Integrating Process with Outcome Data in a Randomized Controlled Trial of Sex Education

Abstract

Integration of process with outcome evaluations has been recommended as a way of addressing some of the criticisms of Randomized Controlled Trials (RCTs) for evaluating social interventions. This is the second of two articles (the first was published in issue 10(4) of this journal) discussing the design, implementation and analysis of a process evaluation as an integral part of the RIPPLE (Randomized Intervention of Pupil Peer-Led sex Education) Study. The methods for integrating process and outcome data are outlined and findings are presented to illustrate how integration of process and outcome data enabled exploration of: 1) variation between schools in the extent and quality of the intervention implemented, and any impact of this on outcomes; 2) the processes by which the intervention might affect outcomes; and 3) the extent to which different subgroups of students or schools may have responded differently to the intervention. The article concludes with a discussion of the methodological issues arising from this attempt at integrating process and outcome evaluations within the design of an RCT.