Data supporting "Simple study designs in ecology produce inaccurate estimates of biodiversity responses"

A zip file containing an .R script with R code to run simulations and a .csv file containing data to load into R to run simulation. This data was extracted from 47 ecological studies and characterises the change in control and impact sites, as well as differences between these sites before an impact occurred, as per detailed methodology in the paper Christie et al. (2019). Data is also present in another .csv file showing results of when the accuracy weights presented in the paper Christie et al. (2019) were used to rerun the results of several meta analyses. We compare the results obtained from these accuracy weights with those obtained by using inverse variance. There is also a .txt file containing a link to a shiny app that will allow the calculation of accuracy weights based on Christie et al. (2019).

一个包含.R脚本文件的压缩包，该脚本文件中的R代码用于运行模拟，以及一个包含数据的.csv文件，用于在R中加载并运行模拟。该数据源自47项生态学研究，并描述了受控区和影响区的变化，以及在这些区域发生影响之前这些区域之间的差异，具体方法详见Christie等（2019）的论文。此外，还存在另一个.csv文件，显示了当使用论文中Christie等（2019）提出的准确性权重重新运行多项元分析的结果。我们比较了使用这些准确性权重获得的结果与使用逆方差获得的结果。此外，还包含一个.txt文件，其中包含一个链接，可以访问基于Christie等（2019）的方法计算准确性权重的shiny应用程序。