This is all obviously wrapped up in the “replication crisis”, which is impacting most sciences which use some statistics and are characterized by modest and complex causal effects (social and biological sciences in particular).
Obviously, I am no social psychologist, but can I just say that everyone knew there was a problem in the field a long time ago. By everyone, I mean psychologists. I had friends who worked in related fields who told me as early as 2006 not to trust anything coming out of social psychology. Others described how p-hacking and “unconscious” data manipulation was relatively common in psychological experimentation, and the personal stands they had to take to avoid engaging in the practices which were ubiquitous.
Finally, there’s the demand-side problem: ideas like power posing, implicit bias, and stereotype threat, offer neat, clean, and powerful explanations and oftentimes solutions for social problems. Wonkish Left-liberal publications and pundits in particular literally mine the literature to “show what the science says” (don’t worry, it overwhelmingly confirms prior beliefs).
As a testament to the power of the likely wrong (not robust) viewpoints, consider that John Bargh has a book out published this month, Before You Know It: The Unconscious Reasons We Do What We Do. Bargh’s work was one of the first research programs to be critiqued in the early 2010s. Of course, he doesn’t agree with the critics, but it does strike me that the field as a whole (e.g., people like Daniel Kahneman) believe that these subliminal effects are much weaker than originally claimed, at best. Nevertheless, Bargh is going to sell his books, and people in coffee shops and airports all over the country are going to eat it up. The reality is that subliminal effects are probably not that different than Freudianism; there may be something there, but it isn’t nearly the deal that the practitioners claimed it was.
Addendum: Arguably, the candidate gene studies of the 1990s and early 2000s and under-powered GWAS of the mid to late 2000s, fall into the same category. But I don’t know any geneticists who defend these results or engage in that analytic paradigm in 2017.
Addendum II: Since some have asked, The Invisible Gorilla is a very good book. It helped crystallize many of my skepticisms of the psychology to pop social science conveyer belt.