TimelyScholarFree Tools › McNemar Test Calculator
Compare paired yes/no outcomes, such as before and after an intervention in the same people. Get the exact p-value and the chi-square test with and without continuity correction.
FreeNo account neededYour data stays in your browser
| After: yes | After: no | |
|---|---|---|
| Before: yes | ||
| Before: no |
| Discordant pairs | 12 (yes → no) and 4 (no → yes) |
|---|---|
| Exact p-value (binomial) | 0.077Recommended when discordant pairs are few |
| χ² with continuity correction | 3.063, p = 0.080 |
| χ² without correction | 4.000, p = 0.046 |
| Odds ratio (b/c) | 3.000 |
How many answered yes/no before and after.
The test uses the discordant pairs, the people who changed.
McNemar's test checks whether the proportion with an outcome changes between two paired measurements. Only the discordant pairs, people who changed from yes to no or no to yes, carry information about the change.
When the number of discordant pairs is small, the exact binomial version is preferred.
People who gave the same answer both times tell you nothing about change, so the test uses only those who changed.