You are here

Conditions for a good evaluation, checklist


The conditions under which a meaningful and satisfactory evaluation of development research can safely be conducted are the same as in any other evaluation of research – yet with different emphasis on certain aspects.

Three categories of conditions are considered here:

  • adequate knowledge of the context,
  • operational conditions,
  • ethical conditions.

These are necessary conditions, not sufficient conditions. Experience shows however that quite a few evaluations as presently performed do not meet such conditions, not even the majority of such conditions.

1. Adequate knowledge of the context

The specific guide should insist on the need, for the practitioners of evaluation, to have – or seek if necessary – adequate and relevant information on the context in which the research was, or is going to be, conducted. Examples:

  • field conditions, when appropriate,
  • institutional context including supervision of the researchers, management, access to literature and internet, transparency, scientific freedom, etc.

This requisite is indispensable to ensure the dimension of comprehensiveness in any evaluation.

2. Operational conditions

For an evaluation to be properly organised, the following conditions are to be met:

  • the objectives, “stakes”, should be clear to all people involved. Such information must be provided prior to the conduct of evaluation,
  • tools and procedures must be made available. If necessary evaluators should be trained in using them,
  • material conditions for mailing, reproduction, etc. should be satisfied.

3. Ethical conditions

  • Transparency: the rules of the game must be made clear from the start. Yet the evaluators' anonymity will be respected when necessary, but the researchers are entitled to know the evaluators’ profile.
  • Fairness to the researchers: equality of chances between researchers, absence of any discrimination in the application of the specific guide.
  • No conflict of interests: any potential or real conflict of interest needs to be made explicit, and a remedy found.
  • Willingness of the researcher (individual or team) to participate in the evaluation and to accept the participation of external evaluators when so decided.
  • To the point: the evaluation does satisfy only the stated objectives. It does not take other considerations into account, except when previously announced ( examples : researcher’s age or affiliation; geographical origin of the project or proposal ; other resources already available, etc.)
  • Independence: freedom of every actor to give his/her judgment . No pressure of any kind. Clear instructions to that effect should be given to all, and provided beforehand.
  • In the specific case of scientists from the North evaluating a scientist or a research in the South, mutual respect, openness and honesty must be the rule. This of course applies also to the reverse : researchers from the South are more and more frequently involved in evaluations of development research that takes place in the North.

A code of conduct established by J. Gaillard might inspire North and South scientists in their mutual relationships with both colleagues in the South as well as in the North (Gaillard, 1994)

Another interesting reference is : SCRES/ICSU (Standing Committee on Responsibility and Ethics in Sciences/International Council for Science. Kathinka Evers, 2001)