Our project provides a new measure to be used by researchers interested in a complete and exhaustive data set of economic sanctions by the U.S. We use web-scraping and text-analysis techniques to collect a set of government documents where the U.S. Congress or President issued a sanction-related text on foreign nations or entities. We employ supervised machine-learning to identify all documents connected to sanctions-activities by the U.S. and their associated violations. We document the precise nature of the data-gathering process and code for each sanctions case a set of variables that provide comprehensive details, for example, about the senders, targets, issues, measures, conditionality, and trajectory.

Recently, our paper decscribing the data and making it available, has been conditionally accepted at the Journal of Peace Research. You can read it on SSRN here https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3790866.