在单因素试验的基础上,提出多因素试验是生产力发展的必然结果。研究油菜氮磷钾肥合理用量,采用定性的试验方法巳不适用了。用来分析资料的方差分析,在判别不同施肥量的产量结果之间差异是否显著虽很有作用,但也很难从经济角度去寻找合理的施肥量。古典回归分析,长期以来只是被动地处理已有的试验资料,而对试验的安排几乎不提任何要求,对所求得的回归方程精度亦很少研究,这样不仅盲目的增加了试验次数,而且试验往往不能提供充分的信息,特别是在研究多因素问题中达不到目的。我们在前几年试验研究初步明确 Based on the single factor experiment, it is the inevitable result of the development of productivity that the multifactorial experiment is put forward. Research rape nitrogen, phosphorus and potassium fertilizer rational use of qualitative test methods have not applied. The analysis of variance (ANOVA) used to analyze the data is useful in discriminating whether the differences in yields of different fertilizers are significant or not, but it is also difficult to find a reasonable amount of fertilizers from an economic perspective. Classical regression analysis has long been only passive processing of existing test data, and almost no test on the arrangements for any requirements, the accuracy of the regression equation obtained seldom studied, so that not only blindly increased the number of tests, and Experiments often fail to provide sufficient information, especially in the study of multi-factor problems. Our pilot studies in the previous few years initially clear