Abstract
We derive the asymptotic power function of the score test for detecting a common relative risk greater than unity from multiple 2 x 2 tables and formulate methods of sample size determination for use when designing stratified prospective studies. The stratified score test is more efficient than the unstratified test when the latter is unbiased.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have