More on social psychology

If I recall correctly, some time ago I discussed the replication crisis in social psychology. It has been very common in the field to do studies with small sample sizes (often consisting entirely of college students who are taking the investigator's course.) These may be unblinded or have other methodological weaknesses, but they get published anyway. Researchers who have systematically tried to replicate them have found that most of them to not get the same results. Here's a little on-line lesson about the problem, if you are interested.

 

However, the field has a deeper problem. Some of the most famous social psychology experiments, which have had major cultural influence and affected politics and policy, were fraudulent, including the so-called Stanford Prison Experiment, .which you may well have heard of. The "experiment" -- which it was not -- was carried out by Philip Zimbardo, who became the most famous psychologist in America on the strength of it, and later president of the American Psychological Association. Zimbardo claimed that he had randomly assigned the role of prison guards and prisoners to students (all male) who then engaged in a simulated prison experience in a basement. 

 

Zimbardo claimed in his publications that the "guards" spontaneously began to abuse and mistreat the prisoners, demonstrating that  "evil is brought out by circumstance. The authors, in their conclusions, claimed that, per the linked Vox article "innocent people, thrown into a situation where they have power over others, will begin to abuse that power. And people who are put into a situation where they are powerless will be driven to submission, even madness." 

 

But it was a hoax. The guards had been instructed to mistreat the prisoners, which they did with great reluctance, and the prisoners also enacted their submission, because they understood that was what Zimbardo wanted them to do. The entire thing was staged, it was not an experiment at all. Yet it is cited in introductory psychology textbooks, has been the subject of movies and books, and Zimbardo was twice invited to testify before congress.  So yes, Zimbardo has proved that at least some people are capable of evil, but no in the way he claimed.

Other famous social psychology experiments have turned out to be fraudulent, or at least badly misrepresented and misunderstood.  These include the "Robbers Cave" supposed replication of the Lord of the Flies phenomenon, and the very famous Milgram experiment in which research subjects supposedly agreed to deliver painful and even possibly fatal electrical shocks to people they thought were actual research subjects. The latter wasn't exactly fraudulent but it didn't really show what it purported to show. (Read the Vox article for details.)

So why was this bad science accepted so uncritically? It told people what they wanted to hear. The cultural moment encouraged people to think the worst of humanity. The phenomenon of Nazism was not far in the rear view mirror when these took place, and yeah, that proves that people are capable of great evil. However, it does not prove that Nazism is characteristic of human society at most times and places. It can happen, but it can just as well not happen. And right now, we all have important choices to make.