BMJ 1996; 312 doi: (Published 13 January 1996) Cite this as: BMJ 1996;312:96

Estimating sample sizes for binary, ordered categorical, and continuous outcomes in two group comparisons

Editorial errors occurred in this paper by Campbell et al (28 October, p 1145). In formula 5 the denominator should have read: (log OR)2p(1-p) [not log (OR)2]. On p 1147 in the second column, in the worked example for ordered categorical data, the formula should read: CA1+OR(1-CA1[not Ca1+OR(1-Ca1)]. Also, all logarithms are natural logarithms.


View Abstract