Empirical investigation of a novel approach to check the integrity of software engineering measuring processes

We present an empirical investigation of the applicability of Benford's Law (1958) and Digital Statistics (Nigrine, 1995) in the context of software engineering metrics analysis and process validation. We have conducted an investigation to determine under what circumstances various software metrics follow Benford's Law, and whether any special characteristics, or irregularities, in the data can be uncovered if the data are found not to follow the law. Lists were formed from three software metrics extracted from 100 public domain industrial Java projects. These metrics were Lines of Code (LOC), Fan-Out (FO) and McCabe Cyclomatic Complexity (MCC). The results indicate that the first digits of numbers in lists of LOC metrics extracted from the projects closely followed the probabilities predicted by Benford's Law. The FO and MCC metrics did not follow the standard Benford's Law as closely as the LOC metrics.