A 2x2 table from a study of an uncommon adverse drug reaction has several cells with expected frequencies below 5. Which test should be used instead of the conventional chi-square test?
- A Yates corrected chi-square
- B Fisher exact test ✓
- C Student t-test
- D McNemar test
Correct answer: B. Fisher exact test
Explanation
When any expected cell frequency in a 2x2 table falls below 5, the chi-square approximation becomes unreliable and Fisher's exact test, which calculates the exact probability from the hypergeometric distribution, is preferred. The Yates correction applies continuity adjustment for expected counts near but generally not this far below 5, while McNemar is designed for paired categorical data, not independent groups.
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.