The Pygmalion effect: when a leader's expectations become the team's results

In the Greek myth as Ovid tells it, a sculptor named Pygmalion carves an ivory statue of a young woman so beautiful that he falls in love with it. He tends it like a living person, and the goddess Venus is moved enough to breathe life into it. More than two thousand years later George Bernard Shaw borrowed the name for his 1913 play about a professor of phonetics who turns a flower seller into a lady, the story that later became My Fair Lady.

Both stories share one idea: people can become what others believe they will become. In 1968 an American psychologist set out to see whether that held in real life.

The rats that were called clever

Robert Rosenthal started with rats. In 1963 he and a colleague gave students two groups of rats, saying one had been bred to be good at mazes and the other to be poor at them. In fact the groups had been split at random. No rat was special.

The rats called clever learned the maze noticeably faster. Not because they were cleverer, but because the students treated them differently: handling them more gently, with more patience, recording them more carefully. Rosenthal wondered: if this happened with rats, what would happen in a classroom?

Oak School, 1968

Rosenthal teamed up with Lenore Jacobson, the principal of a primary school in the city of South San Francisco, California, given the pseudonym Oak School in the book. At the start of the year every pupil sat a test introduced as a new Harvard instrument that could predict which children were about to make a leap in ability.

It was in fact an ordinary intelligence test. The researchers then gave the teachers a list of about a fifth of the pupils, saying these were the ones about to bloom. The names on the list had been chosen at random.

Eight months later the whole school sat the test again. The children on the list had gained more than their classmates, with the difference clearest in the first and second grades. The book reporting it, Pygmalion in the Classroom, became one of the most-cited psychology studies of the twentieth century.

The part of the story told less often

Stop there and it is a perfect inspirational story. But as soon as the book came out, researchers pointed to worrying features in the data. Some children's starting scores were implausibly low, suggesting the test did not suit children that young.

Over the following decades hundreds of studies tried to reproduce the result. The final picture is far more modest: teachers' expectations do affect pupils, but usually only a little. A 1984 analysis found a telling detail: the effect was clearest when teachers received the expectation within their first two weeks of knowing the pupils. Once teachers knew the children, what they had seen for themselves outweighed any suggestion from outside.

Later reviews also found the effect tends to be larger for pupils who are usually underestimated. So expectation works no miracles for anyone, but for someone used to being written off, being believed in once can make a real difference.

From the classroom to the training camp

In the early 1980s the organisational psychologist Dov Eden took the idea into the Israeli army. He told training officers that some trainees in their platoons had high command potential, based on test scores and earlier assessments. As at Oak School, those trainees had been chosen at random.

By the end of the course the high-expectation group performed markedly better on tests and was rated more favourably. In interviews the trainees themselves said their officers seemed more committed, explained things more carefully, and gave them more chances to try.

Eden called this Pygmalion in management. And the mechanism he found was concrete: the trainees who were expected to do well did not receive some mysterious energy. They received more time, more feedback, and more difficult things to do.

The dark side: the Golem effect

If high expectations push people up, low ones drag them down. Researchers named this side the Golem effect, after the Jewish legend of a clay creature made to protect people that gradually turns dangerous.

At work the Golem effect is usually quiet. A new hire makes a mistake in the first month. The manager starts giving them simpler tasks, checking more closely, asking their opinion less. Six months later the person really is behind their colleagues, because they have had less hard work and less trust. The manager looks at the result and concludes the first impression was right.

What a leader can actually do

The lesson from more than half a century of research is not to think positively about everyone. It is much more specific.

• Watch the first two weeks. The research suggests expectation weighs most when we do not yet know someone well. That is also when it is easiest to pigeonhole a new hire over one small mistake or a first impression.
• Look at how you hand out work. In the last month, who got the hardest and most interesting tasks? If it is always the same few people, your expectations may be deciding their opportunities.
• Separate comments on the work from comments on the person. "This report is missing the risk analysis" is very different from "you are not good at analysis". The first lets someone fix it; the second fixes them in place.
• Revisit the people you have filed as weak. Try giving them something harder than you usually would, with enough guidance. The result may surprise you.

The most memorable part of the Oak School story is probably not the IQ figures, which have been argued over for fifty years. It is that the teachers had no idea they were behaving differently. They were not deliberately favouring anyone. They simply waited a little longer while a child on the list thought about an answer, praised a little more specifically, called that child to the board one more time. Whoever receives signals like those usually reads them more clearly than whoever sends them.