One of the major themes emerging from the literature is the need for a context specific evaluation. Patton emphasizes this again and again throughout his two books, Creative Evaluations (1981) and Practical Evaluations (1982). Compared to the old "hypothetic deductive" paradigm of evaluation, the new paradigm (or Patton's paradigm) represents a method of "choices emphasizing multiple methods, alternative approaches and most importantly, the matching of evaluation methods to specific evaluation situations and questions" (Patton 1982: 21). Furthermore, good evaluations will seldom be guided by and directly built on specific evaluation models. Rather, each evaluation setting is approached as a problem to be solved.Salmen, in Listen to the People: Participant Observer Evaluation of Development Projects (1987), states "if time and space may be thought of as the length and width of a projects fabric, the social dimension may be thought of as its texture". Hence the need to embed project objectives, goals and processes, and the evaluation thereof, in the social, political, historical and geographical context. Salmen characterizes participant observer evaluation as "eclectic, responsive and flexible" (108).

The basic precepts for practitioners of participant observer evaluation are (Salmen 1987):

1. know the setting
2. adapt methods to the project and its needs
3. blend qualitative and quantitative data
4. focus on issues of concern to managers
5. be aware of goals, but not bound by them
6. place a premium on establishing trust
7. serve as a bridge between management and beneficiaries
8. strive for reliability rather than precision.

These precepts are quite applicable to an external evaluator coming in new to a project, hired particularly by management. Some of these precepts are quite applicable to our needs in setting up a framework, especially: the need to know the setting (context), adapt methods to the project and its needs, blend qualitative and quantitative data, be aware of goals but not bound by them, place a premium on establishing trust and strive for reliability rather than precision (Salmen 1987).

The need for evaluations to be context specific was also emphasized by our interactions with local practitioners, specifically Roberts (pers. comm. 1995), who referred to "customized" evaluations. The need for a generic framework for evaluations was questioned by a number of people consulted. However, although an evaluation team or evaluator may not have a framework on which to base their evaluations, they may have guidelines, principles, rules or terms of reference on which to structure their evaluations (Sinclair, pers. comm. 1995).

A number of ways are presented to ensure that evaluations are context specific. One is the establishment of "evaluation teams" or an "evaluation task force". The former was endorsed by Bietz (pers. comm. 1995). He stated that the Mennonite Central Committee (MCC) project evaluations are usually done with a team of approximately four members, and only one of them is usually from the "north". Others on the team are from the project area specifically and other developing countries.

Patton (1982) talks about the "evaluation task force". This is, again, a way to make the evaluation more context specific. Patton uses a number of different "stakeholders" on the task force--in this way, the evaluation can be "user specific"--that is respond more effectively to the needs of the person requesting the evaluation in the first place. Further, if users and participants are involved in the evaluation from the beginning, the probability that the findings will be used is greatly increased (Patton 1982; IDRC 1994). IDRC (1994) notes that "an evaluation [of a project, policy or program] is often so 'independent and objective' that it loses sight of the needs of the participants, beneficiaries and managers, so is not relevant to them".

Patton (1982), in the creation of an evaluation task force, is more interested in producing a "utilization focused" evaluation; that is one that can be used. The criteria for being a participant on the evaluation task force include:

1. authority / power to use findings in decision making
2. believe evaluation is worthwhile
3. care how the results are used
4. willing to make a firm commitment of time.

The task force's mandate includes making decisions with regards to the focus, methods, and purpose of the evaluation and acts as a vehicle for involving key stakeholders in evaluation. This results in an evaluation which is appropriate, relevant and useful.