This is the associated code for Horton and Chilton (2010) ```The Labor Economics of Paid Crowdsourcing.'' The paper is available at:

  1. My website:
  2. ACM Digital Library
  3. arXiv:
  4. SSRN:

Citation Info

 author = {Horton, John Joseph and Chilton, Lydia B.},
 title = {The Labor Economics of Paid Crowdsourcing},
 booktitle = {Proceedings of the 11th ACM Conference on Electronic Commerce},
 series = {EC '10},
 year = {2010},
 isbn = {978-1-60558-822-3},
 location = {Cambridge, Massachusetts, USA},
 pages = {209--218},
 numpages = {10},
 url = {},
 doi = {10.1145/1807342.1807376},
 acmid = {1807376},
 publisher = {ACM},
 address = {New York, NY, USA},
 keywords = {amazon's mechanical turk, crowdsourcing, human computation},


The repository is set up to make it transparent how the final PDF is constructed from the raw data. To replicate, you will need a Linux or Mac OX machine that has the following installed:

  1. R
  2. pdflatex
  3. make

To replicate the data analysis, you will need several R packages. Inspect the file writeup/labor_supply.Rnw to see if you are missing anything.

Download the repository from github

git clone

Build the PDF

From /labor_economics_of_paid_crowdsourcing, run:

cd writeup
make labor_supply.pdf

This should download the necessary data files and decrypt them. It will also run the statistical analysis in R and then produce plots and tables. Finally, it will build the pdf file using pdflatex, leaving the resultant labor_supply.pdf in the /writing folder. To see the actual steps that are being followed, you can inspect writeup\Makefile.

If you run into any trouble replicating, please contact me at