Havelock and later theorists see effective improvement of professional practices as a combination of these four approaches. But the separate models are well worth elaborating, for they represent strong differences in the ways evidence is treated. I will briefly describe each of the first three and then spend a bit more time on the fourth approach (which is usually of particular value to those seeking to influence a specific professional practice).

THE RATIONAL APPROACH

This approach is based on assumptions regarding rationality which lead to heavy investment in basic and applied research and to considerable investment in the formulation, testing and packaging of innovations based on research. This is the approach that seems to underlie most of the current attempts to use evidence in the attempt to improve professional practices—at least during the early stages of the improvement process. Havelock identifies five basic assumptions which underlie the Research and Development strategy. First, the R&D model suggests that dissemination and utilization should be a rational sequence of activities which moves from research to development to packaging before dissemination takes place. Second, this model assumes that there has to be planning, and planning on a massive scale. It is not enough that we simply have all these activities of research and development; they have to be coordinated; there has to be a relationship between them; and they have to make sense in a logical sequence that may go back years in the evolution of one particular message to be disseminated.

Third, there has to be a division of labor and a separation of roles and functions, an obvious prerequisite in all complex activities of modern society, but one that we sometimes ignore. Fourth, it assumes a more or less clearly defined target audience, a specified passive consumer, who will accept the innovation if it is delivered on the right channel, in the right way, and at the right time. Extensive scientific evaluation is needed to assure that this happens. Evaluation is to occur at every stage of development and dissemination. Fifth, this perspective accepts the fact of high initial development cost prior to any dissemination activity, because it anticipates an even higher gain in the long run—in terms of efficiency, quality, and capacity to reach a mass audiences. We can see around us plenty of examples of the rational change model. Cars and planes and other material “products” are made and sold that way.  Many federal government agencies have employed rational assumptions in supporting the research and development of various programs ranging from agriculture to education, and from Internet technology to defense. Evidence-based medicine and psychotherapy would seem to have been founded on this set of assumptions.