ABSTRACT
Predictive methods are deployed by various agencies, including the government, to pretest different programs and policies for their likely impacts, process of implementation, and stakeholder acceptability prior to launching these completely or on a large scale. The underlying motivation while deploying these predictive methods is that these will provide results that are largely indicative of what outputs, outcomes, and challenges can be expected when these programs and policies are implemented in a full-fledged manner.While the importance of pilots in supplying information and learning platforms, pretesting policies, and programs including innovations prior to their implementation on a full scale is well acknowledged, there are challenges in terms of their ability to act as a predictive method beyond the context in which these are applied, when these are scaled-up, and when these are designed for complex issues and rapidly changing policy environments. Appropriate evaluation criteria and indicators while focusing on monitoring and learning aspects, need to be considered in view of these challenges. Political factors including the presence of multiple stakeholders can further influence the “predictive potential” of pilots.
