ABSTRACT
A successful intervention is likely to reduce either the frequency or the intensity of violence. Secondly it may be important not only to determine whether any change has taken place but also to decide whether the change can reasonably be attributed to the intervention rather than to other factors. How, and the extent to which this is done, will depend upon the specific factors in each situation including the purposes for which the information is required. In particular it may be necessary to evaluate intervention with respect to the behaviour of individuals, of groups of individuals or with respect to changes in the amount of violence observed in particular settings. Typically the information available regarding a problem will permit several alternative formulations, with there often being insufficient evidence to choose between them. The more information there exists regarding a formulation the more likely it is to be correct.
