Simulation study of three adjustment methods for the modified augmented design and comparison with the balanced lattice square design

Abstract

SUMMARY Three adjustment methods using control plots to adjust for soil variation in a modified augmented design (Lin & Poushinsky, 1983) were studied by simulation. The results show that adjustment by design structure (row and column correction factors) is best when soil variation is relatively uniform in one or two directions, while adjustment by regression analysis is best when the variation is multi-directional. Adjustment using the control plot as a fertility index is least satisfactory. A modified augmented design in a replicated experiment is generally inferior to a balanced lattice square design but is competitive when the percentage of environmental variation attributable to soil factors is less than 70% of the total variation.