Biometrical Letters Vol. 51(2), 2014, pp. 125-142
|
CONTROL TREATMENTS IN DESIGNS WITH SPLIT UNITS GENERATED BY LATIN SQUARES Shinji Kuriki1, Iwona Mejza2, Kazuhiro Ozawa3, Stanisław Mejza2 1Department of Mathematical Sciences, Graduate School of Engineering,Osaka Prefecture University, 1-1 Gakuen-cho, Naka-ku, Sakai, Osaka 599-8531, Japan 2Department of Mathematical and Statistical Methods, Poznan University of Life Sciences, Wojska Polskiego 28, PL-60-637 Poznań, Poland 3Department of Nursing, Gifu College of Nursing, Hashima, Gifu 501-6295, Japan |
This paper deals with two-factor experiments with split units. The whole plot treatments occur in a repeated Latin square, modified Latin square or Youden square, while subplot treatments occur in a block design within the whole plots. The statistical properties of the considered designs are examined. Special attention is paid to the case where one of the treatments is an individual control or an individual standard treatment. In addition, we give a brief overview of work on the design of experiments using the considered designs, as well as possible arrangements of controls in the experiments.
Latin square, Youden square, split units, efficiency factors, merging treatment method
