Evaluation January 2008 [antikvár]

Editorial There is an increasing flow of international evaluations that derive from the reálist school of evaluation, in particular building on the work of Ray Pawson and Nick Tilley. In this issue, Liliana Leone adopts a 'reálist' approach to the evaluation of an Italian anti-drugs programme aimed at young people - a programme that combines deterrence/threat with dissuasion of future drug use. Within a national legal framework, a new pilot programme was introduced at a régiónál level alongside an established programme. Part of both programmes is a 'mandatory' interview with those apprehended by the police. Failure to attend such interviews triggers further 'administrative' sanctions - for example, withdrawal of passport or driving licence. In somé cases a treatment programme can alsó be required. In the 'pilot', the mandatory interview takes place in a youth centre or a health unit; in the established programme, the interview takes place in the Interior Ministry offices. The evaluation carefully unpicks the programme 'theory', with formulated hypotheses and evaluation questions focusing both on the 'dissuasive' effects of the programmes and on the significance of the location (context) where the interviews take place. The author goes on to identify a differentiated effect on different subgroups of affected young people, and by identifying CMOs (Context + Mechanism = Outcome) configurations, she alsó identifies the different mechanisms at work. In her concluding reflections, Leone reiterates the strength of the reálist approach in identifying mechanisms, notes the limitations of what she has been able to achieve in this single evaluation and sees the advantage of 'reálist synthesis' that Pawson first articulated in his article in this journal in 2002, Volume 8(3). Variants of programme theory, which for somé include 'reálist' approaches, centre on representations of programme logic, usually through simple, linear diagrams. This simplification of what is commonly agreed to be a complex, 'messy' world has often been seen as the Achilles heel of the programme theory approach. Patrícia Rogers asks: 'Is it a problem to represent reality as a simple causal model of boxes and arrows, or should the logic models we use address the complexity of life - and if so, how?' Rogers uses Glouberman and Zimmerman's distinction between 'simple','complicated' and 'complex' and considers how complication and complexity are manifested in different interventions and their evaluation. She identifies and exemplifies 'aspects' of the 'complicated' related to 'governance', 'alternative' or 'simultaneous' causal strands, and aspects of 'complexity' related to 'recursive causality', 'tipping points' and 'emergence' - the latter all concepts deeply rooted in systems thinking and the analysis of complex adaptive systems. Finally, Patrícia Rogers notes that the use of programme theory models has to go beyond diagrams, however complex these might become: they alsó require