EVALUATION OF SOME STATISTICAL METHODS FOR DETERMINING THE RELATIVE CONTRIBUTION OF YIELD COMPONENTS IN FLAX
Abstract: Three statistical procedures of relating yield components to yield namely: multiple linear regression (in the form of full model, stepwise and path analysis),factor analysis and cluster analysis were evaluated for determining the relative contribution of yield factors in flax. A set of six widely grown flax varieties, namely: Giza 9, Giza 10, Sakha 3, Sakha 4, Istro and Jetika were planted at Tag El- Ezz Agricultural Research Station during the two successive seasons 2010-2011 and 2011-2012 in a randomized complete blocks design with three replications for this purpose. Simple correlation coefficients revealed that fiber percentage, number of fruiting branches/plant, 1000 seed weight and number of capsules/plant had the greatest influence on seed yield/plant. Path analysis indicated that technical length/plant, number of fruiting branches/plant and 1000 seed weight were the three factors that exerted the greatest influence both directly and indirectly upon seed yield, when studying correlation or path analysis. Multiple linear regression recordedR2equal(83.6%)and stepwise regression recordedR2 equal (80.9%) denoting nearly accuracy. The test of multicollinearity and linearity in the two procedures between variables indicated that, these models show highly multiple coefficient of determination(R2) value for prediction only and were not valid to interpret the partial correlation coefficient. Factor analysis divided flax characters into two groups that accounted for 88.64% of the total variability. The first group contributed by 45.24% and included, total length/plant, technical length/plant, fiber length, fiber fineness and oil percentage. The second group accounted for 43.40% and contained fiber percentage, number of fruiting branches/plant, number of capsules/plant, number of seeds/capsule, 1000 seed weight and seed yield/plant. Cluster analysis also divided studied traits into two clusters ; the first cluster included(seed yield traits) and second cluster include (fiber treaties).Results of cluster analysis are similar to those obtained from factor analysis so; we could conclude that the two procedures give similar information that help plant breeders in flax breeding programs.
Publication year 2013
Organization Name
Author(s) from ARC
Publication Type Researsh & Applied Activities
↑