Students were also trained to become outreach educators for their family and local communities.
Five experimental schools were randomly selected from schools in Addis Ababa (two senior schools and one junior secondary school) and Debra Birhan (one senior and one junior secondary school). Two comparable junior and two comparable senior secondary schools were designated as controls. Most students in these schools conformed to the normal grade/age range, with 38 percent aged 12-14 and 33 percent aged 15-16, although reported ages ranged from 9 to 22. Examination of baseline data showed no significant differences between the experimental and control schools in terms of demographics, knowledge, and attitudinal variables. Other evaluation design features included a pre and post in-class questionnaire; in-depth student KAP interviews; and random surveys of parents and community members about their HIV/AIDS-related knowledge and opinions and their exposure to student educators and AIDS education materials. Monitoring of the project's implementation was performed by teachers and research assistants through checklists and observations. In addition to frequent supervisory visits by project staff, an evaluation workshop allowed teachers to provide feedback about the success or failure of the program.
Results from the project were positive. While rigorous statistical analysis was not performed, experimental students consistently showed the most knowledge gains. At the post-test, learning differentials between experimental and control groups ranged from 13 to 27 percent for items related to HIV-prevention facts (e.g., condoms reduce HIV risk) and correction of transmission myths (e.g., mosquitoes transmit HIV).
The experimental group also showed a greater change at post-test in their attitude toward classmates and others with HIV/AIDS. For example, at post-test, experimental students showed a 35 percent increase in willingness to care for someone with AIDS as compared with a 15 percent increase among the control students. Experimental students also showed a 26 percent increase in willingness to accept a teacher with AIDS, as compared with a 12 percent increase for control students. Statistical significance, however, was not reported. A "considerable effect" on attitudes toward risky sexual behavior was noted among the experimental students. At post-test, experimental students were more likely than control students to support abstinence (24 percent increase among the experimental groups as compared with a 10 percent decrease among the control groups), to report resisting pressure to have sex (20 percent increase among the experimental groups as compared with a 2 percent increase among the control groups), to think that sex with prostitutes increases the chances of getting AIDS (19 percent increase among the experimental groups as compared with 8 percent among the control groups), and to agree that proper use of a latex condom reduces the chances of getting AIDS (19 percent increase among the experimental groups as compared with an 8 percent increase among the control groups). Adolescent