pseudopeople is a project to develop a simulated version of various administrative datasets, including a simulated version of the confidential data behind the U.S. Census. With this, researchers can develop new techniques for linking datasets together that are compatible with the privacy protections necessary for such sensitive and consequential information - and do so without needing access to the real data.
The project is being developed by a team at the University of Washington, with a focus on transparent and meaningfully accountable processes and a team of advisors representing civil society, research and privacy interests. You can read more about our principles and processes here.
If you are interested in using the data, you can view a sample of it in our Python software package. For access to the full dataset, please see our access process. We also have a mailing list for regular updates and discussion