An open platform for science on Amazon Mechanical Turk.
psiTurk is a community-run project for the benefit of scientific research. There are many ways to contribute.
psiTurk will get better by having users. More users mean more interoperable, shareable code, which means more replicability in science. Like everything new there is a bit to learn, but we have tried to put together helpful guides. It is in heavy use in our lab and others!
If you use psiTurk to develop and deploy your experiment, why not contribute it to the experiment exchange? This will allow other people to easily replicate your design with (basically) the same population and procedures. Sharing your code is a very powerful statement that you support open science and the direct replication of behavioral studies. In addition, many students and researchers may benefit from being able to build new experiments based on your code.
If you find bugs or weird things why not report them? This will help everyone in the community. On the project's github page there is an Issue Tracker which allows anyone to post a bug report. Ideally this should be a genuine bug. General questions about how to use psiTurk can be directed to our Google group.
Remember, Google group for general questions about using psiTurk, Github Issues for genuine bug reports and feature requests. Also good idea to check both places before posting your question to see if it has been answered already.
An important part of any open-source tool is good documentation. We've tried to get things started with a nice set of videos and guides. However, if you have a better way to describe things, or came up against an issue you think would be helpful for other users to know about, please consider contributing to our documentation website. This is fully under version control via Github and anyone can lend a hand. A guide for contributors is available here.
In addition to contributing experiment code, we have a fully open development model. If you have ideas for how to improve psiTurk, please feel free to discuss the idea with the project maintainers (by creating an feature request Issue on github) and then create a pull-request on Github! A full guide to how to contribute is provided here and here (HTML).
Think of all the wasted effort across academic labs and those in industry when programmers re-implement the same experiment design over and over. By sharing your experiment code you help others replicate and extend your findings while also providing a great resource for education. Students and researchers can learn from your code and use the best ideas in their own science projects.