Analyzing Herbicide Interactions: A Statistical Treatment of Colby's Method

Publisher Website
Google Scholar
Abstract
A model and a proposed method for testing herbicide interactions were modified from an analysis of variance (ANOVA) model for a 2 by 2 factorial experiment. Statistical tests for either synergism, antagonism, or additivity of herbicide combinations were developed through transforming growth data to logarithms followed by significance tests of 2 by 2 contrasts of the form μii - μi0 - μ0i + μ00 with respect to the log-transformed data. Using actual experimental data, heterogeneity of variance was less severe on the log scale compared to the original measurement scale. An expedient SAS(R) program for obtaining the desired significance tests was developed.