This paper reports the results of an experimental study modeled after a study conducted by Kaplan et al. The hypothesis tested was: Adolescents who receive a social learning intervention (SLI) will show greater improvement in metabolic control of their diabetes than will adolescents who receive only traditional diabetes instruction. Thirty-four adolescents between the ages of 12 and 16 who attended a midwestern camp were randomly divided into two groups. Both groups attended a daily 1-hour teaching session about diabetes. Later each day, the experimental group received an SLI consisting of role modeling to help the subjects circumvent peer pressure. To determine metabolic control, HbA1 values were determined before and 3 1/2 months after the intervention. The control group showed no significant difference in HbA1 values, while the experimental group showed a decrease in metabolic control.