Analysis of Variance

Analysis of variance is introduced using a worked analysis of some data collected by Charles Darwin on inbreeding depression in experimental maize plants. ANOVA is used to compare the heights of a group of cross-pollinated plants with a group of self-fertilized seedlings. The least squares process is explained, including the calculation of sums of squares and variances and the calculation of F-ratios. The organization of the results of a linear model into an ANOVA table is explained. The R code necessary to perform the analysis is worked through and explained in terms of the underlying statistical concepts, and the interpretation of the R output is demonstrated.