Early validation of software quality models with respect to minimality and completeness: An empirical analysis

Validation of quality models is an important and necessary task in order to assure that the quality of a software product is being assessed using the appropriate model. Capture recapture models are employed to provide a quantification of the completeness characteristic of the model content. This article presents an approach for simultaneously validating software quality models with respect to their completeness and minimality. The validation process can be performed without the need of applying the model to a set of software products. The approach is an adaptation of capture-recapture models for estimating software defect content based on largely formalized reviews with defined instruments (questionnaires and templates). We discuss the feasibility of the approach and illustrate its application on four quality models. In conclusion, the application of this static approach has led to valuable information regarding enhancements for the validated quality models, although its statistical power is limited due to the low number of participants.