论文部分内容阅读
决策与判断研究中(甚至是实验心理学研究中)的许多问题关注某效应是否真实存在,及其背后的解释是什么。这些问题不关注该效应在某一特殊群体中是否显著。因此,可以通过分析单个被试来检验效应的显著性。如果有一个被试表现出了该效应,那么,这个效应就是存在的。根据这一观点,有时也可通过跨案例或者轮次(across cases or rounds)分析来验证效应的显著性,而不需要进行跨被试分析(across subjects)。这一观点也暗示在一些实验中可能存在反方向的效应。本文建议通过进行基于被试个体的统计分析来检验这样的效应,并介绍了一些不同形式的方法:PP概率图(probability probability plots);P值分布检验(tests of the distribution of p-values);分层取样多重检验的矫正(correction for multiple testing with step-down resampling)。这些方法都可以用于处理在对同样假设进行多重检验时无法避免的问题。另外,本文也列举了一些例子,其中有一部分例子存在反方向的效应,另一部分例子不存在。
Many of the questions in decision-making and judging research (even in experimental psychology) focus on whether an effect is real and what the underlying explanation is. These questions do not concern whether this effect is significant in a particular group. Therefore, the significance of the effect can be tested by analyzing individual subjects. If one participant shows this effect, then the effect exists. From this point of view, it is sometimes also possible to verify the significance of an effect across analyzes across cases or rounds without the need for cross subjects. This view also suggests that in some experiments there may be adverse effects. This paper proposes to examine such effects by conducting statistical analysis based on the individuals tested and introduces some different forms of methods: probability probability plots; tests of the distribution of p-values; Correction for multiple testing with step-down resampling. These methods can all be used to deal with problems that can not be avoided by multiple tests of the same hypothesis. In addition, this article also lists some examples, some of which have the opposite direction of the effect, the other part of the example does not exist.