How to use the code and data to replicate paper results:

  • Inside the archive there are two files: and
  • The code expects the following directory layout:





  • The name of project_root_directory is irrelevant, only relative paths matter for proper code execution.
  • The file should be unzipped into the code/ directory.
  • The file should be unzipped into the data/ directory.
  • The plots/ directory should exist for the code to generate plot files correctly.
  • From the code/ directory, open the file IncorporatingRefereeSuggestions.R.
  • In the file `IncorporatingRefereeSuggestions.R, edit the value of variable pathToData in line 10 – it should provide a full absolute path to project_root_directory/.
    • For example, on Windows it could look like this: pathToData <- "C:/Papers/RandomizatonInOnlineExperiments/"
    • Whereas on a Mac or Linux, it could look like this: pathToData <- "/Users/kgolyaev/Papers/RandomizatonInOnlineExperiments/"
  • The other file, FunctionsLibrary.R, does not need to be adjusted.
  • A number of R packages must be installed for the code to work correctly. They are listed in lines 13 through 20 of the IncorporatingRefereeSuggestions.R.
  • Running the code should produce all the numbers and plots that are used in the paper.
  • If anything does not work, please contact me at

There are no views created for this resource yet.


Field Value
Format ZIP
License CC-BY 4.0
Type data
Version 1
Publication Date 2018
Availability Delivery
Geographic Area (free)
Temporal Coverage (free)
Unit Type
Number of Units
Sampled Universe
Number of Variables
Last updated July 12, 2018
Created July 12, 2018