Regression sum of squares of symmetric balanced incomplete block design consisting of another one missing observation by accident

Abstract

A balanced incomplete block design (BIBD) is the effective way to help analyze a treatment variable and one block variable under the condition where experimental units are limited. This paper considered the symmetric balanced incomplete block design (SBIBD) with t treatments and t blocks of size t-1. The trouble of analysis is caused if another one missing value unintentionally occurs in the experiments. The SBIBD with another one missing value was analyzed by means of the exact approach, i.e., the general regression significance testing procedure. There was no ready-made formula in the past. Hence, the paper provided the mathematical formulae for the fitted parameters and the regression sum of squares for the full effect model of experimental data.