The model of evaluation from the health field that those in policing and crime prevention have been invited to attempt to emulate is the randomized clinical trial (RCT). The RCT focuses on a specific treatment and determines whether or not it is effective in treating a particular condition. It allocates potential recipients randomly to control and treatment conditions and measures the change in each before and after the treatment has been provided. The difference in outcomes for the two groups provides a measure of effect size. If the effect size is positive and sufficiently large in relation to costs, then the treatment can be endorsed. If it is negative or the effects are too small in relation to costs, it can be discouraged or prohibited. Warburton and Shepherd (2000) conducted a study that applied RCT to a crime-prevention issue. Their study examined the effectiveness of introducing toughened "nonic" pint glasses in bars in Britain as a means of reducing injuries. Nonic glasses are those without a handle, but with a bulge toward the top. The simple RCT undertaken yielded little practical findings of the effect of the intervention. Most other crime prevention interventions are much more diverse than the replacement of one type of pint glass with another in the expectation that when used as a weapon, fewer disfiguring injuries will result. There are developments that challenge the dominance of the RCT, and they may be more promising as models for evaluating policing and crime prevention. This chapter briefly reviews some of these models, and recommendations for future work are offered. 1 table, 2 notes, and 25 references