This article describes an evaluation of a school-based peer education intervention for HIV prevention among students in 27 high schools in Aden, Yemen. The intervention was developed after a survey among the same population in 2005, which revealed a high level of stigma toward people living with HIV (PLWH) and a low level of HIV knowledge. In a quasi-experimental design, students who received the peer education intervention (78.6%) were compared with students who did not receive the intervention (21.4%). No systematic procedure was applied in selecting students for the intervention condition. Data were collected using a self-administered questionnaire from a sample of 2,510 students from all 27 high schools in Aden governorate. To increase internal validity, students were also compared with a cohort control sample surveyed in 2005, which was a random sample of 2,274 students from the same schools. Sixty-eight percent of students targeted by peer education had good knowledge scores, compared with 43.3% of students not targeted by peer education. Multi-level regression analysis revealed that, although there was a significant difference among schools, the intervention effect of peer education at the individual level was significant; students who received peer education had a statistically higher knowledge score compared with those not targeted. Compared with the 2005 cohort control sample, students targeted by peer education had better knowledge on the modes of transmission and prevention and fewer misconceptions; knowledge on the use of condoms increased from 49.4% to 67.8%. In addition, students who received the peer education interventions suggested significantly more actions to provide care and support for PLWH. Also, the levels of stigma and discrimination were much higher among the 2005 cohort control group, compared with those who received the peer education intervention.