An Application of Multiple Covariance Analysis to the Estimation of Fertilizer Response Functions

Abstract

A procedure to test whether or not significant differences exist in fertilizer response functions is important in the statistical analysis of experimental data. Analysis of covariance is often used in the analysis of linear functions. This paper extends the analysis of covariance procedures to quadratic and higher order least squares regression functions by applying the statistical techniques to the results of an actual experiment.