Contents - Index
Generate a covariate-effect biplot from 3-way or 4-way data
For a 3-way or 4-way dataset, covariate-effect biplots can be generated in two ways, under the 4-Way main menu:
1) Covariate-effect biplot based on genotypic values, which will first generate a genotype by trait+environment two-way table and then utilize the module Covariate-effect biplot based on genotypic values.
2) Covariate-effect biplot based on phenotypic values, which will first generate a covariate-effect table of correlation coefficients, and then display it in a covariate-effect biplot.
Which one is better
A covariate effect biplot based on phenotypic values probably makes more sense, because it looks into the target trait vs. explanatory trait relations in the individual environments. However, since it is common that not all traits are measured in all environments, there could be many missing cells. When this is the case, covariate effect biplot based on genotypic values may be more feasible. The latter takes the same form as the data a QQE biplot is based, in which the explanatory variables are genetic markers ("genotypes"!).
Compare: VBV biplot.