An Empirical Analysis of Lack of Cohesion Metrics for Predicting Testability of Classes

The aim of this work is to explore empirically the relationship between lack of cohesion metrics and testability of classes in object-oriented systems. We addressed testability from the perspective of unit testing. We performed an empirical analysis using data collected from two Java software systems for which JUnit test cases exist. To capture testability of classes, we used different metrics to measure some characteristics of the corresponding JUnit test cases. In order to evaluate the capability of lack of cohesion metrics to predict testability, we used statistical analysis techniques using correlation and logistic regression. The performance of the predicted model was evaluated using Receiver Operating Characteristic (ROC) analysis. The achieved results provide evidence that there exist a relationship between lack of cohesion and testability.