A.K. Kaushik
Department of Statistics, J.V. College, Baraut (U.P.) India
E-mail: ashok_kshk@yahoo.com
Abstract: When the size of the experimental material is not sufficient to accommodate all the treatments, we require incomplete block designs to test various treatments under study in agricultural and biological sciences. The author of the present paper has discussed one of the incomplete block design, namely, Balanced Incomplete Block Design ( BIBD ). He has tried to present the review of the available literature on BIBD in brief, its analysis in case of complete data, and in case of one missing observation as well. The subject matter discussed here is not entirely new, but its presentation is new. However, the method for the analysis of BIBD in presence of one missing observation has been developed by him in 1992 in his unpublished Ph. D thesis. The Complex mathematical expressions are avoided in the present paper, and only simple expressions are provided to analyze the data. The methods are also supported by suitable examples. This will be of great help to the investigators engaged in agriculture and biological sciences.
Keywords: Adjusted treatment total, Adjusted treatment mean, Bias, Varietal trials
REFERENCES
Bartlett, M.S. (1937). Some examples of statistical methods of research in agriculture and applied biology. J. Roral. Soc., B4,137-183.
Biometrics (1957). Special issue on analysis of covariance, Vol. 13 No 3.
Das, M.N. (1955). Missing plots in partially balanced and other incomplete block designs. J. Ind. Soc. Agri. Statist, 7, 111-126.
Kaushik. A.K. (1992). Analysis os balanced designs in presence of several missing observations. Unpublished Ph. D. thesis submitted to Meerut University, Meerut.
Kaushik. A.K. (2010). Analysis of balanced incomplete block design in presence of one missing observation. Int. J. Agricult. Stat. Sci., Vol 6, No. 2, pp 615-621.
Kshirsagar, A.M. (1971). Bias due to missing plots., The American Statistician, 25, 47-50.
Kshirsagar, A.M. and Mckee Bonnie (1982). A unified theory of missing plots in experimental designs., METRON, Vol. XL-N, 3-4.