Purpose
This metric helps determine the effectiveness of test cases in identifying defects. In addition to determining
effectiveness, you can also use it to reduce testing costs by reducing the number of times that a test is run.
Definition
Count these instances:
-
Total number of defects found per test case
-
Number of times the test case has been run
Then calculate the average defect yield per test case run (total number of defects found by a test case / number of
times that the test case has been run).
Analysis
If you run the test frequently, it is a candidate for automation. If you run it frequently, but it never detects any
defects, it is a candidate for retirement. If it rarely results in defects, it is a candidate for refactoring into a
different test case.
Note:
This requires traceability to be established between defects and test cases.
Frequency and reporting
Review Test Case Defect Yield at the end of each iteration during the team's iteration assessment.
Collection and reporting tools
Data for this metric can be collected with IBM® Rational® Quality Manager®.
|