An analytical comparison of the fault-detecting ability of data flow testing techniques

Compares several data flow based software testing criteria to one another and to branch testing. The fact that criterion C₁ subsumes criterion C₂, does not guarantee that C₁ is better at detecting faults than C₂. However, if a certain stronger relation between the criteria holds, then for any program and any specification, C₁ is guaranteed to be better at detecting faults than C₂ in the following sense: a test suite selected by independent random selection of one test case from each C₁ subdomain is at least as likely to detect a fault as a suite similarly selected using C₂. It is shown that under those conditions, the expected number of failure-causing inputs in the C₁ test suite. These results are used to compare a number of data flow testing criteria to one another and to branch testing