School of Psychology, University of Sussex, Brighton, UK.
Centre for Health Services Studies, University of Kent, Canterbury, UK.
Addiction. 2018 Feb;113(2):240-246. doi: 10.1111/add.14002. Epub 2017 Sep 18.
To illustrate how Bayes factors are important for determining the effectiveness of interventions.
We consider a case where inappropriate conclusions were drawn publicly based on significance testing, namely the SIPS project (Screening and Intervention Programme for Sensible drinking), a pragmatic, cluster-randomized controlled trial in each of two health-care settings and in the criminal justice system. We show how Bayes factors can disambiguate the non-significant findings from the SIPS project and thus determine whether the findings represent evidence of absence or absence of evidence. We show how to model the sort of effects that could be expected, and how to check the robustness of the Bayes factors.
The findings from the three SIPS trials taken individually are largely uninformative but, when data from these trials are combined, there is moderate evidence for a null hypothesis (H0) and thus for a lack of effect of brief intervention compared with simple clinical feedback and an alcohol information leaflet (B = 0.24, P = 0.43).
Scientists who find non-significant results should suspend judgement-unless they calculate a Bayes factor to indicate either that there is evidence for a null hypothesis (H0) over a (well-justified) alternative hypothesis (H1), or that more data are needed.
说明贝叶斯因子如何有助于确定干预措施的有效性。
我们考虑了一个基于显著性检验得出不当结论的案例，即 SIPS 项目（明智饮酒筛查和干预计划），这是在两个医疗保健环境和刑事司法系统中进行的一项实用、集群随机对照试验。我们展示了贝叶斯因子如何澄清 SIPS 项目中的无显著发现，从而确定这些发现是否代表缺乏证据或缺乏证据。我们展示了如何对预期效果进行建模，以及如何检查贝叶斯因子的稳健性。
单独来看，三个 SIPS 试验的结果大多没有提供信息，但将这些试验的数据合并后，就有中度证据支持零假设（H0），即与简单的临床反馈和酒精信息传单相比，简短干预没有效果（B = 0.24，P = 0.43）。
发现非显著结果的科学家应暂停判断——除非他们计算出贝叶斯因子，表明存在支持零假设（H0）而不是（有充分理由的）替代假设（H1）的证据，或者需要更多的数据。
