| OCR Text |
Show 1938] Walker: Factorial Designs in Agricultural Experiments 3 factors. This type of information, which is equally important as a knowl edge of the single factors operating entirely independently, cannot be obtained from the experiment when it is designed on a single-factor basis. The third advantage of the factorial design, although less obvious than the other two, has an important bearing upon the utility of the experimental resuts in their practical applications. Any conclusion has a wider inductive basis when inferred from an experiment in which the variables of one factor have been studied in the presence of the variable conditions of the other factors than it would have from any amount of experimentation in which these have been kept strictly constant. Possibilities and Applications It would not be far-sighted to assume that designs of thi= tvne nre apr li only to field experiments where studies are being made of methods of cultivation, irrigation, soil fertilization, and other field practices. This method of experimental design is equally applicable to laboratory studies where cable chemical methods are being applied (1), to animal feeding trials, to manu facturing processes and, in fact, to practically all phases of biological and agricultural research. N either should one consider the methods to be limited to the combination relatively few variable factors. For example, rather comprehensive experiments could be set up to study (1) the yielding ability of promising varieties of a crop, (2) the fertilizer of the of a requirements soil, (3) time and methods of cultivation, (4) time and methods of and (5) resistance irrigation, to disease and insect attacks. Thus an experiment may be made extremely complex and comprehensive, and at the same time its efficiency and precision will be increased. Experiments of this type call for cooperative effort and will lend them selves well to the cooperative work of investigators in several fields. The result would be increased efficiency and economy in the use of land, equip ment, and in the expenditure of funds. With the limited facilities available to most investigators, and owing to the desirability for increased efficiency in research, it is questionable if one can afford to pass up the opportunities offered by factorial designs. References (1) (2) (3) Brandt, A. E. 667. 1937. Factorial Design. Jour. Amer. Soc. Agron., 29: 658- Fisher, R. A. The Design of Experiments. 2d ed. Edinburgh, Oliver and Boyd, 1937. Yates, Frank. The Design and Analysis of Factorial Experiments. Imperial Bureau of Soil Science. Technical communication no. 35. 1937. |
