论文部分内容阅读
本研究结合近20亩玉米空白试验,通过不同抽样方法,进行了玉米大面积田间测产的研究与探讨。本文引述了已故谭启栋教授提出的典型众数抽样的理论依据和方法步骤。这一方法不仅在林业上适用,而且在农业方面起到同样作用。试验结果表明,典型众数抽样较同时同地进行的随机抽样平均亩产量更接近空白试验所测得的每亩平均产量。从亩产总体平均值95%的置信范围看,典型众数抽样亦较之随机抽样为精确。即使两种抽样方法结果相近似,但典型众数抽样在布点、经费、人力及时间等方面比随机抽样为少,仅及随机抽样的3/5与3/20。本文通过统计估计与分析阐述了典型抽样的误差小于同大小的随机抽样,且是最能代表总体的抽样。最后论述了典型抽样的必要性,并且结合农业生产的特点,提出了改进这一方法的设想。
In this study, a blank test of nearly 20 acres of maize was carried out. Through different sampling methods, a large area field corn yield test was studied. This paper quotes the theoretical basis and method steps of typical multivariate sampling proposed by Professor Tan Qidong. This approach is not only suitable for forestry but also for agriculture. The test results show that the average mu yield of a random sample collected at the same time and in the same mode is more similar to the average yield per mu measured by the blank test. Judging from the confidence range of 95% of the overall average per-mu yield, the sampling of typical patterns is also more accurate than random sampling. Even though the results of the two sampling methods are similar, the typical sample size is less than the random sample size, funding, manpower and time, etc., and is only 3/5 and 3/20 of the random sample. Through statistical estimation and analysis, the paper expounds the random sample whose error is less than the same size of the typical sample and is the sample which can best represent the population. Finally, the necessity of typical sampling is discussed. Based on the characteristics of agricultural production, the idea of improving this method is put forward.