Social learning theory, social inoculation theory, and differential association theory have been used as rationales for peer programs, which is an umbrella term for programs that involve peer training, peer facilitation, peer counseling, peer modeling, or peer assistance; however, there is limited empirical evidence to support or refute theoretical arguments for the value of peer programs. In addition, the varied aims and procedures of existing peer projects make comparison and evaluation difficult. A discussion of peer programs in correctional settings addresses their use in HIV/AIDS and health education, drug and alcohol abuse treatment, sexual assault/offending, prison orientation, and suicide/violence prevention. A discussion of the risks of prison-based peer programs notes the lack of sound empirical studies that show the preference for and the enhanced effectiveness of peer education approaches over professionally delivered interventions, the critical importance of training peer leaders for program responsibilities, peer leaders' unresolved problems, ethical concerns, and confidentiality issues. In spite of the risks involved, however, prison-based peer programs can have the benefits of expanding the number of rehabilitation programs, drawing upon the behavioral dynamics of positive peer influence, and using the instructive power of shared peer experiences. Process and impact evaluations must be conducted to ensure that the program has been implemented as planned and to determine whether or not the program's goals and objectives are being achieved. The authors conclude that with careful planning, implementation, and evaluation, peer programs may be the key to developing an alliance among offenders, correctional staff, and correctional management. 2 tables and 38 references