What we are all on about - representing causal models April 19, 2009
Posted by Paul Duignan in : Outcomes systems architecture, Use of terms, Doing evaluation more efficiently, Strategic planning, Using the approach, Communicating outcomes models, Outcomes models , trackbackWhether we know it or not, a lot of us in evaluation, monitoring, social programs, philanthropy etc. spend a lot of time working with ‘causal models’. We call them all sorts of things - program justifications, rationales, program activities and objectives, logic models, logframes, intervention logics, strategy maps etc. - and most people who work with them don’t think of them as causal models. But that’ s what they are if we see causal models as just being an attempt to set out ‘what it is believed causes what in the world’. In the case of a program, the model is going to be a model of the steps which you think a program needs to take in order to cause high-level outcomes to occur. We really should get our heads around the best way to represent such models because at the moment I think that there is a great deal of wasted talk and effort about all of this. And it distracts us from getting on with the job of implementing good programs as fast as possible. Every dollar or every hour spent on struggling with an inefficient way of representing our program is a dollar or hour wasted.
I’ve just put up an article on the Outcomes Theory Knowledge Base on Causal models - how to structure, represent and communicate them. This sets out in detail my thinking on the topic. If you are not feeling in an intense mood and just want to find out how to work most efficiently with stakeholders when building such models as visual models then just check out the article on Simplifying terms used when working with outcomes.
In the causal models article I’ve tried to take us back to basics and ask the question - of the various ways of representing causal models (narrative text-based, tables, visual diagrams and mathematically) which is the best one to use? In the article I argue that visual diagrammatic methods (supplemented by mathematical models where possible) are superior because they allow us to make, and communicate, a richer picture of the causal processes occurring in the real world.
The great thing about this approach is that it not only provides a richer model of the world, it is also an easier approach to use when working with stakeholders and easier to communicate, why spend your time struggling with other more inefficient ways of working? Anyway, if you have a moment, check out either or both of these articles and leave some comments on them.
Paul Duignan, PhD
Outcomes and Evaluation Blog (OutcomesBlog.org)
Comments»
no comments yet - be the first?