Hi all,

 

I think there is a solid case to be made for what has been coined “methodological individualism” in Realist Evaluation -- specifically in terms of the definition of mechanism as ‘how people respond or react to resources’. We may ask:  ‘should Realist Evaluation focus mechanisms on the individual-level agency of actors involved in programmes? Or should we be also looking at the emergent properties they create and mechanisms at different levels of a system?’ The debate, in my opinion, should be pinned around questions of pragmatism. What kind of analysis will lead to ideas for the resolution of complex entrenched problems? Perhaps “methodological individualism” as it has been coined, is adequate in many instances for these pragmatic goals – with some exceptions.

 

The case to be made for “methodological individualism” is not in contradiction to an appreciation of emergent properties. Acknowledging the emergence of properties within complex interventions is to appreciate the complexity of reality itself. But if we see mechanisms as what happens in the hearts and minds of people who design, manage, implement and receive interventions – the argument here is that this provides the kinds of insights needed for assessing and innovating the design of programmes. Pragmatism is absolutely key in Realist Evaluation.

 

Realist Evaluation also proposes alternatives to ‘causation via counter-factual’ by illuminating the inner workings of the human element of programmes. Humanizing our understanding of intervention functioning has been Pawson and Tilley’s incredibly valuable contribution to evaluation science. And although that appears to suggest that Realist Evaluation places an over-emphasis on agency of actors, this is not the case. The activity of Realist evaluation is not about isolating mechanisms but rather understanding context-mechanism interactions. There is freedom in using the CMO heuristic to account for the laminated layers of a complex system. Practically speaking however, baby steps into complexity theorizing is needed for many of us - with the first steps being around how individuals buy-in/or not  - this is a very good path forward for many people and may be adequate.

 

Take the following example: Say an evaluator is studying interventions to address gender pay inequity across organizations. Say, hypothetically speaking, that it has been found that women on average are paid less than their male counterparts for jobs that have equal, equitable or even greater roles and responsibilities. Just as with the example of how hydrogen and oxygen when mixed create an emergent property of ‘wetness’, it could be theorized that the mix of differential salary amounts across genders for same work in the workplace creates an emergent property of ‘toxic work environment’. This ‘toxicity’ may be defined as friction amongst staff, feelings of resentment, feelings of being under-valued, feeling entitled, feeling guilty, feeling trapped in gender roles and expectations, – and so on. This may lead to outcomes such as reduced communication between staff, increased inefficiencies and errors, social anxiety and reduced morale and productivity. Now the point here is that even with the emergent property of a ‘toxic work environment’, in order to make pragmatic sense of this and develop solutions, the analysis cannot remain at the level of ‘toxic environment’ even if the concept is at some level useful. The analysis has to go back in to how people think and feel – how employees feel about negotiating their salaries, how managers feel in regard to negotiating across genders, how department leads feel about gender equity sensitivity training, how employees feel when they receive such training etc. etc. So while still retaining an appreciation of the complexity inherent in realist accounts of ontology, “methodological individualism” as it has been coined can be a very valuable emphasis that Realist Evaluation can push forward for understanding where pragmatic innovations can occur in the system to improve conditions and resolve entrenched problems.

 

Justin   

 

Justin Jagosh, Ph.D

Director, Centre for Advancement in Realist Evaluation and Synthesis (CARES)

www.realistmethodology-cares.org

&

Honorary Research Associate

Institute for Psychology, Health and Society

University of Liverpool, UK