Data and code from: Correcting for complexity: Incorporating trait-numbers enhances the performance of EMMLi in investigating modularity

Adams and Collyer (2019) evaluated the statistical performance of several approaches for quantifying morphological modularity and found that EMMLi had inflated type I error rates and a bias towards more complex models compared to the Covariance Ratio (CR) approach. They suggested that this may have been at least partly driven by the fact that AICc values from EMMLi do not incorporate trait numbers, but this was not verified. Here I present a performance analysis of a trait-number corrected EMMLi approach (“EMMLip”), showing that this ameliorates rates of false discovery and produces conservative results that favor less complex models. The corrected EMMLi approach was effective at differentiating models of modularity with varying between- and within-module covariation especially when effect size or dataset size were sufficiently large. While CR tests remained more effective at specifically detecting overall modularity, I found that CR tests are sensitive to varying within/between module ..., , # Data and code from: Correcting for complexity: Incorporating trait-numbers enhances the performance of EMMLi in investigating modularity (published in the journal Evolution) Authors: Jessica H. Arbour correspondence: Jessica H. Arbour; [[[e-mail]](mailto:[e-mail])] The associated dryad repository contains four files: (1) An R script (\"EMMLip.R\") that contains the updated function code for EMMLip. (2) An R script (\"EMMLi_simulations.R\") that provides the definitions for covariance matrices for conditions listed in the manuscript and loops for simulating datasets and applying the EMMLi and CR analyses. Select the appropriate conditions (2 or 3 modules, within or between module covariation) from the selections available in lines 13-102 and copy and paste that code block into the \"Simulation Loops\" code below (lines 113-192) where marked. The \"Simulation Loops\" section currently has the \"3 modules, different between module but same within module covari..., ,Adams与Collyer（2019）评估了多种量化形态模块化（morphological modularity）的方法的统计性能，发现EMMLi相较于协方差比（Covariance Ratio, CR）方法存在一类错误率虚高的问题，且倾向于选择更复杂的模型。他们推测该现象至少部分源于EMMLi的AICc值未纳入性状数量，但该推测未得到验证。本研究开展了一项针对校正了性状数量的EMMLi方法（记为"EMMLip"）的性能分析，结果表明该方法改善了假发现率，并生成了偏向更简单模型的保守性结果。经校正的EMMLi方法能够有效区分不同模块间和模块内协变模式的模块化模型，尤其当效应量或数据集规模足够大时。尽管CR检验在特异性检测整体模块化方面仍表现更优，但本研究发现CR检验对模块内/模块间协变的变化较为敏感…… # 数据与代码来源：《校正复杂度：纳入性状数量可提升EMMLi在模块化研究中的性能》 （发表于《Evolution》期刊） 作者：Jessica H. Arbour 通讯作者：Jessica H. Arbour；邮箱：[e-mail] 相关Dryad仓储库包含4个文件： 1. 一份R脚本（"EMMLip.R"），内含EMMLip的更新版函数代码。 2. 一份R脚本（"EMMLi_simulations.R"），提供了手稿中所列条件的协方差矩阵定义，以及用于模拟数据集、运行EMMLi和CR分析的循环代码。 请从第13至102行的可选配置中选择合适的条件（2或3个模块、模块内或模块间协变模式），将对应代码块复制粘贴至下方"模拟循环"部分（第113至192行）的指定标记位置。当前"模拟循环"部分的示例代码为"3个模块，模块间协变不同但模块内协变相同……"