Determination of the optimum plot size in agricultural essays

Abstract

This study aims to provide a theotrical basis for research on experimental plot size. The optimum plot size estimate is focused on two aspects: In terms of the minimum cost of the experiment, in basic units and in terms of the number of replications necessary to obtain a predetermined difference of mean, fixed "a priori", at the a level of probability. The b coefficient or soil heterogeneity coefficient was estimated using a regression model in which the data are correlated and have unequal variances and minimizing the sum of squares of the errors through the generalized least square method. A methodology was developed to estimate the ω^ij weight to the appropriate data. Such a procedure is valid for uniformity trial data as well as for other trial experimental data. The first case was studied through a hierarchical classification scheme and the second through of a randomized block design, with split-plot model