Raw Data and Analysis R Code for Stewart, Ungemach, Harris, Bartels, Newell, Paolacci, & Chandler "The average laboratory samples a population of 7,300 Amazon Mechanical Turk workers"

This analysis is done using the R programming language. https://www.r-project.org/

README.txt --- This file.

analysis.pdf --- Human readable file of the analysis with all R code and output, and all figures. All the numbers in the paper are taken from here.

analysis.Rnw --- The master file with all of the analysis code. The file is for knitr (http://yihui.name/knitr/) for reproducible analysis. Typing 'make' runs the analysis, generates the PDF figures and produces the output file analysis.pdf.

analysis.R --- Just the R code without the knitr markup. Generated by 'make analysis.R'. You can then run just the R code using 'R CMD BATCH analysis.R', which will generate the figures and the output of the analysis in analysis.Rout. Note, if you regenerate analysis.R you need to delete lines 11-21. These are the bits of code that take the non-anonymous data and swap the WorkerIds for UUIDs.

makefile --- Instructions for the make program.

HITs.RData --- The raw data in an R data.table (https://cran.r-project.org/web/packages/data.table/index.html). data.tables are awesome.

HITs.csv --- Data as a CSV file. Column headings are described in analysis.Rnw and analysis.pdf. This file is not used in the analysis, but would be your starting point if you are doing something outside R.

The figures from the paper:

joint_capture_prob_density.pdf
lab_capture_prob_density.pdf
lab_details_2.pdf
no_batches_plot.pdf
no_labs_plot.pdf
no_labs_plot_with_average.pdf
open_population_by_batch_size.pdf
open_population_by_pay.pdf
open_population_plot.pdf
worker_capture_prob_density.pdf