Test Data Selection and Quality Estimation Based on the Concept of Essential Branches for Path Testing

Abstract

A new coverage measure is proposed for efficient and effective software testing. The conventional coverage measure for branch testing has such defects as overestimation of software quality and redundant test data selection because all branches are treated equally. These problems can be avoided by paying attention to only those branches essential for path testing. That is, if one branch is executed whenever another particular branch is executed, the former branch is nonessential for path testing. This is because a path covering the latter branch also covers the former branch. Branches other than such nonessential branches will be referred to as essential branches.