医学正常值研究中所用样本含量的剖析  

A Dissection for the Sample Size in the Study of Range of Normal Values in Medicine

作　　者：田凤调[1] 张秋萍[1] 

机构地区：[1]中国预防医学科学院计算机与卫生统计研究室

出　　处：《中国卫生统计》1989年第2期29-33,共5页Chinese Journal of Health Statistics

摘　　要：本文就有关医学正常值的文献提供的三类四种正常值,针对不同研究目的,联系 CV 与RE_x 的分析,对 n 的选取作了分析讨论。事实表明:适量的样本含量是完全可以达到理想的结果的。作者认为:那种以大取胜,以多取胜,忽视测量、调查误差及经济开支的作法是不可取的。作者主张应在严格控制系统误差的前提下,本着少而精的原则来确定样本含量。The authors selected 3 groups and 4 subgroups inmedical literatures and analysed the CV and RE_xwith the sample size used.We noticed the complica-tion of n,some facts existed as follows:to controlthe systemetic error is not rigorous,the researchspsychological status take effect frequently with enlar-ging the sample size for safety.The authors believe that how to decide the sam-ple size must follow the principle which should selectnot much more but must assure precise,that is tosay,to control strictly the systemetic erro■s as ■ss- ■ble.Sometimes we should calculate the sample size,but we don't rigidly adhere to calculate n by for-mula,we should adjust the n in practice.In short,we should change the traditional ideawinning in majority and striving for cost effective-ness.

关 键 词：医学正常值范围 样本含量 

分 类 号：R1[医药卫生—公共卫生与预防医学]