Premium
Sample size estimation for comparing two or more treatment groups in clinical trials
Author(s) -
Day Simon J.,
Graham David F.
Publication year - 1991
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4780100109
Subject(s) - sample size determination , nomogram , statistics , factorial , sample (material) , mathematics , estimation , computer science , population , group (periodic table) , medicine , mathematical analysis , chemistry , management , environmental health , chromatography , economics , organic chemistry
Methods for estimating required sample size for comparing two population means have been published. Most involve the use of complicated formulae and tables. These methods are limited to comparing two groups. Although techniques exist to determine sample sizes for comparing more than two groups, they are intrinsically far more complicated. A simple linear nomogram is proposed as a solution to these problems, and its use is illustrated with examples of parallel group, ordered parallel group and factorial designs.