A novel approach in robust estimation of optimum size of plots

Article Type

Research Article

Publication Title

International Journal of Agricultural and Statistical Sciences


The problem of determination of Shape and Size of Plots remains at the foundation of planning and executing controlled field experiments. During sustained research investigations (more than a century) on experimental fields it has been established with confirmation that the premier action-point to be addressed before setting up of evaluation trials is to explore ways and means to minimize the inherent variability existing among the yield observations (role of determination of optimum shape and size of plots offers a milestone-discovery in the precinct of such controlled field experiments with the pioneering work of Smith (1938)). The concept of robust optimum plot size is introduced in the paper [Pal et al. (2015)] and subsequently, it is also considered in another paper in Pal and Basak (2016). An extensive review on the aspect of robust optimum plot size is available in Pal et al. (2016). In this paper, the theoretical model employed in Pal and Basak (2016) is considered under the set-up of a superimposed probability distribution (which mimics the real-life situations arising under the perspective of the class of field experiments) on it. The robust optimum plot sizes (determined basing on the overlaid exponential probability density for the intra-class correlation coefficient) in the form of three layers (in squared meters) are, Layer I (14 (2×7), 18 (3×6), 21 (3×7), 28 (4×7) and 30 (5×6), robust very near to optimum plot sizes are, Layer II (10 (2×5), 12 (2×6), and 24 (4×6)), and robust near to optimum plot sizes are, Layer III (8 (2×4),12 (3×4) and 15 (3×5)), respectively.

First Page


Last Page


Publication Date


This document is currently not available here.