What use is a theory of change? 6 benefits, and some things to avoid.

Whether in the back of a 4×4 in Tanzania, or in seminar rooms in Oxfam house, I seem to spend an increasing amount of my time discussing theories of

Not recommended.....
Not recommended.....

change. Oxfamers seem both intrigued and puzzled – what are they? What are they for? The answers aren’t simple and, as social scientists like to say, they are contested. But here’s what I currently think.

What is a theory of change? A way of working and thinking, and a set of questions. Aerobics for the imagination – not a form to fill in (and most definitely not logframes on steroids). Nor is it a typology or (a personal bête noire) an insanely complicated diagram that no-one coming after you can understand (see example, right). More here.

How does (or should) a good theory of change improve our work (or ‘add value’ as the marketing wannabes insist on saying)?

  • It encourages deep observation of the system – how power is distributed; how decisions are made; what are the coalitions for and against any given change; how is change likely to happen in this system. The longer you can refrain from the jump to ‘so what do we do’, the more likely you are to come up with some good ideas to test.
  • It makes you think far more about critical junctures/windows of opportunity as an essential component of any change process. These are almost uniformly absent from our pre-intervention plans, and yet when you watch a change process taking place, they often play a primary role. They are often then airbrushed out again in the subsequent rewriting of history.
  • It highlights the importance of working with ‘non-usual suspects’, brokering discussions between groups who may initially have low levels of trust, (even hostility), but who can come together and find new answers to old problems. In Oxfam, faith groups and traditional leaders often get overlooked, but there are many others (discussing potential places of discussion on social issues in rural Tanzania we came up with kiosk owners, soccer fans, women attending hairdressing salons and loads more).
  • It helps you identify and open up the ‘black boxes’ in our thinking – intellectual leaps and assumptions, like the recent discussion with Twaweza on their initial assumption that access to information would be enough to trigger citizen action.
  • It shifts the emphasis away from a huge exercise in pre-intervention planning, followed by a long and unremitting process of implementation (with a mid and end term evaluation thrown in). Instead we are into something more experimental and iterative – come up with some initial hypotheses to test, but the smart thinking will be around taking stock and correcting courses as the project evolves. This is true because our understanding of a context improves as the project evolves, requiring adaptation (Oxfam programmes such as Raising Her Voice have really benefited from a theory of change discussion a couple of years into their work), but also because high levels of staff turnover mean that in any case, we need to constantly re-learn the theory of change behind a programme.
  • Linked to this, a good theory of change should shift the centre of intellectual engagement from monitoring to learning (see Claire Hutchings’ recent post).

complexity signAll of these points apply both to programmes that are already under way, as well as to ‘clean sheet’ discussions on new work.


Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


8 Responses to “What use is a theory of change? 6 benefits, and some things to avoid.”
    Hi Duncan, I am a fan of the approach, but but have never liked the term ‘theory’ (def: A set of statements or principles devised to explain a group of facts or phenomena, especially one that has been repeatedly tested or is widely accepted and can be used to make predictions about natural phenomena)- ‘theory’ is normally used to describe something that is proven beyond reasonable doubt – in Mathematics beyond all doubt. Instead I would suggest we should use the term ‘hypothisis of change’ – describing something that is suspected but may not be entirely correct – as you say in your article.
    Cheers, Tim

  2. Aldo Matteucci


    take out of the last drawer of your desk the photo of your graduation. Your’re different now – you’ve changed. When did you change? Change is silent, yet relentless – something Western thought has always rejected as contrary to the “principle of non-contradiction.”

    Your (mis)use of the term “theory” betrays your basic conceptual difficulty in framing what can’t be framed.

    I’ll give you the “uncertainty principle of change” instead: you can either go for direction, or process,but never for both.

    Please note that our social sciences hardly discuss processes and “path-dependent outcomes” – they are seen as “acts of a(malevolent and bureaucratic) god” that can only destroy the timeless plan.

    Now the Sinic civilization always focused on transformation and had no use for categories and divisions. They focused on process through time – the evolution of the situation – and left direction to Heaven. In this view, personal commitment is more important than the plan.
    Wu wei – effortless change, is the way to go.

    Francis I, the new Pope,uses a term which, for all the “theorizing” had been lost: DISCERNMENT. Discernment is the application of guidelines to the situation. And between guidelines and situation, the situation – social and material reality – always win.

    All you say is right, of course.
    Just use “discernment” instead of “theory”. Every time you use the word, you reinforce what you wish to weaken.

  3. Kevin Savage (@KevinSavage11)

    I think the most important question for programmers in the 4×4 is to what extent should we adopt a ToC approach in replacing what might be called strategic planning or business plans and the logframes that flow from them, and how to do it. So, how does it ‘add value’ but how much value in comparison to what we have already to answer how much should it replace what we have now?

    The old style logframe approach I suppose had an underlying assumption of an extremely simple theory of change, or simple cause-effect logic, but most importantly gave all the people responsible a simple plan and understanding of how to make decisions and implement interventions and it continues to fit with all the ways organizations work and behave and internally distribute power that affects how things are done.

    How do we turn the rhetoric of this or say too the ideas that Ben Ramilingam is discussing in Aid on the Edge of Chaos (though I’m on only at Chapt. 4 so maybe he’s got some ideas of how), so that organizations and the people in them, who have real constraints and reasons for their behaviour and tool choices, to be “more experimental and iterative… taking stock and correcting courses as the project evolves”.

    How do we get aid organizations to really use ToC and allow the people in them to really be experimental and iterative? And how do we explain that it isn’t ‘just muddle through any old way’ (obviously it isn’t but that’s how those used to planning may feel about being iterative and experimental).

    I think there’s still much research to do on organizational change and behaviour in aid agencies specifically.

    Thanks for the blog post. Really good thoughts to help people in explaining ToC.

  4. Agree with most of this. I strongly sympathize with this idea that ToC should be a ‘way of working and thinking, and a set of questions’. The problem, from what I can tell, is that it simply isn’t that for the majority of people who work with it.

    Putting aside the semantic issue of using the word ‘theory’ (which is a big issue I know you are wary of, preferring ‘how change happens’), the way in which Theories of Change operate in development policy and practice appears to be already in linear, story-telling, donor-pleasing territory, often characterized by simplistic if…then statements.

    A friend recently told me how their organization contracted out the ToC element of their M&E. It was done well and it was in-depth. In that sense it was a ‘good’ Theory of Change. But it had no effect on their work. It was not translated into French for the programme staff in the country where the programme was implemented (!) and DFID didn’t ask for revisions as the programme went on. What’s the point of that excercise?

    I would say Theories of Change don’t necessarily encourage any of the things you mention. One of the few consistent things identifiable across the diverse but sparse guidance literature is that Theories of Change should ‘question underlying assumptions’. But the 6 (very good) points you make aren’t really part of the Theory of Change discourse.

    My concern is that, given the lack of overall clarity of what a Theory of Change ‘should’ be, different actors will use Theories of Change in whatever way suits their pre-exisiting model of working. Those who are comfortable with more linear and tool-based models of working and thinking will just complete the ‘form’. Others like Oxfam may use it to ask the ‘right’ questions but do we just end up in the same place overall?

    What I want to know (and am researching this with the Asia Foundation right now) is can the introduction of Theories of Change into the development industry actually improve policy and practice or not? Even if it is given to people as a rigid tool, does it create informal conversations behind the document which are useful? More research needs to be done on this, but its clear to me that moving beyond this ‘complete the form’ framework at least requires a genuine desire from donors to give space/time/money for learning and reflection to take place. There needs to be more incentives for people to work in this way.

    Overall I agree with your framework of how to think about ‘how change happens’ but we need to be hearing it from those who make Theories of Change mandatory in their reporting requirements.

  5. Forgot to do some shameless sharing (plugging) of the work by our Justice and Security Research Programme and Asia Foundation collaboration on ToCs.

    Some of the arguments above and more are debated in this literature review of ToCs in international development here: http://www.lse.ac.uk/internationalDevelopment/research/JSRP/JSRP%20Papers/JSRP-Paper-1.aspx

    and these blogs here: http://blogs.lse.ac.uk/jsrp/2013/08/27/can-theories-of-change-reflect-the-realities-of-international-development/

    and here: http://blogs.lse.ac.uk/jsrp/2013/05/28/reflections-on-theories-of-change-in-international-development/

    We would welcome comments and debate, look up @JSRP2 and @craigvalters on twitter.

  6. Carl-Gustav Linden

    I believe the fifth and sixth points are the most relevant as change is about learning and not being locked into a theory about the world. That vision includes to many assumptions about how people should live their lives. The instance we think we know how the world works is also the moment when we stop adding to our understanding. I struggle to see any great ideological difference between modernization theory (despite its Cold War context) and theory of change and tend to think that they are basically the same – with some flavours added.

  7. I too have found theories of change very useful in my work and in work I and my colleagues at CDA do with others reflecting on how to enhance the effectiveness and impacts of their programming. And I would agree that theories of change, in themselves, do not encourage the very good processes you identify as benefits. But theories of change can, at a minimum, encourage people to think about change at all and encourage them to connect their programming to an analysis of the context. In my experience in the peacebuilding and conflict resolution field, people often do not formulate their goals in terms of changes they would like to see, and often don’t connect their program decisions to an analysis of the broader situation. The very process of thinking about change and how the programme can promote change can improve the relevance and appropriateness of the programme to the context.
    How do you know a good theory of change? We have tried to identify some parameters of good theories of change – ones that include the kind of questioning, out of the box thinking, and adaptation Duncan’s 6 points underline.

    * It has a clear conceptualization of impact and pathways to it: in other words, it makes explicit the intended changes and the assumptions about the “road map” to change, including, for peacebuilding, a clear connection to key drivers of conflict or peace.

    * It is coherent: it is logical and shows how the effort will lead to desired changes without big leaps or gaps.

    • It is plausible: beliefs and assumptions about how change will come about have been articulated and explored—and there has been some challenge of comfort zones in thinking them through. Good theories of change should reflect research results and learning from past experience.

    * It is grounded in context: it takes context as the starting point and reflects the reality of change processes in that setting. This implies, of course, that the context has been analyzed deeply and that the analysis is continuously updated as the situation changes and as more is learned.

    * It is testable: it is specific enough to be tested for validity over time, and a process is in place to test whether the assumptions and conclusions in the theory of change are correct.

    * It is dynamic: in other words, it changes as the situation changes or as more is learned about whether and how the desired changes are coming about.

    Not a panacea, of course – but perhaps having some criteria for what a robust theory of change is can help counteract the box-ticking, form-to-fill-in approaches to theories of change.

    CDA has been involved in developing guidance on theories of change for peacebuilding work—more for those just starting to use them:

    Practical Approaches to Theories of Change in Conflict, Security and Justice Programmes (Part I), http://www.cdacollaborative.org/media/89738/Practical-Approaches-to-Theories-of-Change-in-Conflict-Security-and-Justice-Programs-Part-I.pdf

    Theories and Indicators of Change: Concepts and Primers for Conflict Management and Mitigation, http://www.cdacollaborative.org/media/89750/THINC-Full.pdf.

    Comments always welcome–www.cdacollaborative.org.