An outcomes model is a visual model consisting of boxes setting out an intervention’s (a project's program's, organization's, policy’s or strategy's) high-level outcomes and all of the lower-level steps that may influence achieving them.
Outcomes models go by names such as: ends-mean diagrams; causal models; results chains; program theories; theories of change; intervention logics; and program logics.
As a conceptual tool
When used as an conceptual tool, any intervention or outcomes system is viewed as having an underlying outcomes model. This model is conceived of as being made up of a series of boxes with high-level outcomes at the ‘top' of the model. (In fact, outcomes models in outcomes theory are represented as flowing from the left-to-right with the ones on the far right of the model being thought of as at the ’top’ of the model).
Lower-level boxes with outcomes models ‘influence’ higher-level boxes in a general ‘hierarchy of influence’ which flows in the models from left to right. However despite this general flow of influence, any box within an outcomes model can potentially influence any other box within the model. Influence in this sense is taken to include the full range of causal relationships that could possibly exist between boxes (uni-causal, multi-causal, positive and negative, and includes feedback loops etc.). The word influence is used in preference to causes because causes is sometimes used in a narrower sense and may not include the full range of influence-types listed above.
It is not assumed that it is always possible to fully specify either the boxes or the influence relationships between boxes within a particular outcomes model.
There may be different perspectives on what boxes and what influence relationships exist between boxes within any outcomes models when coming from the perspectives of different groups of people (e.g. program staff, funders, participants, stakeholders).
Any influence relationship between one or more boxes within an outcomes model, or may not, have been proved. To be proved means that the relationship has been established as credible by a particular group of people (e.g. program staff, funders etc.) to a particular standard of proof determined by that group.
Conceiving of all interventions and outcomes systems as being underpinned by an outcomes model such as is described above is a useful way of looking at such systems. It has the advantage of allowing outcomes theory to provide technically concise definitions and principles. For instance, several examples of how the concept of an outcomes model can be used are:
- Typically boxes relating to the term vision lie in a outcomes model above those relating to the term mission in strategic planning usage.
- Those wishing to make particular distinctions between terms (e.g. results and goals) can use the concept of an outcomes model in a similar way to the example give above regarding vision and mission. It should be noted that outcomes theory is usually agnostic on the use of many of the terms used in outcomes work. The purpose of outcomes theory is to provide a clear and technically precise way for people to specify the distinctions they want to make between such terms.
- More complex outcomes theory principles rely on the use of the concept of an outcomes model lying behind interventions and outcomes systems. For instance, the following outcomes theory principle is used to clarity confusion in many real-world outcomes systems and incorporates the concept of an outcomes model. The principle is that ‘Impact evaluation is the only option for high-level outcome attribution if no controllable indicators reach to the top of the outcomes model’. This is provides a quick way of describing a principle which would take longer without having the benefit of using the concept of an outcomes system within the principle.
As a practical tool
As a practical tool, outcomes models are an essential building-block in any outcomes system and are usually the best way of specifying what a program consists of as part of: strategic planning; prioritization; performance management; evaluation etc. Their fundamental role is illustrated by their place as the first building-block within the outcomes system diagram.
An outcomes model can be seen as like a ‘roadmap’ of a jouney
What a real-world Outcomes Model looks like
Real-world outcomes models usually take up more than a single page, they are broken up into different 'perspectives' or 'points-of-view'. In the case of the road-trip above you could have a driver, passenger and car point of view. Click within the webpage version of a outcomes model embedded below to see how you can drill-down within a real-world outcomes model. The point is, outcomes models can be as large as you like.
Setting up a group to build an outcomes model (described as a ‘DoView’ below)
Use Duignan's Outcomes model rules to draw your outcomes model
How outcomes models are used in practice
Download the free trial now of DoView software.
Anyone can use the above material, with acknowledgment, in their own for-profit or not-for-profit consulting work if they, or their clients, are using DoView Software. Anyone can use it to teach anyone else how to use DoView software. If you want to use it for other purposes please contact us using our contact form.
Source: Duignan, P. (2009c). Rejecting the traditional outputs, intermediate and final outcomes logic modeling approach and building more stakeholder-friendly visual outcomes models. American Evaluation Association Conference 2009, Orlando, Florida, 11-14 November 2009.