Factor analysis of nursery seedling data in different compost substrates

Abstract

Analyzing units described by a mixture of sets of quantitative and categorical variables is a relevant challenge.
Principal components analysis was used to include these two types of variables in order to study the
correlations of a number of forest soil variables by grouping the variables in factors. Seedlings of four economically
important and ecologically different species, Quercus pubescens, Pinus maritima, Pinus nigra and
Pinus brutia, were grown in paper pots filled with either mixtures of sawdust, straw, byproducts of rice or sugar
beet substrates. The variables used in the analysis were: tree species, composts, height, diameter, weight above
ground and underground weight. The principal components method of factors extraction (PCA) begins by finding
a linear combination of variables (a component) that accounts for as much variation in the original variables
as possible. It then finds another component that accounts for as much of the remaining variation as possible
and is uncorrelated with the previous component, continuing in this way until there are as many components
as original variables. A few components accounted for most of the variation, and these components were used
to replace the original variables. In the nursery, species responded primarily to substrate type. However, there
were no interactions with nursery treatments.