We conducted a meta‐analysis on the most commonly used forensic polygraph test, the Comparison Question Test. We captured as many studies as possible by using broad inclusion criteria. Data and potential moderators were coded from 138 datasets. The meta‐analytic effect size including inconclusive outcomes was 0.69 [0.66, 0.79]. We found significant moderator effects. Notably, level of motivation had a positive linear relationship with our outcome measures. Information Gain analysis of CQT outcomes representing the median accuracy showed a significant information increase over interpersonal deception detection across almost the complete range of base rates. Our results suggest that the CQT can be accurate, that experimental studies are generalizable, and no publication bias was detected. We discussed the limitations of the field research literature and problems within polygraph profession that lower field accuracy. We suggest some possible solutions.

中文翻译：

---

比较问题测谎仪测验的综合荟萃分析

我们对最常用的法医测谎仪测试（比较问题测试）进行了荟萃分析。通过使用广泛的纳入标准，我们捕获了尽可能多的研究。数据和潜在的主持人来自138个数据集。包括不确定结果的荟萃分析的影响大小为0.69 [0.66，0.79]。我们发现了重要的主持人效应。值得注意的是，动机水平与我们的结果指标呈正线性关系。代表中位数准确性的CQT结果的信息增益分析显示，在几乎完整的基本费率范围内，与人际欺骗检测相比，信息量显着增加。我们的结果表明，CQT可以准确无误，实验研究是可推广的，并且未检测到发表偏倚。我们讨论了田野研究文献的局限性以及测谎仪行业中降低田野准确性的问题。我们建议一些可能的解决方案。