What is a “theory of change” and why do we need it?

A theory of change describes a process of desired social change by making explicit the way we think about a current situation or problem, its underlying causes, the long-term change we seek, and what needs to happen in society in order for that change to come about.
As shown in Diagram 1 below, a theory of change helps us uncover the thinking that guides our interventions/actions. By making explicit our thinking, we can be proactive in testing and adapting this thinking, which, in turn, helps us improve our interventions/actions in a more timely manner over time.

Diagram 1: The link between theory and action


Because social change is a complex affair, we need tools like theories of change to help us:
  • Build a common understanding of our collective thinking with regards to the process needed to achieve a desired social
  • Identify potential weaknesses or gaps in our collective thinking, such as certain hypotheses or assumptions that need to be tested, refined or discarded.
  • Develop more coherent program strategies that are constructed from logically robust theories of change
  • Engage in better learning that brings together theory and action. We need to be more adaptive, iterative, and non-linear in the way we think so that we can be more coherent, nimble and effective in the way we act. In the most extreme case, we may gather evidence that requires us to reframe our thinking all together, and this reframing will require major retooling and transformation of our actions.
There is no set format for a theory of change. It can be expressed in diverse ways-as cyclical processes that show the systemic nature of the different factors, as formulas, etc

Back to top

Main elements of a theory of change

So, we can define a theory of change as: a set of hypotheses, critical assumptions and indicators that make up causal pathways of change needed to bring about a desired long-term goal. Hypotheses are ‘if-then’ statements between different levels of a pathway of change.

A theory of change consists of seven elements, which are shown in Diagram 2 below.
  1. A statement of the current situation and major underlying causes affecting the program impact group
  2. A desired long-term goal
  3. Domains of change and main stakeholders
  4. Pathways of change, which include breakthroughs and incremental changes
  5. Stakeholders
  6. Indicators
  7. Assumptions

Diagram 2: Elements of a theory of change
Below is a quick explanation of each element.

1. Current situation and major underlying causes
A statement of the current situation or problem, including the underlying causes/barriers to change, is needed before a group can group can proceed to talk about the long-term change they seek, and what needs to happen in society in order for that change to come about. It may seem obvious that people working together to design a program understand the current situation in the same way, but one will be surprised that this is usually not the case. More often than not, multiple situational analyses have been conducted as part of or prior to the program design process. These analyses have yielded many findings, which are often interpreted in different ways by staff depending on their own practical experiences, training and personal biases. Some staff are not even aware of the findings of these analyses. As a result it is important to make sure that all folks who are developing the theory of change are on the same page by asking them to articulate the current situation/problem they seek to address/change. This statement of the current situation and underlying causes/barriers to change should reflect the findings of the situational analyses that underpin the program.

2. A desired long-term goal and impact population group

This is a 10-15 year ambition, and specifies the kind of enduring impact we would like to see achieved in the lives of the impact population group, at broad scale. An impact goal should be framed in terms of the desired change, such as women’s improved access to and control over assets, and their greater equality in various forms of decision forums, as well as the reshaping of institutions so that women’s voice is represented equitably.

Again, as is the case with the current situation above, groups working together tend not to have a common understanding of the desired change, unless they are asked to identify it together.

3. Domains of change
Domains of change are main areas in which change must occur in order to be able to reach the desired long-term change. Domains of change are identified based on the underlying causes identified as part of the statement of the current situation (element 1 above).

4. Pathways of change
A pathway of change is a map that illustrates the series of major breakthroughs and related incremental changes that are needed within a domain of change in order to reach the desired long-term goal. A breakthrough is a change that represents a leap forward or an advance on the pathway of change that is not easily reversed. For instance, a key breakthrough related to reducing violence against women could be the enactment of a law that makes the practice of dowry illegal.
To identify breakthroughs, it is important to distinguish between actual breakthroughs and incremental changes, which tend to represent preconditions that are needed in order for the breakthrough to occur. Note that breakthroughs and incremental changes should be nouns, not verbs, since they show outcomes (changes, achievements, results) rather than actions. Also, it is important to be selective and prioritize only those breakthroughs, and related incremental changes, that will leverage the greatest outputs.
Also, because pathways to change depict what needs to happen in society in order for the desired long-term change to come about, breakthroughs (and domains of change, for that matter) should not be limited only to those changes that CARE can do. This is the time for stakeholders to think big, without the limits of what any individual organization can do.

5. Stakeholders
Critical stakeholders that must be involved in order for the changes called for by the domain to come about should be identified. Stakeholders can include specific groups in communities, community-based organizations, specific government agencies at different levels, international NGOs, donors, and international financial institutions.

6. Indicators
Indicators tell the story of how success will be recognized at each step in the pathway of change. Indicators should be defined for each prioritized breakthrough on the pathway of change, as well as the long-term impact goal. In addition, indicators should be operational. By operational we mean that they include enough detail for us to be able to measure it.

7. Assumptions
Finally, a theory of change consists of assumptions that stakeholders use to explain the change process they have envisioned. There are three types of assumptions:
1) Assumptions related to the connections between the underlying causes and the problem that the stakeholders are trying to address.
2) Assumptions that explain why each breakthrough is necessary to achieve the long-term impact (assumptions behind the if-then hypotheses)
3) Assumptions about the context/environment in which the theory of change is situated
4) Once program strategy is defined based on the theory of change, there will be assumptions related to the connection between particular activities and the breakthroughs they are expected to generate.

Back to top

Testing and adapting a theory of change over time

A common question about the theory of change is, ‘How do we know that it is, at least initially, adequate?’ A theory of change is adequate when it: provides a logical and coherent explanation of how the major underlying causes of poverty or barriers to change are to be addressed; and specifies the major required areas of change, paying special attention to the structural dimension. The theory of change should be easily understood and verifiable by all those stakeholders who use it to guide their interventions and learning processes.
Because a theory of change should be tested and adapted over time, a theory of change is not:
  • A framework that everyone is 100 percent in agreement with. Nuances and differences in opinion are very welcome in theories of change
  • A set of hypotheses and assumptions written in stone that cannot change, but rather something that is evolving permanently
  • A framework that is so complex that no one except for the author(s) can understand it. People involved in the work should be able to think about the theory of change on a daily basis as they carry out their activities
  • Something that is reviewed only as part of an evaluation or other after action reviews
  • Something that is in a rigid format; the format should be adapted to the
  • Difficult to do reflective practice in relation to that theory of change. Too complicated
  • The format in which the theory of change is expressed should be flexible enough to adapt to the different ways in which the people involved in doing the work organize their thoughts. We should not impose one way of representing a theory of change.

To test a theory of change, stakeholders need to:
§ Prioritize those assumptions that are critical for the causal pathways to hold and determine ways to gauge the level of trustworthiness of those assumptions (through interviews, impact measurement, or other information gathering activities)
Compare the predictions of the theories of change (the if-then statements) with the results of the actions. To do this, we will need to identify indicators that will enable us to know if the predicted results of actions occurred (to verify if the hypothesized causal relations occurred or not).
  • Review available data that support or disprove elements of our theory of change, or the theory of change as a whole (comparing our theory of change with other approaches and new ideas that others bring)
  • Need to be interested in discovering weaknesses and inconsistencies in the theory of change. Karl Popper – contribution of Austrian philosopher. Until him, people tried to prove hypothesis. In the effort of doing this, they found components that were correct, and components that were not correct. Popper shifted the approach to one that aims to discover how hypotheses could be false. Experiments/analysis should be done to falsify a hypothesis – only way to really test a hypothesis.
Back to top

From theory of change to program strategy

At the heart of any program strategy is a theory of change. In starting to develop the program strategy from the TOC, we move from nouns (in the change pathways) to verbs, from the things we wish to achieve to the activities and actions needed to achieve them
In this, we need to weigh up risk carefully
We start by look for the opportunities we can use as entry points to start the process


When we first start to look at a program strategy, one of the obvious realizations will be that there are many different kinds of activities that could be implemented in pursuit of the program goal. The theory of change helps us select the most critical or strategic from among these possible activities. At times, small inputs in a social system can result in surprisingly big outcomes, whilst big, expensive interventions often yield small, dismal outcomes. The breakthroughs in the theory of change will help us prioritize strategies.

Note that the development of theories of change and program strategies is an iterative process. Program strategies are constructed from theories of change, and theories of change can be refined/improved once the program strategy has been crafted.

Back to top

Difference between a theory of change and a log frame

It is important to understand the difference between a theory of change and a project log-frame. A project log frame is a management tool that project teams use with their donors to communicate, negotiate and evaluate projects.
The log frame does not provide a method for designing projects, but it does provide a format that serves as a repository of information about the project. This format has implications for the method of designing the project, but it is not a method for designing projects in and of itself.
A log frame is a tool that donors use to make decisions on investments - should we invest in this set of activities or not? The indicator of the actions of the logical framework is the budget (what will the activity yield and what will it cost -- the budget is used as an indicator for making decisions).The log frame was created to help determine the budget of a project and to estimate the cost-benefit of a project
To the novice, a logical frameworks may seem to have elements of a theory of change (mostly in its first column), but we must look closely in order to see that they are actually quite different!
There are three main components in a log frame (include here a table with the headings of the columns showing the following breakdown):

  • A column for specifying causal relations.
But these causal relations are only related to the activities of the project.
All the other components of the theory of change that are not directly related to the actions of the project are missing or are fuzzy. Log frames are not good for supporting systemic thinking. All the other elements of the change process that are happening that are not directly connected to the actions of the project are left out.

  • Columns for management purposes
  • Columns to specify assumptions and risks (flow of time). Log frames make project teams assume stability of the context – it does not try to predict changes in context or how to adapt to/capitalize on the changes in context. This is the opposite of scenario planning. It is absurd to assume stability of context in the long run.

It is useful primarily to learn around activities and methods (single loop learning)
What is missing? The causal relations/pathways of the social change we seek. Only looks at the causal relations of the activities of a project - assumes that social change is a predictable, linear process. Therefore, it is not possible to make evident the theory of change that orients the project.
It is not useful for more developed levels of learning (double-loop learning) Does not help reframe – on the contrary, makes people mentally rigid – “Lets me see if I am doing my activities well or not.” But does not let me see if I am doing the right activities
Log frame is not a good tool for learning because it– it was not created for that purpose. It does not make evident the assumptions that are behind the causal relations. We need a tool that enables us to be adaptive, iterative, and non-linear.

Back to top

Ideas on mapping log frames to the theory of change and program strategy

Once a theory of change and program strategy have been developed for a program, it will be necessary to align current project portfolio to the program (identify which projects are aligned, and which need to be phased out, and where the gaps are so that we know what kinds of new initiatives/projects we need to establish)
In this process of aligning current projects to the program, it may be useful to map the activities listed in the log frames of relevant projects to the program strategies (this way we will know which project activities are not aligned, or where the gaps are in program strategies).
Other possible mappings that could be helpful are:
1) goals of projects mapped to long-term goal of program
2) project indicators mapped to domains of change or breakthroughs??

Back to top


Back to top

Learning Labs Home | Impact Measurement Wiki | How to Use a Wiki