Experimental evaluation of school-based HIV programs in sub-Saharan Africa

School-based adolescent health education programs represent a durable strategy in reducing the spread of HIV because they can leverage pre-existing social and organizational structures to reach large fractions of students at critical life stages. Many evaluations of school-based HIV programs draw on multilevel study designs that assign schools to treatment conditions or assign students to treatment conditions within blocks defined by school membership. Under multilevel designs, the power or probability with which a design can detect an effect is proportional to the degree of clustering among students within schools. A priori knowledge of plausible values of this clustering is necessary to ensure the effective and efficient design of a study. The purpose of this study was to improve the design of school-based studies of HIV education programs in sub-Saharan Africa by compiling empirical estimates of this clustering using representative samples in 15 countries.