In software testing, among the various types of tests, two commonly conducted ones are unit and integration tests. Unit tests verify individual functionalities, and integration tests verify the combination of multiple functionalities. If we can identify unit/integration tests and measure them as ordinal values, such as the degree of integration-ness, we can utilize them to improve testing efficiency. However, the definitions of unit/integration are ambiguous, making it difficult to distinguish between them. To the best of our knowledge, there is currently no method for detecting this distinction. In this study, aiming to support the testing process, we will consider a measurement method for unit/integration tests. The key idea is to utilize an inclusion relation, which naturally exists among test cases. As an application of the inclusion relation, we propose a method for ordering failed tests to streamline debugging. We conducted a mutation analysis to evaluate how much our proposal reduces debugging effort compared to a naive method. The results showed that our proposal was effective in 29.7% of cases and confirmed an average reduction of 20.7% in debugging effort.