ANALISIS KERAGAMAN PADA DATA HILANG DALAM RANCANGAN KISI SEIMBANG

Abstract

In a research, it is required a design experiment that obtained a conclusion desired. Balanced Lattices design is an experiment which the number of groups is k, the number of treatment is equal to square of the number of groups (k2) and the number of replication is equal to the number of groups plus one (k+1). In Balanced Lattices design often occurs the missing data. There are two methods estimate missing data. The first method is use equation. The second method is iteration. The second method is used if there was more than one missing data. Analysis of varians in of missing data in Balanced Lattices design is counted as in the complete, but the total degrees of freedom and the error degrees of freedom are substracted by m, respectively where m is the number of missing data. Based on the data used, a research conducted to determine the influence of 16 varieties fertilizer to the product of grain. The result of the research shows that in the case of with one missing data resulting that there is influence of varieties fertilizer to the product of grain with the F count is 5,092. The results of the study with two missing data resulting that there is influence varieties fertilizer to the product of grain with the F count is 4,246. The pairwise test of means by LSD test resulting that the variety fertilizer is varieties fertilizer 14. There is no significant different of the influence with varieties fertilizer 12, 11, 15, 3, 8, 4 and 6. Keywords: Balanced Lattices Designs, Analysis of varians, LSD Test