In practice, empirically derived evidence is concerned with estimating the extent to which an observation is true in a scientific sense and what is the probability that an intervention, when implemented, will produce a particular type of outcome. There are three types of evidence that will help community services focus on outcomes. The first is reliable and usable evidence about the contexts in which policies are realized and practices are undertaken. Such evidence recognizes the importance of measuring diversity and taking this into account. The second type of evidence pertains to the type and level of effect produced by various types of interventions in the real world. The third kind of evidence is that which links what practitioners do, that which they are able to document, with the eventual outcomes for clients, which are often difficult to measure within the world of daily practice. It is here that the findings of empirical research must be linked to practice outputs, so that it becomes possible to estimate the probable effects of actions. Linking the research effort to the practice effort through collaboration is fundamental to the eventual success of an evidence-based approach. Of equal importance is linking the people who are affected by policies and practices to the enterprise of evidence making.