Thanks for presenting this paper!
While the randomized experiment artificially generates a circumstance that people get informed of whether other people were also provided with information, I am wondering how to distinguish the situation of scattering information with CK and without CK in the real world. The paper finds that broadcasting works better than seeding in a world without CK, however, how could a policy-maker effectively delivers information to the whole population while making people believe such information is not publicly available? I guess it is hard to figure out the clear-cut of "CK" and "No CK" in reality since this concept is really what people believe/assume. More importantly, it seems that the efficiency of modern information dissemination channels, e.g. television, news media, cannot create a "No CK" environment like the field experiment.
shugamoe
commented
6 years ago 
Thanks for presenting this paper! While the randomized experiment artificially generates a circumstance that people get informed of whether other people were also provided with information, I am wondering how to distinguish the situation of scattering information with CK and without CK in the real world. The paper finds that broadcasting works better than seeding in a world without CK, however, how could a policy-maker effectively delivers information to the whole population while making people believe such information is not publicly available? I guess it is hard to figure out the clear-cut of "CK" and "No CK" in reality since this concept is really what people believe/assume. More importantly, it seems that the efficiency of modern information dissemination channels, e.g. television, news media, cannot create a "No CK" environment like the field experiment.