Timeline for How do you test your unit-tests for false negatives?
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 3, 2015 at 12:30 | answer | added | Kate Paulk | timeline score: 1 | |
| Mar 3, 2015 at 12:09 | comment | added | Kate Paulk | @Nope, user246 has the correct definition of false positive and false negative. In the case of test code, a "positive" means the code has found a bug - so a false positive is the test code saying something is wrong when there is no bug (that is, the test code is incorrect). A false negative is the test code reporting no bugs when there are bugs present (again, the test code is incorrect). | |
| Mar 2, 2015 at 18:36 | answer | added | Dale Emery | timeline score: 2 | |
| Mar 2, 2015 at 15:52 | history | edited | user246 | CC BY-SA 3.0 |
added 12 characters in body
|
| Mar 2, 2015 at 12:09 | history | edited | user246 | CC BY-SA 3.0 |
added 21 characters in body; edited title
|
| Mar 2, 2015 at 12:08 | comment | added | user246 | I edited the question to use "positive" and "negative" in the conventional way. | |
| Mar 2, 2015 at 11:13 | comment | added | StudentsTea | In my question, I describe two scenarios. In Secenario A, the code under test is entirely broken, but a subset of unit tests continue to pass--false positives. In Scenario B, the code under test is partially broken, and the unit tests written for that particular part of the code passes--false positives. I recognize the wording is confusing, though. Feel free to edit as you wish. | |
| Mar 2, 2015 at 11:05 | vote | accept | StudentsTea | ||
| S Mar 2, 2015 at 9:29 | history | suggested | Viktor Malyi | CC BY-SA 3.0 |
The author has mentioned only about unit tests in his question, so it would be better to stick to this term. Additionaly the "test-cases" can't be tested for false positives since they are just textual representation of actual tests which are implemented as unit-, integration- or system tests
|
| Mar 1, 2015 at 22:24 | comment | added | user246 | Terminology note: in medicine, the purpose of a test is to detect a disease; if it detects the disease, the outcome is positive, otherwise it is negative. Similarly, the purpose of a test is to find bugs, so FAIL is a positive outcome and PASS is a negative outcome. A false positive would be a FAIL that does not actually correspond to a bug, and a false negative would be a PASS that fails to detect a bug. You seem to be using the terms in the opposite sense. See for example "Positive or negative" in en.wikipedia.org/wiki/Medical_test. | |
| Mar 1, 2015 at 21:15 | review | Suggested edits | |||
| S Mar 2, 2015 at 9:29 | |||||
| Mar 1, 2015 at 21:09 | answer | added | Viktor Malyi | timeline score: 7 | |
| Feb 27, 2015 at 17:48 | history | tweeted | twitter.com/#!/StackSQA/status/571366197549240320 | ||
| Feb 27, 2015 at 12:28 | review | First posts | |||
| Feb 27, 2015 at 14:01 | |||||
| Feb 27, 2015 at 12:26 | history | asked | StudentsTea | CC BY-SA 3.0 |