Author(s): A. SAHA, A. MAJUMDER, G. R. MANJUNATHA AND 1P. K. DHARA

Abstract: The study mainly aimed at the comparison of analysis of experiments through ANOVA model and ANCOVA model. The selection of covariate is one of the most important parts of the analysis through ANCOVA model to reduce the residual error. The study includes the necessary conditions for selection of best covariate for analysis of covariance (ANCOVA) model in field experiments using the homogeneity of regression slopes (HOS) test (Karen ,2004) and the test based on the correlation coefficient value of dependent variable (y) and the covariate (x) (Cochran, 1957). The relative gains in efficiencies of ANCOVA models in the above field experiments were also calculated. The layout of the conducted experiment was done under randomized block design set up with 6 treatments and 3 replications. The used covariates, taking a single covariate at a time, for the above experiments were soil pH, Organic Carbon (%), amount of Nitrogen (kg ha-1) in soil, amount of Phosphorus (kg ha-1) in soil and amount of Potash (kg ha-1) in soil. In experiment, soil pH for the year 2012 and amount of Potash (kg ha-1) in soil for the year 2013 were selectedrnas best covariates, respectively. The relative efficiencies were 425.392 and 996.295 per cent for the years 2012 and 2013, respectively.

Download Full Article: Click Here