This will occur if
a) The tests have not been instrumented
b) All the tests have not been run
Look in your output for something like the following
The tests may not be instrumented if you have excluded them.
You also may not have run all the tests, and without running the tests. If the tests are not run, then the covereage will be incorrect and per-test coverage cannot work.
Modify the build configuration or the command you are running to ensure that all tests are run