselves' (1992: 1). Indeed Brown's review of NW was specifically intended to 'identify factors associated with the successful establishment, continuation, operation and practices of individual NW schemes or types of scheme' (1992: i).

What is needed is to marry the neighbourliness and social cohesion potential of NW schemes with a stronger emphasis on police-citizen contact. This creation of partnerships would, potentially, go some way to 'reassuring the public; such partnerships could be evaluated accordingly. The reduction of crime may become a benefit of the program, rather than the principle objective. As Willis has pointed out, 'working with communities to reduce their concern about crime and disorder is as important as bringing down the actual level of crime and disorder prevalent in neighbourhoods' (2004: i). A recent 'reassurance policing and local management of community safety' project in the UK has suggested that partnerships which encompass police, other agencies, community organisations and the community can affect 'modest but nonetheless positive' gains. Such gains indicate that 'working together can make a difference' (Willis 2004: i).

A well organised, focused and effective NW unit, whose aims seek to improve the degree of personal and household security, enhance the relationship between police and the community and expand the community's involvement in wider community safety and crime prevention initiatives, could make a difference. A partnership encompassing police, business organisations, other government agencies and members of the community could target high crime areas, places of social deprivation and communities where local residents have concerns about specific issues. These areas are where the potential benefits of such an initiative are the greatest. As a group it would act as a forum for comprehending local problems, facilitating activities, evaluating progress and ensuring both the commitment of the partners and the capacity of the project to deliver. As an organisation, this activity would potentially keep the membership active and engaged. Such a venture is a move away from traditional NW imperatives but research suggests that significant progress can be made in reducing crime while reassuring the public, building community capacity, improving the quantity and quality of police citizen-contact and enhancing police legitimacy. As a secondary consideration the project may even minimise the incidence of preventable crime. In testing such hypotheses, rigorous systematic evaluation will be crucial.

References

All URLs correct at September 2005

• Bennett T 1990. Evaluating Neighbourhood Watch. Aldershot: Gower.
• Bennett T, Holloway K and Farrington D 2005. The effectiveness of neighbourhood watch, in The Campbell Collaboration reviews of intervention and policy evaluations (CP-RIPE). Pennysylvania: Campbell Collaboration.
• Bright J 1991. Crime prevention: the British experience. In K. Stenson and David Cowell (eds) The politics of crime control, 62-86. London: Sage.
• Brown L 1992. Neighbourhood Watch: a literature review. Edingburgh: Scottish Office, Central Research Unit.
• Dalgleish D and Myhill A 2004. Reassuring the public - a review of international policing interventions, Home Office Research Study 284, London: Home Office Research, Development and Statistics Directorate.
• Davis RC and Taylor BG 1997. A proactive response to family violence: the results of a randomized experiment. Criminology 35: 307-33.
• Farrington DP, Gottfredson DC, Sherman LW and Welsh BC 2002. The Maryland Scientific Methods Scale. In L.W. Sherman, D.P. Farrington, B.C. Welsh and D.L. MacKenzie (eds) Evidence-based crime prevention. London: Routledge.
• Hope T 1995. Community crime prevention, in Tonry M and Farrington D (eds) Building a safer society. Crime and justice 19. Chicago: University of Chicago Press.
• Hope T and Trickett A 2004. Angst essen steele auf... But it keeps away the burglars! Private security, neighbourhood watch and the social reaction to crime, Klner Zeitschrift fr Soziologie und Sozialpsychologie. Issue N. 42 'Neue Criminalsoziologie'.
• HM Inspectorate of Constabulary (Scotland) 2002. Narrowing the gap. Police visibility and public reassurance - managing public expectation and demand. Edingburgh.
• Husain S and Bright J (eds) 1990. Neighbourhood Watch and the police, Crime Concern, Swindon, UK.
• King M 1991. The political constriction of crime prevention: a contrast between the French and British experience. In K Stenson and D Cowell (eds) The politics of crime control, 87-108. London: Sage.
• Laycock G and Tilley N 1995. Policing and neighbourhood watch: strategic issues. Crime detection and prevention series, Paper 60. London: Home Office, Police Research Group.
• Lindsay B and McGillis D 1986. Citywide community crime prevention: an assessment of the Seattle program. Community crime prevention: does it work? In D Rosenbaum (ed) Beverly Hills, CA: Sage.
• Loader I 2000. Plural Policing and Democratic Governance, Social and legal studies, 9 (3) 323-345.
• Millie A and Herrington V 2005. Bridging the gap: understanding reassurance policing, The Howard journal, 44 (1) February 41-56.
• Mukherjee S and Wilson P 1987. Neighbourhood watch: issues and policy implications, Trends and issues in crime and criminal justice, No. 8. Canberra: Australian Institute of Criminology.
• Pate A, McPherson M and Silloway G 1987. The Minneapolis community crime prevention experiment: draft evaluation report. Washington, DC: Police Foundation.
• Pate A, Wycoff M, Skogan W and Sherman L 1986. Reducing fear of crime in Houston and Newark: a summary report. Washington, DC: Police Foundation.
• Paternoster R, Brame B, Bachman R and Sherman LW 1997. Do fair procedures matter? The effect of procedural justices on spouse assault, Law and society review, 31 (1) 163-204.
• Povey K 2001. Open all hours: a thematic inspection on the role of police visibility and accessibility in public reassurance, Her Majesty's Inspector of Constabulary. London: Home Office.
• Rosenbaum D 1987. The theory and research behind neighbourhood watch: is it a sound fear-and-crime-reduction strategy? Crime & delinquency 33: 103-34.
• Rosenbaum D, Lewis DA and Grant JA 1986. Neighbourhood-based crime prevention: assessing the efficacy of community organizing in Chicago. Community crime prevention: does it work?, D Rosenbaum (ed). Beverly Hills, CA: Sage.
• Sherman LW and Eck JE 2002. Policing for crime prevention. In LW Sherman, DP Farrington, BC Welsh and DL MacKenzie (eds) Evidence-based crime prevention. London: Routledge.
• Sims L 2001. Neighbourhood watch: findings from the 2000 British crime survey, Findings 150. London: Home Office.
• Singer L 2004. Reassurance policing: an evaluation of the local management of community safety, Home Office research study 288. London: Home Office Research, Development and Statistics Directorate.
• Skogan W and Frydl K (eds) 2004. Fairness and effectiveness in policing: the evidence, National Research Council of the National Academies. Washington, DC: The National Academies Press.
• Skogan W 1996. Community policing in Chicago, year three. Chicago: Illinois Criminal Justice Information Authority.
• Skogan W 1995. Community policing in Chicago, year two, NIJ research preview. Washington DC: US Department of Justice.
• Skogan W 1990. Disorder and decline: crime and the spiral of decay in American neighbourhoods. New York: The Free Press.
• Sunshine J and Tyler T 2003. The role of procedural justice and legitimacy in shaping public support for policing, Law and society review, 37 (3) 513-547.
• Tyler T 1990. Why people obey the law. New Haven: Yale University Press.
• Victoria Sentinel: Official journal of Neighbourhood Watch Victoria Inc. 2004 Neighbourhood watch 21st anniversary, July, 8-9.
• Weatherburn D 2004 Law and order in Australia: rhetoric and reality, Leichhardt: Federation Press.
• Willis C 2004. Foreword. In L Singer Reassurance policing: an evaluation of the local management of community safety, Home Office research study 288, London: Home Office Research, Development and Statistics Directorate.
• Wycoff MA and Skogan W 1993. Community policing in Madison: quality from the inside out. An evaluation of implementation and impact. Research report. Washington, DC: Police Foundation.

Endnotes

1. The Australian Research Council funded part of the research for this paper (Grant number LP0346987).
2. For a full discussion about the theory that underpins NW approaches to crime prevention, see Bennett 1990: Chapter 3.
3. As yet there is no definitive description and analysis of the composition of NW boards. However, there is some evidence that there are low levels of commitment or involvement from members generally, although 'high commitment from many coordinators' (Laycock & Tilley 1995). For a discussion about NW participation rates generally, see Bennett 1990: 47-409.
4. Mukherjee and Wilson's discussion about Neighbourhood Watch in Victoria (1987) subjected data to two statistical tests. In their regression analysis the authors cautioned that 'the lack of sufficient number of data points' meant that the results could not 'be used with a great deal of confidence' (1987: 4-5).
5. Crime prevention generally and NW specifically has a small evaluation literature. The most rigorous evaluations are usually based on the Maryland Scientific Methods Scale. This five-point scale ranges from level 1 (the weakest design) to level 5 (the strongest design) in terms of overall internal validity. Evaluations should be at least at level 3 in order to conclude the program worked. This level requires that the evaluation must include a comparison of one or more units and one or more comparable control units over time (Farrington et al. 2002). The research projects discussed here (with the exception of the Husain and Bright research) are those that have used the standard of at least two consistent findings from level 3, that is, before and after studies with a comparison group and consideration of other factors that may support the same conclusion.
6. For a comprehensive review, which includes research that does not necessarily meet the rigorous evaluation standards set by MSMS, see Brown, 1992.
7. For another review of the published literature, see Laycock and Tilley 1995: 21-43.
8. Although recent pilot programs in the UK are targeting confidence in police and enhanced legitimacy of police in a series of 'reassurance strategy trials'. The findings of such initiatives are as yet to be realised (Millie & Herrington 2005).

Note: Trevor Bennett and his team have recently concluded (in a yet to be reviewed Campbell Review) an assessment of the effectiveness of NW in reducing crime. The strongest finding of that review is that 'across all studies combined neighbourhood watch is associated with a modest reduction in residential burglaries'. However, the authors advise caution about such conclusions because, apart from such variations in results obtained across different studies, 'little is known about the mechanisms by which neighbourhood watch might have contributed to these reductions' (2005:35).

About the authors

Dr Jenny Fleming is a Fellow at Security 21, Regulatory Institutions Network, at the Australian National University.
Dr Fleming is currently working with ACT Policing on the Policing in the 21st Century project.