Problem-oriented policing initiatives are one important form of crime prevention, and they offer opportunities for learning about implementation success and failure. Problem-oriented policing initiatives can succeed or fail for a variety of reasons, among them: inaccurate identification of the problem, inaccurate analysis of the problem, inadequate implementation, or application of an incorrect theory. This paper draws upon both the research literature and reports on problemoriented policing initiatives to identify those factors that best explain why action plans do or do not get implemented. It identifies and provides examples of five clusters of factors that help explain implementation success or failure: (1) characteristics, skills, and actions of project managers; (2) resources (3) support and cooperation external to the police agency; (4) evidence; and (5) complexity of implementation.

One of the common features of recent evaluations of community safety and crime reduction initiatives has been the observation that project management has often been weak and ineffectual. The implications of this are that improving project management skills of the staff involved in such initiatives and implementing effective management processes would increase the likelihood of success. This chapter explores the various weaknesses found in project managing community safety initiatives and presents a model of the "dynamic project lifecycle." This model helps to identify how projects evolve and helps to identify ways in which common problems can be resolved

It has been recognised for some time that implementation is not as easy as was once assumed. This chapter looks at some of the conflicts and tensions associated with implementation and suggests that they form major obstacles to achieving effective action. An appreciation of these difficulties can assist with their resolution, which on occasion might result in the need to apply leverage against individuals, companies or even governments to change the contingencies under which they operate and persuade them that crime prevention is a high priority and ultimately in their interest. The problem of "carding" - i.e., the placing of advertisements for prostitution in telephone kiosks - is discussed as a means of illustrating the complex issues that can arise in the course of attempting to implement a crime prevention scheme.

Organizations or persons with the power to affect the opportunity structure for crime are not always ready to act. One should be prepared to encounter little enthusiasm and, in some instances, even open resistance. Given these negative expectations, the process of implementation went surprisingly smoothly in one case involving the practice of problem-oriented policing in a small town in Norway, and the project was successfully implemented. A more detailed examination shows that the organization of the project, the choice of measures and the process of putting them into practice were all carried out in an efficient way, thus explaining the success.

An analysis of the processes used for implementing the U.K.'s Crime Reduction Programme, together with some recent crime prevention efforts in Australia and New Zealand, demonstrates that many of the problems found in the implementation of local crime prevention projects are the result of a number of common and frequently repeated errors in the way that central and local agencies relate to each other in the planning, delivery and evaluation of initiatives. Some of these errors include the use of centrally defined crime targets to establish priorities; short-term, non-recurrent project funding; project management failure resulting from too many simultaneous interventions; a failure to identify and plug gaps in implementation skill and capacity; and a hands-off approach to implementation support. It is argued that a stronger process of active and ongoing engagement between central and local agencies is necessary to overcome this prob- lem. The model proposed treats the policy and program delivery chain as a single integrated system rather than a fragmented process with disconnected ownership and responsibility. In other words, it promotes the equal importance of vertical policy and service delivery integration and support (i.e., through all layers of government and the community) as well as the more common horizontal cross- agency integration (e.g., at the local, regional or central levels).

This paper advances an explanation for the very widespread implementation failure that has been seen in British initiatives to reduce crime through partnership work. It suggests that a narrow form of rationalism has been applied to crime problems that oversimplifies the nature of these problems and misidentifies priorities for action. This reductionist process is the consequence of the interaction between imanagerialist methods associated with the reform or "modernisation" of public services and a theory of knowledge which admits a narrow range of "evidence" about what works. The various crime reduction partners have not seriously engaged with crime reduction efforts, because this rationalist perspective has privileged tactical solutions of a particular sort at the expense of strategic solutions with which senior managers in partner agencies are actually preoccupied.

It is of vital importance that practitioners plan ahead to help avoid future implementation problems. By thinking strategically, it should be possible to foresee at least the more common problems that might be faced. Here we attempt to assist in this process by producing a typology of schemes and discussing the elements of different sorts of scheme that both guard against and accentuate the risk of implementation failure. For example, schemes that are very specific about the crime they are targeting are likely to experience fewer problems than those with more general aims. We argue that this will generally be true even when the implementation team is less experienced and where problems arise in relation to under-resourcing or red tape. The reason for producing the typology is to help practitioners more easily identify the types of scheme which will be less risky to implement. However, it is important to consider that there is likely to be a trade- off associated with choosing to implement schemes that are likely to be easily delivered. For instance, implementing high-risk approaches can encourage innovation and generate new knowledge. Whichever the strategy taken, we recommend that the practitioner is at least aware of potential implementation problems prior to implementation. Consequently, we discuss approaches that can aid practitioners in considering the risk and protective features of their particular planned operation(s).

The most frequent reason for implementation failure is asserted to be that crime reduction is typically peripheral to the purposes of those tasked with it, a state of affairs which they cannot acknowledge overtly. Three examples are given of marginalisation of the crime reductive purpose in practice. The neglected role of cognitive social psychology in understanding the implementation process alongside the political mindset of implementers is highlighted and asserted to be a necessary element in translating sound crime reduction theory into practice.

Good practice involves knowing what to do and being able to do it. There are a number of generally accepted attributes of good practice: it must be ethical, effective, without significant negative side-effects, and economical. There have been, and continue to be a number of efforts to provide catalogues of good practice or good practice guides. What is to be reproduced as good practice can, however, be construed in differing ways: for example as a specific set of preventive activities or interventions; as a methodology for selecting preventive interventions to be adopted, or the informed activation of tested principles or mechanisms by whatever means are available; or as the exercise of well-honed tacit craft skills. Likewise, good practice can be identified in a number of ways: for example from rigorous experimental studies associating specific practices and intended outcomes, from first principles, from informed opinion, or from coherent and tested theory. For the policy-maker attempting to disseminate good practice, for the practitioner committed to implementing it, and for the researcher or educator concerned to inform practice these are important issues. This paper identifies and compares two major approaches to good practice guidance: "What works?" and, "What's to be done?" It reviews their relative strengths, weaknesses and potential uses. It also discusses common shortcomings, in particular neglect of the tacit, and what might be done about it.