A study compares an uncommon adverse outcome between a new vaccine and placebo using a 2x2 contingency table. Three of the four cells have expected frequencies less than 5. Which test should be applied instead of the ordinary chi-square test?
- A Fisher exact test ✓
- B Student t-test
- C Yates corrected chi-square
- D Mann-Whitney U test
Explanation
The chi-square approximation is invalid when any expected cell frequency falls below 5 in a 2x2 table, because it relies on large-sample theory. Fisher exact test calculates the exact probability from the hypergeometric distribution and remains valid at any expected frequency, so it is the correct substitute. Yates correction is a continuity adjustment for marginal cases around an expected value near 5, not a remedy for grossly small expectations.
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.