Practical application of variance analysis of four-factor experience data as a technology of scientific research

Abstract

Abstract The article shows the possibilities of practical application of the four-factor analysis of variance in research with fruit and berry crops, since the multifactorial experience, in contrast to the one-factorial experience, allows us to reveal not only the influence of two or more factors in one stationary experiment, one or another effective indicator and their combination, but also establish the presence of interaction of factors. The use of multifactorial experiments makes it possible to obtain significantly more information than single-factorial ones, which makes it possible to reduce the cost of funds, labor and time for the scientific solution of certain issues. Algorithms for mathematical processing of the results of a multifactorial experiment are given on specific examples of comparing the length of annual growth of branches of 9-year-old apple trees when studying the effect of four factors and eight options (variety, crown shape, crown orientation and sample size) in two gradations each (2x2x2x2). A compact, reduced algorithm for analysis of variance is presented, which has a number of advantages and can be used for various factorial schemes of multivariate stationary and other experiments, primarily due to a significant reduction in the time for data processing and analysis. Calculations according to the full scheme of the four-factor complex coincide with the final estimates of the differences in the mean values of the NDS of the proposed compact scheme.