synoptic, national picture, but other metrics would be required to measure the progress of individual communities.
Metrics are an important tool of administration. They allow targets to be established and set clear goals for improvement. The very act of defining a metric, and the discussions that ensue about its structure, help a community to clarify and formalize what it means by an abstract concept, thereby raising the quality of debate. The general concept of resilience is one with which most people are familiar, but resilience is not something that communities have much experience in measuring. Resilience is also clearly influenced by multiple factors, making precise measurement very difficult. This immediately suggests a strategy of combining various factors, using appropriate weights, into a composite index. The set of factors, how they are measured, the weights given to each factor, and the operations used to combine them into a composite index all present issues that can be the subject of lengthy debates and contention. At the same time, the translation of an abstract concept into a rigorous procedure for measurement—the formalization of the concept—allows for monitoring, the comparison of progress in different communities, and the prioritization of actions and investments, all of which can be extremely helpful. The effects of actions and policy changes can then be monitored through time to produce more desirable outcomes in the future by comparing improvements in resilience that result from those actions to what was promised or predicted, iteratively modifying actions and policies, and perhaps recalibrating metrics.
To be useful in this context, a resilience metric needs to be open and transparent, so that all members of a community understand how it was constructed and computed. It needs to be replicable, providing sufficient detail of the method of determination of a community’s resilience so that it can be checked by anyone using the same data. It must also be well documented and simple enough to be used by a wide range of stakeholders.
Metrics may be quantitative, but metrics with no more than ordinal properties still allow resilience to be ranked and progress to be monitored. For example, a metric might set the qualitative levels “unsatisfactory,” “marginal,” and “satisfactory” resilience, without specifying quantitative measures or ranges for each level, as long as the procedure for arriving at a rating was open, transparent, and replicable. A scale similar to those used in academic report cards with designations of A, B, C, D, or F could also be used to indicate progress. In recent years, much of this process of defining a metric has been the subject of extensive research, often under the rubric of multicriteria decision making (MCDM). Many of these methods have been devised for problems embedded in geographic space, such as the selection of a site for a new public facility, or of a route for a new highway. The geospatial nature of such problems raises additional issues such as estimating environmental, social, and economic impacts of site selection for the new development and the way in which the necessary data to gauge these impacts can be incorporated into a collective planning process, as several texts make clear (see, e.g., Massam, 1993;