Since he is an insider, but not one of those people circling the wagons around frauds while the whole frontier is on fire, he has some good insight into how to restore social psychology.
The big problem in all psychology is the fetish with doing surveys and then compiling statistics and calling it science. If that is science, it is "a science of undergraduates" as legendary psychologist Paul Ekman phrased it earlier this year. But that doesn't mean it can't be valuable.
All fields have people that are guilty of losing their minds over false positives and massaging data, it just happens a lot more in psychology because the standards are subjective, the methodology is subjective and the results are rarely confirmed. That means psychologists have to be more rigorous in their work. Here is the chart Dr. Nussbaum (edit:
Basically, his point is that social psychologists need to have rules for the same reason homes have locks. Most people are honest, but although 1% will always try to break in and 1% never would, in between those poles locks keep people from the temptation of getting something through cheating.
He does a terrific job of discussing data peeking and basically being a trusted guide for young researchers who are wondering if the field is overrun by charlatans. It isn't, most people do good work, it is just that the charlatans figured out they could produce nonsense like that people with messy offices are more racist and get attention for it.
Crimes and Misdemeanors: Reforming Social Psychology by Dave Nussbaum, Random Assignment