假设检验与P值的再认识

Recognition of Hypothesis Test and P Value

假设检验是统计推断的一项主要内容,而P值在基于概率性反证法的假设检验中又扮演者重要角色。在之前的学习过程中,我们往往在熟识假设检验的基本步骤之后,却对P值的相关理论认识不清,没有正确理解利用P值进行检验的思想。本文首先在假设检验的背景下,进行对检验的P值与显著性水平的比较讨论。其次在对P值不同角度的解释中加深对其的认识,厘清对P值的错误认识,深化对P值的正确理解,了解其在统计研究的优势和局限,以此规范P值在显著性检验中相关表述。最后将P值计算的理论方法应用于实践,通过运用Excel和R软件实现在实例中P值的计算,更好地理解P值在假设检验中的作用。

Hypothesis testing is one of the main contents of statistical inference, and P value plays an important role in the hypothesis test based on probabilistic disproportion. In the previous learning process, we often get familiar with the basic steps of hypothesis testing, but we don’t know the relevant theory of P value, and do not correctly understand the idea of using P value to test. In this paper, under the background of hypothesis test, we discuss the comparison between P value and significance level. Secondly, we should deepen our understanding of P value from different angles, clarify the wrong understanding of P value, deepen the correct understanding of P value, and understand its advantages and limitations in statistical research, so as to standardize the expression of P value in significance test. Finally, the theoretical method of p-value calculation is applied to practice. Through the use of Excel and R software to realize the calculation of P value in the case, the role of P value in hypothesis testing can be better understood.