Premium
A Method of Statistical Analysis of a Factorial Experiment Involving Influence of Fertilizer Analyses and Placement of Fertilizer on Stand and Yield of Cannery Peas 1
Author(s) -
Davis J. F.,
Cook R. L.,
Baten W. D.
Publication year - 1942
Publication title -
agronomy journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.752
H-Index - 131
eISSN - 1435-0645
pISSN - 0002-1962
DOI - 10.2134/agronj1942.00021962003400060003x
Subject(s) - fertilizer , agricultural experiment station , soil water , agriculture , yield (engineering) , mathematics , statistical analysis , engineering , environmental science , agronomy , statistics , archaeology , geography , soil science , physics , biology , thermodynamics
I the attempt by investigators to increase the efficiency of field experiments, a number of complicated experimental designs have been suggested by which the amount of information that may be secured from an experiment is measurably increased. However, the statistical analysis of the data resulting from such an experiment likewise becomes more involved and misuse of statistical methods is more likely to occur. This paper is presented with the idea that agronomists would be interested in a detailed statistical analysis and interpretation of results of a "factorial" design experiment concerning fertilizers on cannery peas. The purpose of this experiment was to study the effect of different fertilizer mixtures, placed in various locations with respect to the seed, on the yield and stand of cannery peas. Of special interest was the method for obtaining the interaction between fertilizers and placements. In an experiment of this kind it is necessary for reasons associated with the practicability of mechanical operations in conducting field experiments to limit the number of treatments to a minimum. The number of plots in an experiment with cannery peas is especially important because the quality of the peas changes very rapidly at harvest time and prices are dependent on quality. It is important, then, to have an experimental design which will furnish the desired information and still include no more plots than can be harvested within a 36-hour period. Because of this limit on the number of plots which could be cared for, the fertilizers were limited to three, o-2o-o, o-~6-8, and 4-~6-8. The placements were also limited to three in number: Placement x, fertilizers applied in contact with the seed; placement 2, fertilizers applied in bands ~/~ inch out and i~/~ inches below the seed; placement 3, fertilizers applied in bands e inches out and i ’/~ inches below the seed. The rate of application was 30o pounds per acre for the 4-I6-8 and the o-16-8 analyses and 240 pounds per acre for the o-2o-o. This arrangement provided for equivalent quantities of phosphate on all plots receiving fertilizer. By the use of three fertilizers and an unfertilized control and three methods of placement here are i2 treat-