An investigator evaluates a new sputum smear reading method by testing 100 samples twice, once by the standard method and once by the new method, with every sample read by both methods. Each sample is classified simply as positive or negative. The appropriate test to compare the proportions positive by the two methods is:
- A Standard Pearson chi-square test
- B Fisher exact test
- C McNemar test ✓
- D Mann-Whitney U test
Explanation
Each sample contributes a pair of binary observations because both methods are applied to the same specimen, so the two proportion estimates are not independent. McNemar test uses only the discordant pairs, samples called positive by one method and negative by the other, and ignores concordant pairs. Applying ordinary chi-square, the natural wrong choice, treats the groups as independent and inflates the apparent information by counting every sample twice.
Reference: Park's Textbook of Preventive and Social Medicine, 27th ed.
High-yield for: NEET PGINI-CETNExTFMGEUSMLEPLABMRCP
Written and medically reviewed by the StethoPrep medical team.