In a cluster randomised trial of a school-based health intervention, the intra-cluster correlation coefficient (ICC) is 0.05 and cluster size is 40. What is the design effect (DEFF)?
- A 1.95
- B 2.05
- C 3.25
- D 2.95 ✓
Explanation
Design effect (DEFF) = 1 + (cluster size − 1) × ICC = 1 + (40 − 1) × 0.05 = 1 + 39 × 0.05 = 1 + 1.95 = 2.95. The DEFF reflects the loss of efficiency due to clustering; a DEFF of 2.95 means the effective sample size is approximately one-third of the actual sample size. Ignoring ICC and DEFF in cluster trials leads to underestimation of standard errors and inflated type I error rates.
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.