Why bother with Logic Models and a Theory of Change?

I had a very interesting chat the other day with a researcher.  I won’t say who, and I won’t say where.  So essentially this could be part of an elaborate storytelling ploy.  

It centred on why we are using logic models and theories of change as the basis of our new evaluation strategy when they’re not perfect tools to understand change.  At least not when compared to deep, longitudinal ethnographic research.

As anyone will tell you, whether they are a secret evaluation geek or not (just me with my hand up?) evaluation can be a bit tricky.  The fact is human and societal change is complex and messy.  But if we want to progress our field, like any field of endeavour, we have to learn from what we have done to move forwards.The key for me here is that while social change can appear to be random, programme design is not.  Programme design is logical, it has aims, it has assumptions (often based on years of experience) about what works, both the process and practice. So, we need structures in place to help map, articulate and reflect.  Such structures need to help us recognise and test our assumptions about what we think works.  They need to help us communicate why we are approaching a social challenge in a particular way, both internally, and to others who don’t have a nuanced understanding of all the issues.  It means having an accessible coherent framework to do this across projects, and using this framework to understand what worked and what didn’t.  Such a framework needs to make evaluation easier to do, without making it less robust.

That is what logic models and theories of change aim to do.  They are useful tools for this. If I had to give you 3 reasons why we’re using them it would be that: they are accessible tools people can start to (and continue to) use easily to help evaluation, they are cost effective, and frankly, they do the job.  They are not perfect, but they are the best we have.

Nominet Trust is beginning to use logic models as a framework for monitoring and evaluation.  They help draw an outline of why a project has chosen to work in the way that it does and give a foundation of appropriate indicators for evaluation. For some of the projects that we work with, a logic model is enough.  For example, the project may be a continuation of a tried and tested model that we have previously invested in, or it may be a small investment that leaves minimal resources for extra evaluation approaches.  In either case they may not to need to engage with a theory of change or more complex evaluation processes.  

To support everyone to take this logic model snap shot, we have structured our application questions around the four key components of a logic model:

You won’t see this when you fill in an application form because we recognise that logic models are not the way that our brains work when you are telling the story of a project you want investment in.  So we’ve tried to keep the application questions in a format that corresponds to how you would naturally share your project.   But in the background each question corresponds to an area.  That way we can take an application and view it in a logic model as another way of understanding the process of a project.  At the moment we manually do this, in the future we will make it automated so we can share this with each applicant as a way of reflecting on their application.

However, for those projects that are testing a new area, working with untested assumptions, or operating a large scale, more complex project, then a theory of change facilitates a deeper mapping of the change process that goes on in a project.  I believe they are very valuable.  Already in the short time we have employed them at the Trust I have seen how they can be useful for an organisation’s staff to internally communicate the goals they have for their work (which individuals may not always share), can support project planning, and crucially, highlight the steps that need to be evaluated.  If our partner projects who want to develop their theory of change (which we recommend for larger or more complex projects) we support them to do this through our funder plus model.

Obviously both logic models and theories of change have limitations; they are just representations of what is going on in the real world.  They cannot possibly hope to express the true complexity of the societies they engage with.

But if we’re serious about developing our work and pushing to do better then we need a method and approach to test assumptions, to coherently map out and communicate why we think something will work, and to understand when it did or didn’t.  I started off this post talking about storytelling, which I think is crucial to the work we do. But every story has a structure, and while some people can work from memory and intuition, when intuition fails us or we need to share our work, we’ll be damn glad we wrote it down somewhere.