A large number of metrics for evaluating the quality of software have been proposed in the literature. However, there is no standard terminology or formalism for defining metrics and consequently many of the metrics proposed have some ambiguity in their definitions. This hampers the empirical validation of these metrics. To address this problem, we generalise an existing approach to defining metrics that is based on the Object Constraint Language and the Unified Modelling Language metamodel. We have developed a prototype tool called DMML (Defining Metrics at the Meta Level) that supports this approach and we present details of this tool. To illustrate the approach, we present formal definitions for the Chidamber and Kemerer metrics suite.