At this stage, we are attempting the complex task of identifying change and attempting to link this change back to a specific "agent of change"; that being the use of output. When this link is made with some degree of certainty, we can move forward and evaluate whether or not the change is in accordance with the broader goals of sustainable livelihoods.

It is proposed that one agent of change will be the user-specific outputs produced by the initial project process. Another may be the process which produced these outputs. It must be recognized that a community (be it the policy community or a village group, etc.) will always be subject to the forces from a countless number of agents of change. Some examples include: natural disasters; budget cutbacks; new roads; the removal of a subsidy; a death in the village; change in decision-makers within a government; and the presence of an NGO working within the village.

The vectors (lines with both magnitude and direction) moving throughout the plane, CHANGE, can be thought of as these external agents acting on the process of change within communities. All the forces that impact on that plane will manifest themselves in change, yet only one of these vectors represent the Project, and more specifically, output use. Only after we have evaluated the change and linked this change with a degree of certainty to the output use can we call this change RESULTS. Therefore, a challenge inherent in evaluating this portion of the project is the link between change and the Project itself. The recognition and acceptance of an innate level of uncertainty is necessary before we proceed; otherwise, the task is paralyzing.

Agents of change as vectors impacting change.

A framework can be made to structure the process of identifying changes in the community from the time when the outputs were received (the starting point for the evaluation) and linking these changes, with some degree of certainty, to output use. The main components of this process are:

1. identifying change;
2. identifying factors which lead to that change;
3. determining if outputs are identified as a factor, and if so;
4. ranking and weighting of these factors;
5. determining if the output was a sufficiently significant factor influencing change; and if so
6. considering this change "results".

At this stage the evaluation makes the assumption that a community (be it policy, rural, etc.) can effectively engage in self-reflection and identify the major changes which have taken place in their immediate environments. Furthermore, they should be able to have some preliminary insight into what has caused these changes. The major points to be made are:

Participatory ranking exercises Stage 4

Stage 7

Choose evaluation facilitator and team

Identify output users

Determine whether outputs were received by intended users

Re-examine project mechanics

Determine whether outputs were used

Identify changes linked to outputs

Determine whether changes promote sustainable livelihoods

1. the community should be the main identifiers of change;
2. the actors external to the community should also be involved in identifying and verifying change; and
3. both communities should be involved in pointing to the factors which have lead to that change (Henerson, Morris, and Fitz-Gibbon, 1987 and Richards, 1985).

When a community is asked to identify major causes of change, the use of outputs may not be cited as a factor, be cited as the sole factor influencing change, or more likely, could be cited among a myriad of other factors. As we are only interested in determining change that is the outcome of output use, only change that is significantly linked to output use is considered results. It is up to the evaluation facilitator or team, on a case by case basis, to determine what is "significant".

There are many participatory methods which allow communities to rank and weight the significance of factors they have identified previously.