Abstract
The importance of considering wider contexts when evaluating the success or failure of programs has been increasingly acknowledged with the shift towards culturally responsive evaluation. But one of the important advantages of contextual approaches has been mostly overlooked-that they can provide more "realist" evaluations for why programs fail or succeed. The careful identification of causal mechanisms involved in program delivery is important for avoiding spurious conclusions about the effectiveness of programs. Drawing on findings from a mixed-methods study conducted in Western Kenya among the Luo to evaluate the impacts of a HIV prevention program involving voluntary male medical circumcision (VMMC), it is shown that the VMMC program was one of several variables that contributed to the desired outcome, being not so much the cause but a catalyst for accelerating the desired behavioral change that the surrounding context was already amenable to and contributing to, even before the program was introduced. The need for context evaluations is particularly obvious when programs are part of broader campaigns involving scale-up from one context to another.
