The comparison of three models applied to the analysis of a three-factor trial on hybrid maize (Zea mays L.) cultivars

Summary The aim of this paper is to present the comparison of three models applied to the analysis of a one-year study of protein yield for two types of hybrid maize cultivars under different forms of nitrogen fertilizer and doses of magnesium. The field trial was conducted in 2010 at the Agricultural Experimental Station in Swadzim (Poland). The results obtained were analyzed in terms of three randomization-derived models of observations. Consideration was given to three mixed systems, which are combinations of split-plot design, split-block design and randomized complete block design. The relative efficiency of the designs with respect to estimation of some comparisons among treatment combination effects was examined. Particular attention was paid to the interaction between cultivars and nitrogen fertilization.