DesignLibrary: A Library of Common Research Designs
DesignLibrary provides simple interface to build designs using the package DeclareDesign. In one line of code users can specify the parameters of individual designs and diagnose their properties. The designers can also be used to compare performance of a given design across a range of combinations of parameters, such as effect size, sample size, assignment probabilities and more.
Installing the design library
To install the latest stable release of DesignLibrary, please ensure that you are running version 3.4 or later of R and run the following code:
If you would like to use the latest development release of DesignLibrary, please ensure that you are running version 3.4 or later of R and run the following code:
devtools::install_github("DeclareDesign/DesignLibrary", keep_source = TRUE)
Contributing designs and designers
We welcome contributions to the library!
- You can submit static
DeclareDesign, which will live as properly attributed entries in the library on our website
- Or you can submit designer functions that generate designs, which may be added to the CRAN version of the package
This project is generously supported by a grant from the Laura and John Arnold Foundation and seed funding from Evidence in Governance and Politics (EGAP).