Abstract

VARGAS-ROJAS, Jorge Claudio. Uniformity trials simulation to determine the statistical power in yield rice trials. Agron. Mesoam [online]. 2021, vol.32, n.1, pp.196-208. ISSN 2215-3608.  http://dx.doi.org/10.15517/am.v32i1.40870.

Introduction. Prospective analysis of the statistical power of a hypothesis test should be one of the most important stages in any experiment, however, it is frequently omitted. In Costa Rica, within the literature consulted, no research related to this topic was found for yield experiments in rice cultivation. Objective. To simulate uniformity trials to determine the power of a completely randomized design for rice yield experiments in Bagaces, Costa Rica. Materials and methods. The parameters of the spatial correlation process of a blank trial established in Bagaces, Guanacaste were estimated. Then, the estimates were used to perform 10 000 simulations of larger random fields, this allowed to superimpose different number of repetitions and estimate the power achieved to detect a difference of 10 % with respect to the mean in a completely randomized experiment at a significance level of 5 %. Results. The power of 80 % was obtained with five repetitions. Conclusion. In rice yield trials, to detect a mean difference of 10 % at a significance level of 5 %, this investigation required five or more repetitions.

Keywords : test power; number of repetitions; geostatistical simulation; random fields.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )