The following pointers are a summary of the suggestions provided by members of Evaltalk, the American Evaluation Association Discussion List in June 2002. These suggestions are based on the assumptions that the primary purpose of evaluation is to improve programs or initiatives, and that people are more open to learning and change when they are not feeling threatened.

1. Use a participatory approach from the start.
Engage stakeholders in describing program logic, defining evaluation questions, identifying indicators of success and selecting appropriate data collection methods and tools. When these are defined by stakeholders, evaluation results are more likely to be in line with their expectations.

2. Discuss possible negative results in the early contracting and design stages.
Encourage clients or stakeholders to articulate their concerns and expectations early on about what the evaluation will reveal, and plan with them about how best to handle these results if they do occur.

3. Inform clients immediately and often - a 'no surprises' approach.
The worst way for people to learn about negative results is in the evaluation report or in a near-final presentation. As soon as any negative results begin to emerge, gently inform the client through a phone call or a meeting. Continue to communicate this both formally and informally as the evaluation progresses. This approach provides time for people to come to grips with negative findings, to decide how to handle them, and to question the methods or data while there is still time to make adjustments.

4. Build in time for course correction.
Recognize from the start that negative findings may occur, and build time into the evaluation plan for clients to initiate action to address them before the evaluation is complete. The final report can then tell the positive story of how a problem was id entified and has been corrected.

5. Question the evaluation plan.
In cases where evaluation questions, indicators or data collection tools have been imposed on the program, question whether they are appropriate. If not, develop alternative criteria and tools, and tell both stories: how the imposed methods show no progress but locally relevant methods do.

6. Emphasize the positives.
Every initiative will have some positive results, even if they are not very relevant to the funders' priorities. Make sure that your evaluation captures all positive outcomes, and highlight these. Begin and end reports and presentations with the positives, sandwiching the negative findings in the middle.

7. Tell the truth.
Ethically, negative findings must be fully reported. Most of the stakeholders will already be aware of the problems and will appreciate the fact that they have been brought out into the open and can now be addressed.

8. Present results in terms of lessons learned.
Identify what is working, what might need tweaking, and what needs to go back to the drawing board.

9. Provide suggestions for addressing deficiencies.
Provide clients with concrete suggestions for addressing the issues, drawing on your own experience and the research literature. Refer to best practices and to how others have successfully handled similar issues. When available, provide contacts who have agreed to speak with them about how they dealt with these issues.

10. Involve stakeholders in identifying obstacles and ways to overcome them.
There are often many good reasons why work has not been carried out as planned or objectives have not been achieved. Use a participatory process such as a force field analysis to engage stakeholders in identifying what internal and external forces were working against them, and describe these in your report. Involve stakeholders in identifying ways to overcome these hindering forces and to strengthen the forces that support their work.

Catherine Bingle Barrie, Ontario CANADA Bill Collins Dartmouth, Nova Scotia CANADA Susan Lilley Dartmouth, Nova Scotia CANADA

Joyce Morris Oklahoma City, Oklahoma USA Burt Perrin La Masque, Vissec FRANCE Eileen Stryker Kalamazoo, Michigan USA