Understand causes of outcomes and impacts
Most evaluations require ways of addressing questions about cause and effect – not only documenting what has changed but understanding why.
Impact evaluation, which focuses on understanding the long-term results from interventions (projects, programs, policies, networks and organisations), always includes attention to understanding causes.
Understanding causes can also be important in other types of evaluations. For example in a process evaluation, there often needs to be some explanation of why implementation is good or bad in order to be able to suggest ways it might be improved or sustained.
In recent years there has been considerable development of methods for understanding causes in evaluations, and also considerable discussion and disagreement about which options are suitable in which situations.
When choosing between these different options, consider the different types of causal inference that might be involved:
One cause producing one effect – it is necessary and sufficient to produce the effect
Two or more causes combining to produce an effect (for example, two programs or a program when combined with other factors such as particular participant characteristics) – one of the causes alone is necessary but not sufficient
Two or more causes being alternative ways of producing an effect – either of them are sufficient and neither is necessary
Different labels might be used for these different types of causal relationship - ‘causal attribution’ implying a single cause, ‘causal contribution’ implying a package of causal factors, and ‘causal inference’ being used to refer to all of these.
It is also important to consider the different types of questions that might be asked about cause and effect:
Did the intervention make a difference?
For whom, in what situations, and in what ways did the intervention make a difference?
How much of a difference did the intervention make?
To what extent can a specific impact be attributed to the intervention?
How did the intervention make a difference?
You can explore the three broad strategies for causal inference shown below.