Further research is needed into all types of youth programs and in a variety of countries—particularly research that will contribute more to improving program design and implementation. For instance, programs could benefit from learning more about (1) which elements of the program are most important to reaching program goals; (2) how much participation or exposure is needed to increase knowledge, change attitudes, and influence behavior; (3) costs and cost-effectiveness; and (4) efficacy of different training and teaching methods. Of course, evaluation methods and indicators must suit the type and size of program. For example, evaluation of mass-media communication can measure recall, understanding and approval of messages, perception that others approve, intention to act, action, and advocacy with others (275).

Several recent initiatives seek to identify the most effective approaches to encouraging safer behavior among young people in developing countries. A working group on reproductive health services for young adults, representing a number of agencies working in this area, many with support from the United States Agency for International Development (USAID), has developed a manual of indicators to be used in evaluating youth programs (459). In 1996 the World Health Organization will issue a technical paper on accelerating health programs for young adults. The report will identify successful interventions and discuss strategies, including ways of mobilizing resources for expanding small or pilot programs (551). Also in 1996 Pathfinder International, in collaboration with the Futures Group and Tulane University, will begin a 5-year project supported by USAID to promote understanding of the health needs of young people and to identify program strategies that increase the practice of abstinence and other safe sexual behavior.