Application of geostatistical and neighbor analyses to data from plant breeding trials

In plant breeding trials and other experiments with many treatments, there may be spatial effects due to large-scale trends and small-scale autocorrel ation. Geostatistical analysis may be used to investigate the extent of spa tially structured variation. When spatial structure is present, neighbor an alysis can be superior to classical analysis of variance (ANOVA). The objec tives of this study were to assess the validity and efficiency of neighbor analysis and to investigate the extent of spatially structured variation in cereal breeding trials. Three neighbor analysis methods (the Papadakis pro cedure, the iterated Papadakis method, and the first differences with error s in variables [FD-EV]) were applied to data from uniformity trials and to data from a set of 361 cereal breeding trials. The iterated Papadakis metho d consistently underestimated the variance, producing tests with highly inf lated Type I error rates. Thus, its relative effiiciency could not be estim ated correctly. Geostatistical analysis indicated that spatially structured variation was frequently present in the cereal breeding trials, and that f irst differencing was effective in removing it. The FD-EV analysis consiste ntly improved the accuracy and precision of the estimation of entry effects compared with classical analysis of variance and Papadakis analysis. Effic iency relative to classical analysis of variance averaged 152% for FD-EV an d 116% for the Papadakis procedure. Considering both validity and efficienc y, FD-EV was the best method. In the presence of spatially structured varia tion, FD-EV can improve the interpretation of data from field trials. In th e absence of spatial structure, FD-EV causes no loss of efficiency.