This paper is made up of two parts. The first is critical assessment of current models for software metrics, namely Halstead's and McCabe's. The two approaches are deeply and critically examined in order to point out their weakness, both in the underlying theory and in their experimental validation. The second part describes an application of software metrics to Pascal, and proposes four different counting strategies. Finally, experimental results are given. The best counting strategy is chosen depending on the length of the program analyzed and a possible correlation between the main models is checked.
Software metrics: a critical evaluation and an application to PASCAL
MORASCA, SANDRO
1986-01-01
Abstract
This paper is made up of two parts. The first is critical assessment of current models for software metrics, namely Halstead's and McCabe's. The two approaches are deeply and critically examined in order to point out their weakness, both in the underlying theory and in their experimental validation. The second part describes an application of software metrics to Pascal, and proposes four different counting strategies. Finally, experimental results are given. The best counting strategy is chosen depending on the length of the program analyzed and a possible correlation between the main models is checked.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.