摘要
当消除了系统误差时，则量值的分布遵正态分布。根据这一基本原理，本文对小样本可疑值的取舍提出了一种新方法。该方法先检验小样本是否遵从正态分布。凡遵从正态分布的小样本，全部则量值都应该保留；对不遵从正态分布的小样本，再用统计量Ｆ进行舍弃，舍弃后的新样本，若遵从正态分布，余下的则量值都应该保留；若不遵从正态分布，还得继续舍弃，如果舍弃的则量值个数较多，仍不遵从正态分布，说明则量过程存在较大的系统误差或样本来自非正态总体。
When the systematic erros is eliminated ,The measuvemenst obey the mormal distribution,by the basic idea,This Paper advence a new method of the rejection of a Vesuts in small Samples The method is first to test small sample's normal distribution,If the small samples obey the normal distribution,All the measurments should be retained;If the small samples don't obey the normal distribution,They should be rejected by the meaus of statistic F,For remaiming new sampses ,If they obey the normal distribution,Remining mesurements should be retained; otherwise,It will Continue rejection,If rejecting measurement is relatire much and don't obey normal distribution, It is prored that there are relative large systematic error or the sample come from non-normal total.last,This Pepat appraise on the rejection of a vesuts in small samples.
出处
《赣南师范学院学报》
1997年第3期62-68,共7页
Journal of Gannan Teachers' College(Social Science(2))
