Most programmes can say what they will do and what they hope to achieve. Far fewer can explain, step by step, why the first should lead to the second. A theory of change is the tool designed to close that gap.
This guide explains what a theory of change is, how it differs from a logical framework, how to draft one, and what the evidence says about where it helps and where it falls short.
What a theory of change is
The term was popularised in 1995 by the evaluator Carol Weiss, in work for the Aspen Institute’s Roundtable on Community Change. According to the Center for Theory of Change’s account of its origins, Weiss argued that complex programmes were hard to evaluate largely because the assumptions behind them were poorly stated.
Her answer was to make those assumptions explicit. A theory of change maps the sequence of changes expected to lead to a long-term goal, and states why each step should lead to the next.
The United Nations Development Group’s 2017 theory of change companion guidance describes it as a method that explains how interventions are expected to lead to a specific development change, drawing on a causal analysis of the available evidence. It asks that assumptions and risks are clearly analysed and spelled out.
Why it matters now
Theories of change are no longer optional extras in international development. The UN’s internal guidance for Cooperation Frameworks, first issued in 2019, requires country programmes to rest on “a clearly articulated, evidence-based, robust theory of change”. It adds that the theory “needs to be adaptable” and lists its continued validity among the things UN entities should monitor.
Evaluators also expect one. The OECD Development Assistance Committee’s revised evaluation criteria, adopted in December 2019, refer to an intervention’s underlying theory of change when judging relevance, and ask that effectiveness be assessed along the “results chain / causal pathway”.
In practice, a funder, partner or evaluator is likely to ask the same question: how exactly is this supposed to work?
Theory of change or logframe?
The two are related but do different jobs. A 2012 review for the UK Department for International Development, written by Isabel Vogel, found that a theory of change maps the wider process of change and its context, while a logical framework (logframe) maps the intervention itself.
The review notes that the logframe’s assumptions column is often treated as the “too hard” column. A theory of change puts those assumptions at the centre. The majority view among contributors was to keep both: a logframe for performance management, and a theory of change for analysis and critical thinking.
How to draft one: five steps
The steps below draw on the guidance cited above. They work for a single project, a training programme or an institution-wide strategy.
1. Start with the long-term change
Agree the change you want to see in the world, not the activities you plan to run. “Fewer women die in childbirth in three districts” is a long-term change; “train 200 midwives” is an activity. Be specific about who changes, where and by when.
2. Work backwards to the preconditions
Ask what must be true just before that change can happen, then what must be true before that. Working backwards in this way fills what the Vogel review calls the “missing middle” between outputs and higher-level outcomes. Each step should be a change in behaviour, practice, policy or capacity, not a task.
3. Name the assumptions behind every link
For each arrow between two boxes, write down why one should lead to the other. If midwives are trained, will they be posted where they are needed? Will facilities have the supplies to use the new skills? These are the assumptions most likely to break, and the ones an evaluation should test.
4. Ground it in evidence and context
Each assumption should point to evidence: published research, routine data or local knowledge. Where evidence is thin, say so. A good theory of change is honest about which links are well established and which are bets. An earlier Praxis insight on questions leaders should ask of their evidence offers a checklist for that stage.
5. Turn it into a living tool
Link each step to an indicator, so monitoring data can show whether the expected changes are happening. Then set dates to revisit the theory. The ODI researcher Craig Valters, in a 2015 piece on principles for theories of change, urged teams to “think compass not map”: a guide for direction that is updated as the terrain becomes clearer.
What the evidence says
Theories of change are widely used, but the evidence on how well they are used is more mixed. A 2016 systematic review in Implementation Science by Breuer and colleagues examined 62 papers on public health interventions. Of these, 49 (79%) described how a theory of change was developed, but only 18 (29%) described using it to develop the intervention itself.
The authors found that descriptions of both development and use “lacked detail” and proposed a reporting checklist. The review covered published literature up to its search date, so practice may have moved on since. But it suggests that many teams draw a theory of change without letting it shape the design.
Common pitfalls
The strongest criticisms come from practitioners. The 2012 DFID review warned that a mandatory theory of change could become a compliance exercise, and that diagrams can slip back into linear, results-chain thinking. It described the approach as “not a magic bullet”.
Valters raised similar concerns: theories drafted at the start of a project and never revisited, documents written mainly for donor reporting, and theories imposed by a small group without the partners and communities they describe.
Four warning signs suggest a theory of change has become paperwork:
- It has not been opened since the funding proposal was approved.
- Every arrow points neatly forward, with no feedback loops or external factors.
- The assumptions are generic, such as “political stability”, rather than specific to the programme.
- The people expected to change were not involved in drafting it.
There is also a resource cost. Facilitating an honest discussion of assumptions takes time and skill, and the Vogel review notes that time pressure often limits the quality of the process.
Making it useful
A theory of change is most valuable when it is treated as a hypothesis to be tested rather than a promise to be defended. That means involving implementers and communities, revisiting it at planned review points, and using monitoring data to check the links that matter most.
For institutions, the same discipline applies beyond individual projects. A strategy, a capability programme or a digital reform can all be expressed as a chain of expected changes with stated assumptions, and then tested against reliable data.
Praxis Global Institute’s advisory and partnership work helps institutions diagnose problems, agree practical plans and review delivery; to discuss a programme or strategy, contact the team.
Praxis Editorial Team
The Praxis Editorial Team publishes Praxis Global Institute’s explainers, practical guides and analysis of new evidence and policy in global health, education and professional capability, ethical AI and technology, research and data, and institutional strengthening. Every article is researched from primary sources, cites its evidence and is published under the editorial direction of its editors.
Facing a similar challenge in your institution?
We work with teams on research, data, digital systems, and learning that hold up in practice.
