Step-by-step instructions for setting up a crowdsourcing experiment using Experigen and Amazon Mechanical Turk. Experigen was developed by Michael Becker and Jon Levine (https://becker.phonologist.org/experigen; https://github.com/tlozoot/experigen). The scripts for audio stimulus presentation and response recording shared here were customized by Daniel Szeredi (http://wp.nyu.edu/szeredi; https://github.com/aquincum) and Ildiko Szabo (https://wp.nyu.edu/szaboildi; https://github.com/szaboildi). Components include:
1. Preparing files and uploading to AWS S3 (with R script for generating and checking items.txt files)
2. Configuring Experigen files (with sample settings.js and design.js files)
3. Posting and managing HITs on Amazon Mechanical Turk (if you prefer Prolific Academic or another platform, you can follow steps 1-2 and ignore this component)
4. Downloading results (with sample R script)
*Note: This tutorial and the associated sample files come from a study involving rating children's /r/ sounds. For the task from Nightingale et al., see Experigen files uploaded under https://osf.io/z2j65/.*