This code bundle is a package for estimating impacts and cross-site variation in multi-site or blocked RCTs. It also has a bunch of routines for running multisite simulation scenarios, in particular code for generating multisite data parameterized by parameters such as the intraclass correlation coefficient and degree of cross-site impact variation. It is used in a variety of research projects connected with the Miratrix CARES Lab (https://cares.gse.harvard.edu/) and other research that Miratrix is part of.
A core method of the package is compare_methods()
, which uses a bunch of different estimation strategies on multisite data to estimate the ATE, and then makes a table of the results.
Also see block_estimator()
which implements the various variance estimators for blocked experiments as discussed in the Pashley & Miratrix JEBS papers.
You can install the package from github using the devtools
package:
library( devtools )
devtools::install_github("lmiratrix/blkvar")
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.