Nov 27, 2012


This comprehensive school support program aims to reduce HIV risk by supporting regular school attendance among a population of orphaned adolescent girls in rural Zimbabwe.  After 2 years of intervention, girls in the comprehensive support group experienced an 82 percent reduction in school dropout rates and were significantly less likely to be absent from school than control group girls.  In addition, girls in the intervention group reported significantly improved expectations related to future educational attainment and intentions to wait for sex because of consequences compared with control group girls.  Finally, intervention group girls experienced a 63 percent reduction in rates of marriage compared with control group girls.


Target Population: Orphan girls attending coeducational primary and high schools

This school-based, comprehensive attendance support program is designed to reduce future HIV risk by addressing barriers to school attendance experienced by orphaned girls.  The program provides fees, books, uniforms, and other school supplies as well as targeted support for attendance-related problems from trained female teachers.  These teachers, assigned approximately 10 girls each, are trained to monitor participants’ school attendance and provide guidance and assistance in cases of absenteeism.  The program is offered beginning in 6th grade, and girls continue to receive support through the 7th grade and into high school, which begins in 8th grade.  Some girls live with teachers during high school due to the long distance between their villages and schools, but this was not a part of the intervention design.  All study schools were provided a universal feeding program, and school heads and helpers (i.e., teachers assigned to orphan girls) received cash incentives of approximately $15 – $20 per term.  Also, the authors note that they plan to conduct more intensive cost effectiveness analyses.


Hallfors, D., Cho, H., Rusakaniko, S., Iritani, B., Mapfumo, J., & Halpern, C. (2011). Supporting adolescent orphan girls to stay in school as HIV risk prevention: Evidence from a randomized controlled trial in Zimbabwe. American Journal of Public Health, 101(6), 1082-88.

Evaluated Population: A total of 329 orphan girls attending grade 6 in 26 primary schools in rural eastern Zimbabwe participated in the study.   The average age of participants in both groups at baseline was 12 years (range: 10-16).

Approach: Researchers identified five rural geographical clusters, each with two high schools (1 Methodist, 1 government), surrounding Zimbabwe’s capital city.  Primary feeder schools within each cluster (n=26) were randomized into treatment and control groups.  One control group school was dropped from the study sample after randomization because it was discovered that several of the participating girls were not orphans.  Orphan girls were the unit of analysis, and generalized estimating equations were used to account for nesting within schools.  No significant differences were found between intervention and control groups at baseline.  Data on outcomes were collected at three time points (baseline, 12 months, and 24 months).  Self-report questionnaires were used to assess school absence; perceptions of adults’ caring; educational aspirations; future expectations; beliefs about gender equity and wife beating; attitudes toward sex; and sexual debut.  School records and reports from school staff and classmates were used to collect data on dropout, marriage, and pregnancy.

Results: The program had significant positive impacts on school dropout and marriage at the two-year follow-up.  No results were reported for the pregnancy outcome because only two pregnancies were reported during the study period (both in the control group).

Girls in the intervention group experienced significant reductions in school absence.  They also were more likely to expect to graduate from high school and to graduate from college than control group peers.  Findings related to girls’ expectations about making enough money by age 30 were marginally significant in favor of the intervention; however, girls in the intervention were no more likely than control group girls to expect to live to age 35.

Girls in the intervention group reported a marginally significant increase relative to control group peers in the perception that teachers were caring, but no impacts were reported for perceptions of adults in families or the community.  The only significant impact reported for sexual attitudes and behaviors was an increased likelihood of delaying sex because of consequences among girls in the intervention group relative to peers in the control group.  The intervention had no impacts on girls’ beliefs about asking one’s husband to use a condom; having sex as an adolescent; waiting for sex until marriage; or waiting for sex because of values.

Attitudes toward wife beating did not change as a result of the program; however, the intervention did result in a slightly stronger endorsement of gender equity among intervention girls compared with control group girls.



Hallfors, D., Cho, H., Rusakaniko, S., Iritani, B., Mapfumo, J., & Halpern, C. (2011). Supporting adolescent orphan girls to stay in school as HIV risk prevention: Evidence from a randomized controlled trial in Zimbabwe. American Journal of Public Health, 101(6), 1082-88.

Contact Information

Denise Hallfors, Ph. D.

Pacific Institute for Research and Evaluation

1516 E. Franklin St., Suite 200

Chapel Hill, NC 27514


Adolescents, middle school, high school, female only, rural, school-based, attendance, academic motivation/self-concept/expectations/engagement, high school completion/dropout, sexual activity, other reproductive health, family structure/marriage, other relationships

Program information last updated on 11/27/12.