2.2 The dmetar package

In this guide, we want to make conducting meta-analyses as accessible and easy as possible for you as a researcher. Although there are fantastical packages like the meta and metafor package which do most of the heavy lifting, there are still some aspects of meta-analyses in the biomedical field and psychology which we consider important, but are not easy to do in R currently, particularly if you do not have a programming or statistics background.

To fill this gap, we developed the dmetar package, which serves as the companion R package for this guide. The dmetar package has its own documentation, which can be found here. Functions of the dmetar package provide additional functionality for the meta and metafor packages (and a few other, more advanced packages), which we will be using frequently in this guide. Most of the functions included in the dmetar package and how they can improve your meta-analysis workflow will be described in detail throughout the guide.

Although highly recommended, it is not essential to have the dmetar package installed to work through the guide. For each function of the package used in the guide, we will also provide the source code, which can be used to save the function locally on your computer, and the additional R packages those functions rely on.

However installing the dmetar package beforehand is much more convenient, because all the functions will already be pre-installed on your computer.



2.2.1 Installation of the dmetar package

2.2.1.1 Current R version

To install the dmetar package, the R version of your computer must be 3.5.2 or higher. If you have (re-)installed R recently, this will probably be the case. To check if your R version is new enough, you can paste this line of code into the Console, and then hit Enter.

R.Version()$version.string

This will display the current R version you have. If the R version is below 3.5.2, you will have to update your R version. A tutorial on how to do this can be found here.



2.2.1.2 Devtools

If you want to install dmetar, one package already needs to be installed on your computer first. This package is called devtools. So, if devtools is not on your computer yet, you can install it like we did before:

install.packages("devtools")



2.2.1.3 Installation process

You can then install dmetar using this line of code:

devtools::install_github("MathiasHarrer/dmetar")

This will initiate the installation process. It is likely that the installation will take some time, because several other packages have to be installed along with the dmetar package for it to function properly.

During the installation process, the installation manager may ask you if you want to update existing R packages on your computer. The output may look something like this:

These packages have more recent versions available.
Which would you like to update?

1: All                          
2: CRAN packages only            
3: None                          
4: ggpubr (0.2.2 -> 0.2.3) [CRAN]
5: zip    (2.0.3 -> 2.0.4) [CRAN]

Enter one or more numbers, or an empty line to skip updates:

When you get this message, it is best to first tell the installation manager that no packages should be updated. In this example, this means pasting 3 into the console and then hitting Enter. In the same vein, when the installation manager asks this question:

 There are binary versions available but the source versions are later:
 
 [...]
 
  Do you want to install from sources the package which needs compilation?
  y/n: 

It is best to first choose n (no). If the installation fails with these strategies, run the installation again, updating all packages this time.

“Failed to install ‘unknown package’ from GitHub”

After having installed devtools, and then proceeding with the installation of the dmetar package, it is possible that the installation does not start and that you get the following error message:

This error message can have various reasons, depending on your internet connection, location, operating system, and/or R version (see thread for a similar issue). If you get this error message, you can try to download the dmetar package directly from GitHub. On the GitHub site, click on the button Clone or download, and then on Download ZIP.

After downloading, unzip the ZIP-folder, and drag the dmetar-master folder inside to a meaningful location on your computer. You can then install the dmetar package directly from your computer using the code shown below.

  • For Windows:
devtools::install("C:\\Path\\To\\The\\Folder\\dmetar-master")
  • For Mac:
devtools::install("~/Path/To/The/Folder/dmetar-master")

Where you have to replace the path shown in the above examples with the path to the dmetar-master folder on your computer.

After the dmetar package is installed successfully, you can load it from your library:

library(dmetar)

If you are having trouble installing the package, you can copy the entire output of the installation process, including the code you entered, and the error message you receive, and send it to Mathias ().



banner