Much of the evaluation literature centers on the concepts of educational and social programming, focusing heavily on the concepts of efficiency and cost-effectiveness. The majority of the literature reviewed, therefore, applied to the task at hand in only a general sense, but did not provide a specific method for the evaluation of project outcomes. It did, however, provide insight into a number of broad principles which are to be incorporated into every evaluation and are pertinent to the framework which was subsequently developed.

Patton (1988: 301) defines evaluation as:

The practice of evaluation involves the systematic collection of information about the activities, characteristics, and outcomes of programs, personnel, and products for use by specific people to reduce uncertainties, improve effectiveness, and make decisions with regard to what those programs, personnel or products are doing and emphasizes (1) a systematic collection of information about (2) a broad range of topics (3) for use by specific people (4) for a variety of purposes.

This definition is quite useful as it is comprehensive, flexible and broad. Definitions of evaluation vary widely, and in recognizing the variations in and complexities of the field of evaluation you can develop a "situational responsiveness" (Patton 1988). This involves having sufficient flexibility to understand which definitions and models of evaluation are appropriate and meaningful in a particular context. For example, in contrast to Patton's general theoretical definition, the FAO (1988) defines evaluation more practically as "an assessment, as objective as possible, of how the project is going, how well participants are doing, and what effect it is having on intended beneficiaries" (p1).

In the context of participatory research on sustainable livelihoods, an evaluation can be thought of as a dynamic process specific to context in which both qualitative and quantitative measures have roles to play. The stakeholders and various communities must be actively engaged in the evaluation. Therefore, a generic evaluation framework which is guided by these principles will:

• be an iterative, cyclical process involving feedback loops into the project itself;
• have the capacity to be context specific and flexible;
• have the capacity to make use of both qualitative and quantitative measures; and
• actively involve the various communities represented in the project.