We link high-frequency geocoded protest data during Bangladesh’s 2024 Monsoon Revolution, Facebook measures of protest activity and social connectivity, university…

We link high-frequency geocoded protest data during Bangladesh’s 2024 Monsoon Revolution, Facebook measures of protest activity and social connectivity, university enrollment and demographics to establish: (1) student-linked protests precede off-campus mobilization; (2) student concentration predicts both on- and off-campus protest, while non-student youth, educated adults, urbanization, and locations of “placebo” universities authorized but not yet operating do not; and (3) social connectivity to student protest locations predicts off-campus diffusion one week later. Universities appear to incubate mass uprising by making dissent visible, coordinated, and capable of spreading beyond campus. Such patterns are also evident in U.S. Vietnam War protests and in global data. That universities facilitate broader political mobilization can explain why governments often seek to restrict student activism. We thank Ariadna Gozhenko, Maryia Yemialyanenka, Elizaveta Volgina and Yaroslav Prokhorskoy for excellent research assistance, and seminar participants at the University of Houston, University of Michigan, and Emmerich Davies, Islamul Haque, Dean Yang, and Noam Yuchtman for comments. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. In addition toworking papers, the NBER disseminates affiliates’ latest findings through a range of free periodicals—theNBER Reporter, theNBER Digest, theBulletin on Health, theBulletin on the Economics of Alzheimer's, and theBulletin on Entrepreneurship—as well asresearch spotlights, lectures, and panel videos.