What questions help us understand how change happens?

change ahead road signHow do we analyse the stories of change that we all use in development? Such stories shape narratives, illustrate approaches and enrich our understanding of how change happens. Regular readers of this blog will know that this is a running theme, but I’m now about to step it up, working with colleagues across Oxfam and beyond to collect and use case studies of change to sharpen our thinking and practice. What emerges when you do this is the problem of ‘retrospective coherence’. Asked to remember what happened, people rearrange and reinterpret a change story. Typically they downplay the importance of failure and unexpected events, and the role of individuals (eg champions within state institutions). They also tend to minimize the role of actors outside the civil society-state interaction – faith leaders, academics, media, private sector, traditional leaders. What remains is a smooth, well-planned and executed project that bears little resemblance to the messy reality faced by people working in real time. So part of the effort in collecting such stories is to recapture what actually happened. I’ve got case studies coming out of my ears at the moment – working with Oxfam Novib, in East Asia, and with the campaigns and advocacy team – and will be blogging about them as they develop. But in the meantime, here’s the latest version of the guidance questions I send round to kick off the process – I would really appreciate any suggestions for sharpening them up, references etc. They’re also available as a Word document here.

Starting Point

What change did Oxfam seek? Where/how did the idea originate? Was it specific (eg improving livelihoods for X women) or systemic (changing government policy, prevailing norms)? Was it primarily economic, political, social or a combination? Power and Change cycle The remaining questions help you work your way round the power and change cycle, which helps in analysing a wide range of change processes (see graphic)

Power Analysis

What was the nature of the redistribution of power involved in the change? Was it primarily about ‘power within’ eg empowering women to become more active social agents, ‘power with’ (collective organization) or ‘power to’ (e.g. supporting CSO advocacy)? What was the power analysis of the key forces driving/blocking such a change? What economic or political interests were threatened/promoted by the change? Which groups were drivers/blockers/undecided? Was their power formal (eg elected politicians) or informal (traditional leaders, influential individuals)? Was it visible (rules and force) or invisible (in people heads – norms and values) or hidden (behind the scenes influence) Which individuals played key roles, either as allies or opponents?

Change Hypothesis

What aspects of (or changes in) political, economic, social context made the desired change more or less likely (eg functioning institutions, political leadership, new technologies, new threats or opportunities) What was the hypothesis for how the change was likely to come about? What alliances (eg with sympathetic officials or politicians, private sector, media, faith leaders or within civil society) could drive/block the change? What tactics were likely to work best (cooperation v conflict, research v street protest)? What were the pivotal moments/windows of opportunity (eg new governments; changes of leadership; crises and scandals; election timetables)?

Change Strategy

What was Oxfam’s role in promoting change? As an active player or supporting partners? One programme approach, or advocacy/programme only?complexity sign Who were our partners – were they ‘usual suspects’ (local civil society organizations and NGOs), ‘unusual suspects’ (private sector bodies, local/national government, faith leaders) or a mixture of both? What was Oxfam’s contribution eg helping them develop a clearer theory of change; bringing partners together with other actors to build alliances; building particular aspects of their organizational capacity; funding?

Implement and Evaluate

What did we/partners actually do (as specific as possible, please!) What was unexpected? Few change processes go according to plan (although we often rewrite them to make them look that way!) What unforeseen events or realizations (e.g. that something wasn’t working) led to a change of approach? How did the original plan change as the work developed? Were there unintended outcomes and impacts? Were there early wins that helped build confidence and momentum in the work? Looking back, what would you have done differently? How did you monitor and evaluate impact? What evidence can you provide to persuade someone who questions whether your actions actually led to the change described? What are the top lessons you would draw from this experience for development workers in other contexts?]]>

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.


21 Responses to “What questions help us understand how change happens?”
  1. Very comprehensive list Duncan.
    Would like to suggest that you specifically ask for “unintended negative impact (harm)” – both in the course of implementation or post implementation. This is an area, imo, we often neglect and / or don’t want to know about.
    While need to promote a culture that allows us to fail, so long as we know we are failing and are learning from it, we more urgently need an appreciation of the fact that we do cause harm (hopefully unintentionally). I am trying to distinguish ‘harm’ from ‘strategic failure’.
    Let us get talking about this in some of the case studies you are building.

  2. Kate

    Interesting work and look forward to reading the cases.
    Will you be looking back at any earlier examples to see how things have played out over time? Recent comment on a 2009 Malawi example (http://www.oxfamblogs.org/fp2p/?p=313) suggests things might not have gone as well as expected – it would be interesting to understand that more.
    Also, will the analysis just use interpretations by Oxfam staff, or will you also talk to other organisations that were involved? (I realise that would take more time and money!) Conversations with colleagues in Malawi suggest they emphasise different aspects of the domestic violence bill campaign. It would be interesting to understand the views of different organisations on the advocacy processes and key factors in success (or failure).

    • Duncan

      good challenges Kate, another feature of stories of change is the almost complete absence of evidence of impact, attribution etc, so that’s clearly got to be a part of the work

  3. sounds interesting. will be tracking this one.
    One question to add would be ‘what did we plan to but didn’t do in the end, and why’?
    Our development programmes tend to have detailed plans and strategies yet in practice we don’t implement all the plans, and often end up doing what we didn’t plan in the first place.
    I heard Michael Quinn Patten talking about this and he referred to a model from Henry Mintzberg that i found really helpful.
    There’s 5 types of strategy:
    1) the intended strategy – what we planned to do
    2) the deliberate strategy – the bits of the original plan that we actually did
    3) the unrealised strategy – what didn’t get done, for whatever reason
    4) emergent strategy – thinks we did that weren’t in the original plan
    5) the realised strategy – what actually happened in the end.
    the emergent strategy is obviously very interesting. This is the space for creativity and responding to changes in context.
    But is the unrealised strategy even more valuable? We tend to carry on making the same mistakes. But if we were more explicit about what we planned to do but didn’t do, would that help us to see what doesn’t work? Then allow this to address our blind spots?
    I’m really into the idea of LEARNING to improve performance, and this question of the unrealised strategy has been a very valuable discussion starter and eye opener.
    here’s a link to a Mintzberg-esque diagram from IBM.

  4. Paul O'Brien

    Thanks Duncan. Great questions and enjoyed (and agreed with) the set up.
    Didn’t like your power analysis questions however. At the risk of abstraction (precisly my concern) they are all about power as “capacity”. The power that matters in our game is power as “control”. Or power “over”.
    The two things we need from any power analysis are (1) who controls the key decision/resource, and (2) what’s it going to take to change either their control or their decision making.
    We need to be better at and naming, measuring and then changing the political calculus behind policy decisions. I think there may be interesting proxies for measuring the politics around issues, and then our collective impact on those politics.

    • Duncan

      Hi Paul, you’re right, for those activities aimed at influencing policy or the behaviour of those in authority. But often, we are talking about ‘power within’ and ‘power with’ – sense of rights and dignity, and desire to take collective action. The questions try to get at that too – maybe too big a stretch to cover all those kinds of power in one go, but that’s what I’m starting off with anyway

  5. Tracey Martin

    Hi Duncan,
    Some good questions here and great to see Oxfam looking critically at stories of change and using them to inform practice.
    Of course, the best way to reduce selective memory is to reflect and document change processes as they go along – then you have evidence to look back at at the end.
    Failing that, using the questions in a facilitated process involving wide range of stakeholders/staff is more likely to give a full picture – people remember different things and interpret them differently.
    This also enable use of visuals – timelines, maps of people involved and their relationships – which can be built on as the story unfolds.
    The question ‘What did we/partners actually do (as specific as possible, please!)’ is helpful to start with – too often we rush straight to analysis (automatically deprioritising some information) – going through what happened in detail (without analysis) makes sure that important elements are not left out and can be used as a check when you move on to analysis – you said that X was an important player, in what activities and what did they actually do?’
    The action learning cycle is a useful tool and other ideas for enabling people to tell their stories can be found at http://www.barefootguide.org/index.php/component/content/article/39-download/255-the-bfg2-companion-booklet-designing-and-facilitating-creative-learning-activities p.18 and p. 23.

  6. David

    Hi Duncan, I assume you’re familiar with the Global Giving story-telling method, but it strikes me that one of the more powerful aspects of it (where it interviews numerous community members rather than implementing staff) is that it asks people the neutral question “what changed?” It is similar to the Most Significant Change methodology in that way. Both help to avoid a retrospective bias by changing what the topic of the questioning is.
    Toward that end, I wonder if you might precede your questions on “what did we/partners actually do” with a broader “how did the situation shift in the following months/years?” which would anchor the narrative to a broader discussion. It also allows you to probe as to what critical windows opened up and how the effort shifted to take advantage of/mitigate them.

  7. Gabrielle

    Great initiative Duncan. We often think of change from where we sit and what we do, and have a blind spot for other factors & actors and the history of struggle and change that’s come before. It would be good to ask people about these other factors, and their relative contribution, in light of these.

  8. Hi Duncan! Long time.
    Very impressive article. This is from one M&E professional to another.
    I hope you are relieved to come out of a prolonged cold winter which I understand affected Britain’s agriculture. For the first time in several decades, I understand Britain has turned net importer for wheat and for several other food items.
    As you are probably aware, climate alarmism is now in full retreat with all leading global temperature datasets and the UN IPCC admitting that there had not been any statistically significant global warming for more than 17 years. By their own null hypothesis, an absence of warming for 15 years falsifies the theory of anthropological global warming.
    From a M&E context, what had been your theory of change that led Oxfam to adopt climate alarmism as a policy?
    Is Oxfam still retaining climate alarmism as a policy? If so why?
    If no, why is there no formal apology from Oxfam for misleading the public and wasting of funds? What does it mean to the credibility of an organization and how policies are made internally?

  9. Just a short comment/question.
    What would be your response, if I noted that the framework seems to portray change and ‘reality’ in a fashion where there’s a static condition A, and then a static condition B, where process is something happening in between?

  10. Great piece, and very useful framework; thanks very much for sharing this.
    Curious to know, within Oxfam, to what degree these questions are considered and planned around in the earliest stages of an initiative. So would your case studies have an initial ToC, power analysis, etc to return to for reference in completing this? Or will these be reconstructed retroactively?
    Katie McKenna
    Knowledge Impact Team
    Justice and Security Research Programme

  11. Mark

    Great article, as always. My only specific suggestion is to ask who had input into defining the problem or issues that you were trying to impact? And did anyone disagree this was an important problem to work on? Often, community members are not asked what they perceive of as large problems and therefore aren’t invested in prioritizing or investing in the changes being made.
    However, in general, it sounds like you are trying to describe change and intervention within multiple overlapping systems. That said, I think you might want to check out some of the guiding questions on pp. 315-316 that Peirson and colleagues elaborated in this paper. Many of them would need to be reworded if they were to be disseminated, but they might be useful.
    Peirson, L. J., Boydell, K. M., Ferguson, H. B., & Ferris, L. E. (2011). An ecological process model of systems change. American Journal of Community Psychology, 47(3-4), 307-321.

  12. Chris

    Another great post, Duncan.
    Slightly off topic, but the Barefoot Collective are hosting a writeshop conference in Johannesburg in September of this year on this topic in order to write their next book: “What really works? Sharing and developing effective approaches to social change”.
    If you’re free you should definitely participate, I’m sure they would love to have your input and it would be great to have some of your ideas and experience included in their next guide.
    Here’s the link with more info:

  13. AndyB

    Interesting stuff as always. Two thoughts from someone who has tried to ask similar questions of my own colleagues.
    * I’m afraid the question, “What change did Oxfam seek?” might be big too big and abstract for some. I’ve found it helpful to include an element of time, e.g. “If this works, what would be different in the social context from Time 1 to Time 2.”
    * There’s not much in the Implement and Evaluate Section that deals with whether the change actually happened. I know indicators are a quagmire, but I would ask something along the lines of, “During the course of the program, how were you assessing whether the change was taking place.”