Look at the slew of "social and communications research" conducted since the 1960s. All of the findings have quickly become obsolete, and have nothing to do with what works and what doesn't work.