The use of mass media has long been an attractive method for implementing and discharging institutional responsibility for the promotion of good health practices and the prevention of various social and health problems. Although there is a long history of such efforts, relatively little is known about the effects such campaigns might have. The large number of programs currently being conceived and implemented on local, state, and federal levels have not benefited as much as they might have from the experience of past mass media efforts. This paper reviews the history, commonalities, assumptions, and effects of planned largescale campaigns to communicate information to the general population to encourage moderation or abstinence in the use of alcohol, drugs, or tobacco. The concept of experimental design is critiqued regarding the appropriateness of such a model for evaluation of mass media campaigns. Suggestions are offered for making evaluations of such campaigns more able to generate useful information for program and planning purposes. This includes expanding evaluation questions through additional forms of inquiry rather than constricting the information obtained through methods based on assumptions of experimental design.