This repository contains the experimental results and the code we used to analyze them.

• data_analyzer.rb is a Ruby script which calculates all the measures discussed in the article. See comments within the file for more information. Ruby 1.9.3 or higher is required to run the script (install Ruby, and run ruby analyzer.rb from the command line).
• The folder Measures contains the output of data_analyzer.rb (a separate csv file for every measure and every condition, files v364*.csv) and a machine-readable representation of the CHIELD causal graph.
• The folder Languages contains all the 495 languages (files lang*.csv) and basic data (file rough_data2.csv) about the participants. data_analyzer.rb uses these data as input.
• statistics.r is an R script which performs the statistical analysis described in section 3. The comments in the scripts explain, inter alia, how the models were selected and how the assumptions were tested.
• epsilon-graphs.R is an R script which creates the figures in the article.
• The archive clean_code.zip contains the full code for the webpage where the experiment was hosted.