Schmalz Xenia, Biurrun Manresa José, Zhang Lei
Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital, Ludwig-Maximilians-Universitat Munich.
Institute for Research and Development in Bioengineering and Bioinformatics, National Scientific and Technical Research Council.
Psychol Methods. 2023 Jun;28(3):705-718. doi: 10.1037/met0000421. Epub 2021 Nov 15.
The use of Bayes factors is becoming increasingly common in psychological sciences. Thus, it is important that researchers understand the logic behind the Bayes factor in order to correctly interpret it, and the strengths of weaknesses of the Bayesian approach. As education for psychological scientists focuses on frequentist statistics, resources are needed for researchers and students who want to learn more about this alternative approach. The aim of the current article is to provide such an overview to a psychological researcher. We cover the general logic behind Bayesian statistics, explain how the Bayes factor is calculated, how to set the priors in popular software packages to reflect the prior beliefs of the researcher, and finally provide a set of recommendations and caveats for interpreting Bayes factors. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
贝叶斯因子在心理科学中的应用日益普遍。因此，研究人员理解贝叶斯因子背后的逻辑以正确解读它，以及理解贝叶斯方法的优缺点就显得很重要。由于心理科学家的教育侧重于频率统计学，对于想要更多了解这种替代方法的研究人员和学生来说，需要相关资源。本文的目的是为心理学研究人员提供这样一个概述。我们涵盖了贝叶斯统计学背后的一般逻辑，解释了贝叶斯因子是如何计算的，如何在流行的软件包中设置先验以反映研究人员的先验信念，最后提供了一组解读贝叶斯因子的建议和注意事项。（PsycInfo数据库记录（c）2023美国心理学会，保留所有权利）
