A sample size calculation for logistic regression involves complicated
formulae. This paper suggests use of sample size formulae for compari
ng means or for comparing proportions in order to calculate the requir
ed sample size for a simple logistic regression model. One can then ad
just the required sample size for a multiple logistic regression model
by a variance inflation factor. This method requires no assumption of
low response probability in the logistic model as in a previous publi
cation. One can similarly calculate the sample size for linear regress
ion models. This paper also compares the accuracy of some existing sam
ple-size software for logistic regression with computer power simulati
ons. An example illustrates the methods. (C) 1998 John Wiley & Sons, L
td.