The use of a covariance analysis to compare derived linear regressions in stocking rate trials

Abstract

An analysis of covariance allows statistical comparison of derived data. However, it has not as yet been related to the Jones and Sandland (1974) model. In this paper the analysis of variance is therefore assessed in terms of relationships which are used in the Jones and Sandland model. If this method of analysis is used, four to six stocking rates are recommended in order to ensure that statistical significance is meaningful.