The July Effect and the Beginner's Bubble Hypothesis

Several years ago, in what seems like a lifetime ago, I spent a semester working as a student athletic trainer for my college.  Technically, we were "athletic training students in training" (they called us rookies).  I went to college with plans to go to medical school and become an orthopedic surgeon specializing in sports medicine.  After a couple of years, I wasn't as sure that I wanted to go to medical school, so I saw this opportunity as an alternative career in sports medicine.  

Suddenly, I became the de facto "athletic trainer" and "team physician" for every one of my fraternity's intramural sports teams.  I was taping everyone up before their games, and I responded on the field whenever someone was hurt.  Here's the thing though - I had absolutely no clue what I was doing!  My experience in the athletic training room was limited to observing and shadowing the older student trainers, and my knowledge base was limited to one 3 credit hour class in the basic management of sports injuries ("Introduction to Athletic Training 101").    I suppose I had more knowledge than everyone else on our intramural team, but I really had no business trying to diagnose, let alone manage, my teammates' injuries.

You could definitely say that I was overconfident.  As it turns out, my overconfidence wasn't all that unique.  There is a phenomenon that has come to be called the Dunning-Kruger Effect, named after the two psychologists who first described it in a series of experiments published in 1999.  Dunning and Kruger started their paper with a vignette which I think perfectly illustrates the concept.  

Let me give you a story that illustrates this effect.  Apparently, a man named McArthur Wheeler (no relation btw) robbed two Pittsburgh banks in broad daylight in 1995 with no visible attempt at disguising himself.  He was promptly identified and arrested less than an hour after the surveillance videotapes were broadcast on the 11 o'clock news.  When the police later showed him the tapes, Wheeler reportedly responded with a dumbfounded look and said, "But I wore the juice."  He apparently was led to believe that rubbing lemon juice (which actually is one of the ingredients for invisible ink) on his face would render him invisible to the surveillance video cameras! Overconfidence or just stupidity?

Probably a little of both.

Dunning and Kruger showed that participants in four different studies consistently overestimated their level of competence.  Participants routinely scored in the bottom quartile on tests of humor, grammar, and logic (average 12th percentile), even though they rated themselves in a much higher quartile (average 62nd percentile).  Dunning and Kruger argued that "the skills that engender competence in a particular domain are often the very same skills necessary to evaluate competence in that domain - one's own or anyone else's."  As Charles Darwin said, "Ignorance more frequently begets confidence than does knowledge."

Incredibly, the cognitive bias identified in Dunning and Kruger's study appears to be fairly ubiquitous.  For example, 90% of professors at the University of Nebraska reported themselves to be above average in their ability to teach.  A 1976 study found that 70% of more than 1 million individuals reported that that they were above average in leadership ability and 60% reported that they were above average in athletic ability.  Check out the famous video of William Hung's audition for American Idol.  Regardless of the domain, people consistently overestimate (frequently by a lot) their own level of competence, expertise, skill, or ability.

Wharton professor Adam Grant suggested in his book, Think Again: The Power of Knowing What You Don't Know that this "overconfidence among beginners" (i.e., the Dunning-Kruger Effect) may explain what has become to be known as the "July Effect", the observation that the risk of medical errors and complications increase during the month of July, when newly graduated physicians start working for the first time in hospitals.  He offered an additional study by psychologists Carmen Sanchez and David Dunning (yes, the same Dunning in the Dunning-Kruger Effect) as supportive proof.  Study participants played the role of new physicians who had to diagnose "zombie disease" during a "Zombie Apocalypse" scenario.  Subjects were initially tentative and less confident in their diagnoses.  However, as they gained a little experience, they became overconfident with an inflated sense of their level of competence.  Sanchez and Dunning explained their experimental observations with something they called the "Beginner's Bubble Hypothesis" (more on this in the next post).  

Confidence can be a valuable trait for leaders (see, for example, "How confident leadership helps teams thrive" or "20 things leaders with real confidence do differently").  However, overconfidence is not.  Nobel laureate and cognitive psychologist Daniel Kahneman wrote in his book Thinking, Fast and Slow that if he could wave a magic wand to make one of the many cognitive biases disappear forever, it would be overconfidence.  There is certainly a balance here, and Adam Grant defines it as "confident humility" ("a leader's ability to make the right decision while acknowledging that they need others to do it right").  I will return to this whole discussion on the Dunning-Kruger Effect and the concept of "humble confidence" or "confident humility" as an ideal leadership trait in my next couple of posts.