There has recently been a surge of interest in improving the replicability of second language (L2) research. However, less attention is paid to replicability in the context of L2 meta-analyses. I argue that conducting interrater reliability (IRR) analyses is a key step toward improving the replicability of L2 meta-analyses. To that end, I first discuss the foundation of IRR in the context of meta-analytic research. Second, I introduce two IRR measures, S index and Specific Agreement, which aid in improving the replicability of meta-analytic research. Third, I offer a flexible R program, meta_rate, to facilitate the conduct of IRR analyses for L2 meta-analyses. Fourth, I apply the R program to an actual meta-analytic L2 coding sheet to demonstrate the practical use of the IRR methods discussed. Finally, I provide interpretive guidelines to assist both L2 meta-analysts and journals with the transparent reporting of the IRR findings.

中文翻译：

---

第二语言元分析中的互信度

最近，人们对提高第二语言 (L2) 研究的可复制性产生了浓厚的兴趣。然而，在 L2 荟萃分析的背景下，对可复制性的关注较少。我认为，进行交互可靠性 (IRR) 分析是提高 L2 元分析可复制性的关键一步。为此，我首先在元分析研究的背景下讨论内部收益率的基础。其次，我介绍了两个 IRR 措施，S指数和具体协议，这有助于提高元分析研究的可复制性。第三，我提供了一个灵活的 R 程序，元速率，以促进对 L2 荟萃分析进行 IRR 分析。第四，我将 R 程序应用于实际的元分析 L2 编码表，以演示所讨论的 IRR 方法的实际用途。最后，我提供了解释性指南，以帮助 L2 元分析师和期刊透明地报告 IRR 调查结果。