Editorial Comment:

The purpose of a realistic evaluation is to establish whether there is an ‘inequivocable causal relationship between a program and its outcome’. That is, where some change can be measured following the installation of a particular programme, it seeks to establish beyond doubt that it was the actual programme, which caused the measurable change, and not some other, unidentified variable.

Realistic evaluation assumes that there is an underlying theory behind the workings of a particular programme or intervention. This theory explains how the particular programme caused the measured change.

The most important aspect of the realistic evaluation is the overall context in which the programme takes place.

In order to find the underlying theory, it is necessary to configure a series of CMOs for each intervention. A CMO has three constituent parts: a context, a mechanism and an outcome.

‘The context’ signifies the precise circumstances into which a particular intervention is introduced. The mechanism is the precise way in which this measure works within the given context to produce a particular ‘outcome’.

The precise make-up of the context will greatly influence whether a mechanism can work or not. For a programme to work the contextual conditions must be right for allowing mechanisms to ‘gel’ or ‘fire’ to produce a defined outcome.

Each CMO forms the basis of a ‘mini-experiment’, which needs to be measured empirically.

Through measurement of a series of CMOs it should be possible to deduce the features of contexts that allow difference mechanisms to work to achieve particular outcomes. Thus ‘transferable lessons’ may be learned.