【摘要】 研究单因素、双因素的方差分析,提出因素水平差异显著与否,主要取决于试验效应的平方和(或试验效应方差)与随机误差,引入平均相对试验效应,在其已知前提下,给出保证一定概率的因素水平差异显著的最小总试验次数、或每水平的最小平均重复试验次数.更多还原

【Abstract】 After studying analysis of variance on single factor and double factor,this paper reveals that if the difference between levels of the primes exist noteworthily is mainly depended on the sum of its experiment effects or variance of experiment effects and stochastic difference.The notion of average-relative-experiment-effects is introduced,if it’s known we can obtain the lest experimental units and the lest average replicated experimental units of per level to a certainty probability and significance of difference among factor levers are assured.更多还原