Abstract

FLORES, Juan RamónNavarro  and  ROJAS, Jorge ClaudioVargas. Relative efficiency of the randomized complete block design for rice trials in Bagaces, Guanacaste, Costa Rica. InterSedes [online]. 2015, vol.16, n.34, pp.61-70. ISSN 2215-2458.

Selecting the appropriate experimental design is one of the most relevant aspects in scientific research. There are various types of experimental designs that can be used depending on the environmental conditions where the trial is conducted; however, for this research only the two most simple designs were considered: unrestricted design and randomized complete blocks (RCB). In Costa Rica, executing the trial under field conditions is enough to justify a randomized complete block design since its use has spread without assessing its efficiency. The objective of this paper was to assess the efficiency of the randomized complete block design compared to the unrestricted design for rice trials in the Guanacaste area. There are two methods to assess efficiency of an experimental design and both use the mean square error. The analysis was focused on trials with promising materials of the rice varieties and hybrids production program from CoopeLiberia RL between 2010 and 2014. Data from a blank test from the same area was also used to simulate both designs. Efficiency data obtained show that the RCB design contribution was not significant for the conditions in the Guanacaste rice area, where, in most cases, conditions do not require the use of randomized complete blocks. Choosing the RCB design based on the fact that the trial is going to conducted in the field is not enough reason to select this design. The incorrect use of RCB may represent losing accuracy to estimate the effect of treatments.

Keywords : unrestricted random design; randomized complete block design; rice; efficiency of the design.

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