The Diamond Ratio: A Visual Indicator of the Extent of Heterogeneity in Meta-Analysis ------------------------------------------------------------------------ **Abstract:** The result of a meta-analysis is conventionally pictured in the forest plot as a diamond, whose length is the 95% confidence interval (CI) for the summary measure of interest. The Diamond Ratio (DR) is the ratio of the length of the diamond given by a random effects meta-analysis to that given by a fixed effect meta-analysis. The DR is a simple visual indicator of the extent of heterogeneity in the meta-analysis, where increasing values of DR greater than 1.00 indicate increasing heterogeneity. We investigate the properties of the DR, and its relationship to four conventional but more complex measures of heterogeneity. We propose for the first time a CI on the DR, and show that it performs well in term of coverage properties. We provide example code to calculate the DR and its CI, and to show these in a forest plot. We conclude that the DR is a useful indicator that can assist students and researchers to understand heterogeneity, and to appreciate its extent in particular cases. Contents: --------- **Paper:** A pre-print of the paper can be found [here][1] **Files:** - Contour plots - The *[file][2]* contour_plots.R contains the R code needed to create the plots as seen in the paper. - The *folder* Contour Plots Data contains the data used to create the contour plots in the paper. The coverages data is merged together for easy use while the Confidence Intervals folders contains the confidence intervals used to find the coverages. These are split according to the value of Tau (these are interesting if you are interested in things like width of the intervals). - The [file][3] diamond_ratio_wrapper.R is a wrapper function for rma familly of functions in metafor. It is designed to add an object corresponding to the Diamond Ratio and its lower and upper bounds into an rma object. - The *[file][4]* Simulations.R creates the simulations used to generate the data. - Line plots - The *[file][5]* Line_plots.R contains code used to create the line plots seen in the paper. - The *[file][6]* Line_plot_coverages.RData contains the data used to create the plots. - Forest plots - The *[file][7]* ESCI_forest_plot.R contains code to generate the forest plot corresponding to Figure 1 in the paper (you may have to alter the window size to make everything fit). - The *[file][8]* Writing_to_Learn_Interventions_forest_plot.R contains code to generate the forest plot corresponding to Figure 2 in the paper (you may have to alter the window size to make everything fit). - The *[file][9]* Hospital_Stay_of_Stroke_Patients_forest_plot.R contains code to generate the forest plot corresponding to Figure 4 in the paper (you may have to alter the window size to make everything fit). - The *[file][10]* Coverage_simulations.R is an alternative file used to simulate data seen in the plots. **Page 2:** A brief introduction to the ESCI program **Page 3:** Examples of how the Diamond Ratio can be used in a meta-analysis. [1]: https://psyarxiv.com/f4xus [2]: http://osf.io/bvwkf/ [3]: http://osf.io/wtfns/ [4]: http://osf.io/xg2ez/ [5]: http://osf.io/a2v8q/ [6]: http://osf.io/9mv4a/ [7]: http://osf.io/8evhx/ [8]: http://osf.io/w8v9f/ [9]: http://osf.io/kxeyv/ [10]: http://osf.io/zqxwu/