Logic, logic, logic. Effective programmes are underpinned by a well thought out argument, supported by evidence, that explains why an intervention will achieve a small number of outcomes, and to what extent.

Good logic is like the infrastructure of a building. Strong infrastructure can seem expensive and take a long time to assemble.

Not surprisingly, many children's services organisations, and some governments, the UK included, look for short cuts and efficiencies in delivering these programmes.

So the question arises, what is the impact of implementing say 30% or 50% of a proven model? Do you get 30% or 50% of the impact on outcomes, or maybe a little more.

The answer is depressing. The probable effect is zero. There is much about proven models that can be altered. As Mark Greenberg explained, PATHS establishes objectives for teachers and children. Teachers will achieve these objectives in different ways, but to be successful they must achieve the objective.

But changing the logic model is dangerous. Take away one or two of the girders of a building and it will likely fall.