新世纪20年国内假设检验及其关联问题的方法学研究

Methodological research on hypothesis test and related issues in China’s mainland from 2001 to 2020

新世纪20年来国内假设检验方法学研究内容可分为如下几类:零假设显著性检验的不足、p值的使用问题、心理学研究的可重复性问题、效应量、检验力、等效性检验、其他与假设检验关联的研究。零假设显著性检验已经发展成一套组合流程:为了保证检验力和节省成本,实验研究需要做先验检验力分析预估样本容量,但问卷超过160人在传统统计中就没有必要这样做。当拒绝零假设时,应当结合效应量做出结论。当不拒绝零假设时,需要报告后验检验力;如果效应量中或大而检验力不够高,则可增加被试再行分析,但这一过程应主动披露,报告最后的实际p值并对可能犯的第一类错误率做出评估。

In the first two decades of the 21st century,the research of hypothesis test and related topics in China’s mainland can be divided into the following categories:Deficiency of null hypothesis significance test,use of p-value,repeatability of psychological research,effect size,the power of statistical test,equivalence test,and other research related to hypothesis test.NHST has been developed into a set of procedures as follows.To ensure power of statistical test and save costs,experimental research often needs to do a priori power analysis to estimate the required sample size,while questionnaire studies with more than 160 participants usually does not need to do so for traditional statistical analyses.When the null hypothesis is rejected,a conclusion should be made in combination with an effect size.When the null hypothesis is not rejected,the posterior power of statistical test needs to be reported;if the effect size is medium or large and the power of statistical test is less than 80%,more participants could be added for further analysis,but this process should be disclosed,the final p-value should be reported,and the type I error rate should be evaluated.