In three separate NSF-funded studies, Barry Bozeman and his colleagues at University of Georgia and Georgia Tech have been studying the social, behavioral and, now, the ethical aspects of research collaboration.  The projects have involved multiple methods, including (1) survey research of a representative sample of science and engineering faculty in U.S. Carnegie Extensive universities (i.e. "Research I), n=2024,  (2) data from curricula vitae, and analysis of co-authoring, (3) citations (both articles and patents) data from the SSI/Web of Sciences and (4) interviews.   Studies have focused on: (1) numbers of collaborators and characteristics; (2) geographic and social proximity of collaborations, (3) roles of institutions in collaboration (especially university research centers); (4) collaboration and technology transfer between individual scientists and industry (i.e. the Bozeman Industry Involvement Scale); (5) collaboration by type of collaborator, especially graduate students; (6) relationship of collaboration to research productivity; (7) differences between men and women faculty with respect to collaboration patterns and criteria and strategies for developing criteria.  The latest phase of the work, just beginning this month, focuses on power dynamics and ethical issues in collaboration, include "ghost authors."  After a very brief review of the methods and chief findings of the projects, the presentation will focus on those aspects of collaboration of greatest interest to those attending the presentation.