Negotiation Journal 11:4 (October 1995), pp. 329-332.
Summary by Tanya Glaser.
Copyright ©1997 by the Conflict Research Consortium
Sander expresses six reservations about the usefulness of settlement statistics in evaluating mediation programs:
First, what should be considered a "successful" outcome? Sometimes mediation produces a settlement indirectly, by providing parties with the tools to reach their own settlement.
Second, emphasis on settlement overlooks the importance of process. Research shows that party satisfaction rests as much on the quality of the mediation process as on whether a settlement is reached.
Third, settlements statistics do not take into account the quality of the various cases. Cases range from complex multiparty, multi-issue conflicts to relativly straight-forward, limited disputes, to cases which are found to be inappropriate for mediation.
Fourth, an important effect of mediation is that it may help the parties to develop better conflict resolution and prevention skills. This educational process is not taken into account by settlement statistics.
Fifth, settlement statistics do not address the quality of the resolution. Was the agreement coerced or consensual? Was it fair? Could it be successfully implemented?
Finally, even if reaching a settlement is taken to be the primary goal of mediation, the costs of mediation versus other methods of reaching settlement must be considered.
Sander suggests taking a more multi-faceted approach to evaluating mediation. Settlements statistics should be supplemented with other information, such as, "the 'quality' of the settlement reached (eg., whether it is Pareto optimal), cost and time savings, the satisfaction of the parties with the process as well as the outcome, and the extent of compliance with mediated settlements." [p. 331].