The Failure of Social Research

One of the seminal social-psychology studies, at the turn of the 20th century, asked a question that at the time was a novel one: How does the presence of other people change an individual’s behavior?   NY Times, 2017-10-18

You have to read the whole article to get the gist of just how mind-blowing this discovery was:  people behave differently if there are other people in the room.  I know– I was shocked too.

This stunning revelation was made by Norman Triplett– may his name endure forever.  The reverberations continue still.  People behave… differently… if there are other people in the room.

This is the psychology, which has revolutionized the art of discovering things that are already known, repackaging them as “research”, and impersonating science.

It was a revelation to the psychological establishment which had, before then, believed that people behave exactly the same way if there are people in the room or if there are not.

You know where the writer of the article in the New York Times is going when you see:

In 2000, Malcolm Gladwell, the author of the best-selling “Tipping Point,” applied irresistible storytelling to the science, sending countless journalists to investigate similar terrain and inspiring social psychologists to write books of their own.

Gladwell tried to argue that the Beatles were successful because they had practiced for 10,000 hours, over-looking the obvious fact that thousands upon thousands of artists “practice” for 10,000 hours and don’t go anywhere because they don’t have any talent.  Gladwell did have one aspect of genius: when he wasn’t making an outright error, he made people feel smart by packaging obvious truths with smug observation.

Let’s not forget this chessnut:

that once people have made a decision, they curiously give more weight to information in its favor.

I guess most psychology students have never read Shakespeare, the brilliant 17th Century psychologist who discovered this truth in his study of neurological disconnects, “Hamlet”.

The article is about a researcher named Amy Cuddy who claimed to prove that your body language not only expresses your attitudes and your confidence, but actually can change your attitude and confidence.  She urged people to adopt strong poses, to display confidence and assertiveness.  She did a study that, she claimed, proved that adopting the body language of a confident, aggressive person would give you confidence and aggression.

She proved it thusly: she recruited subjects by telling them that she was studying the use of an electrocardiagraph, to see if it worked just as well above the heart as below.  Then she arranged all the students in body poses, half confident and assertive, half shy, diminutive.   Then she studied their responses.  She even tested them to see if they were willing to bet on the outcome of a literal roll of the dice.

And she was just stunned by the outcome!  A bunch of smart students from affluent families who could afford college were recruited for a specious “study” reported that they gained confidence after adopting confident poses!  Quick, publish the results:

“That a person can, by assuming two simple 1-min poses, embody power and instantly become more powerful, has real-world, actionable implications.”

Sensational.  People will now pay this researcher thousands of dollars to deliver this breakthrough in person at conventions and conferences!  Give her a TED talk!  Get her on Oprah!  Calling Dr. Phil!

 Across disciplines, a basic scientific principle is that multiple teams should independently verify a result before it is accepted as true. But for the majority of social-psychology results, even the most influential ones, this hadn’t happened.

No, it didn’t.  It is a miracle of sorts that anyone actually bothered to try to verify the results– you get more headlines if you don’t.  What went wrong?

 Simmons believes that self-reports of power generally reflect what is called a demand effect — a result that occurs when subjects intuit the point of the study. Cuddy believes that studies can be constructed to minimize that risk and that demand effects are often nuanced.

Yes.  Shocking insight here: a study like this creates an artificial environment that can’t reliably be extrapolated to real life.

[whohit]The Failures of Social Research[/whohit]