capable leader with the characteristics described by Tobler would be just as or more effective as a resistance skills training course taught by a teacher.
Timing, duration, and the characteristics of the deliverer are potentially important moderator variables that could explain the wide range of effects observed across studies of prevention activities that are otherwise similar. But at present, we can only guess about which activities and what about each activity is critical to its success. The knowledge base for choosing among the multitude of prevention options is severely limited. Each of the potential moderator variables must be systematically varied in rigorous prevention trials.
Much remains to be learned about the potential of prevention activities for reducing illegal drug use. The committee identified five major areas in which answers from additional research would bridge this knowledge gap. Research is needed to examine
Which of the noninstructional modalities are effective for reducing drug use.
Whether prevention activities affect the subsequent drug use of different user groups differently. To what extent do prevention messages spread to individuals and groups not initially targeted, and can this “diffusion effect” be harnessed to reduce drug use in high risk peer groupings?
Whether prevention activities affect the quantity, frequency or problems associated with use of nongateway substances.
What prevention content is most effective, with which groups.
How the timing, duration, and characteristics of the deliverer condition the effects of prevention programs. Does the effectiveness of prevention effects vary relative to the timing of drug epidemics? Are there important trade-offs between total dosage delivered and timing of delivery of prevention messages?
Table 7.2 shows the areas that must be studied for each prevention modality in order to fill in gaps in understanding of the potential of prevention. Most is known about the ideal content of instructional programs, but in the committee’s judgment, more research is needed even in that cell.
Once these gaps are filled, the next step will be to explore how effects can be enhanced through combinations of the most effective modalities. A number of studies have combined several modalities (e.g., Battistich et al.,