EXPERIMENTAL PRECISION IN CORN TRIALS USING THE PAPADAKIS METHOD

The objective of this work was to verify whether the use of the Papadakis method in competing corn hybrid trials would lead to modifications in the validity of assumptions for the mathematical model of variance analysis and for experimental precision indicators. To achieve this, corn-grain-yield data from 25 competing corn hybrid trials, performed in a design of complete randomized blocks, were examined. Each trial entailed verification of assumptions, variance analysis, hypothesis tests, statistics to identify experimental precision, and analysis using the Papadakis method. This method improves experimental precision indicators. The assumptions were valid for both analysis types (without Papadakis and with Papadakis). Mean figures for the Fasoulas differentiation index increased from 8.5 to 20.7 and selective precision rose from 0.82 to 0.92. Trials with three repetitions analyzed using the Papadakis method enabled the identification of superior corn hybrids in relation to grain yield, with 86.5% precision. To maintain the same precision in conventional analysis, four times the number of repetitions would be necessary.