Field plot techniques for soybean experiments
1990
Latonio, A.M.L.S.
Results from 52 soybean experiments conducted in Central Luzon State University [Munoz, Nueva Ecija, Philippines] from crop years 1960 to 1988 were studied, with regard to determining the suitable experimental design, efficient plot sizes and shapes and sizes of experimental area. The weighted and unweighted Smith's Index of Soil Heterogeneity were estimated and used in the determination of efficient number of replications in combinations with the efficient plot sizes. The study shows that Randomized Complete Block (RCB) and Split-Plot (SP) designs are more efficient than Completely Randomized Design (CRD) for soybean experiments concerning yield, fertilizer application, plant protection and cultural practices. On the average, the use of these two design increased the precision of the experiments by 34.82 percent and 19.09 percent, respectively. The efficient plot sizes range from 6 sq. m to 12 sq. m. Within this range, there is an inverse relationship between the coefficient of variation of the efficient plot sizes are 2m x 3m, 2.4m x 3m, 2.5m x 4m, and 3m x 4m. Experiments with experimental area of sizes within the range 301 sq. m to 500 sq. m give the lowest coefficient of variation. In combination with the efficient plot sizes, the number of replications, for the range 2 to 10 replications, required to detect 5 percent significant level treatment differences of not more than d units (expressed as percent of the mean) and coefficient of variation for basic units CVo's generalized into rates CVo/d for values 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, and 0.9, at specified probabilities 0.8 and 0.9 of obtaining significant results for soybean experiments in RCB were determined. The most commonly chosen plot size in combination with number of replicants requiring less experimental area is 6 sq. m.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by University of the Philippines at Los Baños