论文部分内容阅读
正确估算样本含量是临床试验和现场调查设计中一个不容忽视的重要环节。样本太大,造成人力、物力和时间的浪费;样本太小,检验效能偏低,难以获得真实的结论:所谓样本含量估算就是根据统计学原理在保证研究结论有一定可靠性的条件下,所确定的研究所需的最小观测例数。样本含量估算是涉及概率统计的专门学问,要在有限的篇幅内讲清楚在各种情况下估算样本含量的方法相当困难。本讲仅从生殖医学研究的实际出发,集中介绍以率(例如,治愈率、有效率、生存率等)作为观察指标时,对总体率作估计及两样本率作比较时样本含量的估算。考虑到临床医生的实际情况,有必要先就一些必须具备和掌握的统计概念先加以扼要介绍和解释,然后再说明样本含量估算的具体方法。
Correctly estimating the sample content is an important part of the clinical trial and field survey design that cannot be ignored. The sample is too large, resulting in a waste of manpower, material resources, and time; the sample is too small and the test efficiency is low, and it is difficult to obtain a true conclusion: the so-called sample content estimation is based on statistical principles under the condition that the conclusion of the research is guaranteed to have certain reliability. The minimum number of observations required for a defined study. The sample content estimation is a special knowledge involving probability statistics. It is very difficult to explain in a limited space the method of estimating the sample content under various conditions. This lecture only focuses on the reality of reproductive medicine research and focuses on estimating the sample size when estimating the overall rate and comparing the two sample rates when rates (eg, cure rate, effectiveness, survival rate, etc.) are used as observation indicators. Taking into account the actual situation of the clinician, it is necessary to first briefly introduce and explain the statistical concepts that must be possessed and mastered, and then explain the specific methods for estimating the sample content.