The Estimation of Treatment Means by Designed Experiments
暂无分享,去创建一个
Although it is usual to present treatment means, an experiment primarily estimates treatment differences. Only if certain conditions are fulfilled can importance be ascribed to the means themselves. A method of analysis is presented that enables the blocks to be weighted so that the experiment as a whole is made representative of the population to which results will be referred. The method is available for all connected designs and is much simpler for randomized blocks.