These files (or at least some of them) were probably used to compute the effect of removing outliers on the psychometric qualities of the measures.
It seems that outliers.r and outliers2.r were both used to create the output files that are reported in the tables and graphs papers.
It seems that the csv files have the values reported in the paper, and that they are the outpout files from those r files.
These files are uploaded five years after they were created and I have not tested them now, so the details here are based on my understanding of the code.