Multisite field experiments using the (generalized) randomized block design that assign treatments to individuals within sites are common in education and the social sciences. Under this design, there are two possible estimands of interest and they differ based on whether sites or blocks have fixed or random effects. When the average treatment…