There is discussion on the most appropriate research methodology to examine the efficacy of Family Group Conferencing (FGC). Randomised controlled trials (RCTs), despite their pitfalls, are considered by many to be the ‘golden standard’ but the argument is not compelling. In this paper, the theory on programme evaluation is discussed which offers an alternative methodology to study FGC. It is argued that reaching a comprehensive image of truth in the social sciences is never within reach. A RCT is an abstraction of reality, it only provides a partial image of the complex reality of families and the impact that FGC has on safety issues and the quality of their lives. Moreover, the rigour of a study depends heavily on the researcher's interpretative skills. In studying the efficacy of a complex intervention, such as FGC, it is a challenge to provide a valid and reliable picture of its impact. The context of such a conference, where the lifeworld of families constantly interacts with the system world of professionals, is characterised by multiplicity, polyvalence and interference. The methodology used to examine the efficacy of FGC should meet this ‘interplexitiy’.