MetaVidence tutorials
MetaVidence does meta-analysis without code. You load a spreadsheet, choose the parameters and get publication-ready plots and tables, all in the browser, with nothing to install and without your data leaving your computer.
Where to start
- Get started shows how to open MetaVidence in the browser, and how to install R, RStudio and the package if you would rather run it on your own computer.
- The three-step flow walks the screens end to end: data, parameters and export. Every module works this way.
- Your module’s page covers only what changes in it, such as the required columns and the summary measure.
Every module ships an example dataset with 12 studies. It is worth running the analysis on it before using your own data, so you can see what each screen produces without preparing a spreadsheet.
Choosing the module
| You have… | Use |
|---|---|
| Event counts and sample sizes in both groups | Binary outcomes |
| Mean and SD, or median and IQR | Continuous outcomes |
| The effect already calculated, with CI or standard error | Pre-calculated effects |
| One proportion or mean per study, no comparison | Single arm |
| Three or more treatments compared indirectly | Network meta-analysis |
| 2×2 diagnostic accuracy tables | Diagnostic meta-analysis |
Have more than one outcome to analyse? Once you have picked the module, follow Multiple outcomes at once to run them all in a single pass.
How to cite
If MetaVidence produced results you are publishing, cite MetaVidence and the analysis packages. Citations are what keep this free academic tool alive and maintained, and recording which software produced your numbers is part of a reproducible analysis.