The Pygmalion Effect at Work: How Leader Expectations Shape Performance

By Tara Minh · Updated Oct 11, 2026

The Pygmalion effect is the finding that people tend to perform in line with what their leader expects of them. When a manager genuinely expects strong results, she treats the person differently in small, often unnoticed ways, and those differences push performance toward the expectation. Low expectations work the same way in reverse.

That's the short version. The longer version is more interesting, because the research behind it is stronger in some settings than others, and a manager who believes it blindly can end up with a worse team, not a better one. This article covers where the idea comes from, how it works in practice, how it differs from its cousins (Galatea, Golem, Hawthorne), and a step-by-step way to use it without pretending expectations are magic.

What Is the Pygmalion Effect?

Key Facts

• Rosenthal and Jacobson's 1968 classroom experiment produced the original finding: in first grade, students teachers expected to "bloom" gained about 15 IQ points more than controls, and in second grade about 10 more. There was no significant difference in third grade, and in fifth and sixth grade the bloomers gained slightly less, per Jussim and Harber's 2005 review.
• A meta-analysis of 17 studies (58 effect sizes, 2,874 participants) found an average Pygmalion effect of d = 1.13 in management settings, with stronger results in military settings, among men, and for people initially held in low regard, per McNatt (2000), Journal of Applied Psychology.
• Across classroom replications, only slightly over one third found a statistically significant expectancy effect, per Jussim and Harber (2005).
• Seven field experiments testing workshop training in Pygmalion leadership found a small mean effect of r = 0.13, per Eden and colleagues (2000).

The Pygmalion effect is a type of self-fulfilling prophecy: a leader forms an expectation about someone, behaves in ways that fit that expectation, and the person's performance drifts toward it. The name borrows from the Greek myth of the sculptor who falls for his own statue, and from George Bernard Shaw's play of the same name.

It entered management through J. Sterling Livingston, who taught at Harvard Business School and also founded a management consulting firm (Harvard Business Review). His article "Pygmalion in Management" ran in HBR in 1969, with later versions in 1988 and 2003, according to a university library guide. His central argument was that what managers expect of subordinates, and how they treat them, shapes the results far more than most managers realize.

Here's the part people skip. The effect isn't about positive thinking. It's about behavior. A leader who privately believes in someone but never changes how she delegates, coaches, or gives feedback produces nothing. The mechanism runs through what the leader actually does.

Where It Came From: Rosenthal and Jacobson

Robert Rosenthal and Lenore Jacobson ran their well-known classroom study in 1968. Teachers were told that certain children, picked at random, were "bloomers" likely to show big intellectual gains. The labels were fiction. Later reviews describe the pattern in the results this way: the headline effect came almost entirely from first and second grade, and across both years of the study, self-fulfilling prophecies didn't occur in 8 of 11 grades examined (Jussim and Harber, 2005).

So the original study is both the origin story and the first caution. It showed that inaccurate positive expectations can become real in some conditions. It didn't show that expectations always do.

Livingston carried the idea from classrooms into the office. Managers, he argued, hold expectations of their people constantly, and people read them accurately.

How It Works: Rosenthal's Four Factors

Rosenthal later proposed (in the early 1970s) a four-factor account of how expectations travel from a teacher (or manager) to a person. The factors were developed from the classroom literature, and they translate to work without much stretching. The four are climate, feedback, input, and output. Here's what each looks like at work.

Notice what this means. You can hold a fair-minded view of everyone and still run a team where four small channels quietly favor the people you expect most from. Nobody sees a decision. They see a pattern, and the pattern becomes a gap in results.

Galatea, Golem, Hawthorne and Self-Fulfilling Prophecy

The Pygmalion effect has relatives that get mixed up in meetings. They differ on who holds the expectation and which direction it points.

The Galatea link matters most in practice. Leader expectations often work by changing what a person believes about themselves. That's why the best Pygmalion managers don't just say "I believe in you." They hand over a task that proves it.

And the Hawthorne contrast is worth remembering: Hawthorne is about being watched, Pygmalion is about being believed in. They can overlap in a pilot program, which makes both easy to over-credit.

What the Research Shows (and Doesn't)

The evidence is mixed in an honest way. It's strong in some settings and weak in others.

The strong end. The McNatt (2000) meta-analysis pooled 17 studies and found an average effect of d = 1.13 in management settings, which is large. Effects were stronger in the military, among men, and when people were initially held in low regard (McNatt, 2000). The author called for more work in ongoing work contexts and in for-profit organizations, which tells you where the thin spots were.

The Israeli military studies. Dov Eden and colleagues ran field experiments in the Israel Defense Forces where leaders were told that randomly assigned trainees had higher-than-usual potential. A 1995 pair of experiments by Dvir, Eden and Banjo is a good example of both sides: the effect was confirmed for men led by a man and for women led by a man, but not for women led by a woman, and not at all in the first experiment with female officer cadets (Dvir, Eden and Banjo, 1995). The point isn't that the effect is fragile. It's that it depends on context.

The classroom replications. Across studies that tried to replicate the induced-expectation effect on student IQ and achievement, only slightly over one third found a statistically significant result (Jussim and Harber, 2005). Those authors concluded that self-fulfilling prophecies in classrooms do occur, but the effects are usually small, tend not to accumulate, and may fade. They also raised a point that matters for managers: expectations can predict outcomes mostly because they're accurate, not because they cause the outcome.

The training problem. If the effect is real, can you train managers to produce it? Seven field experiments on workshop training in Pygmalion leadership style found little evidence of change in leaders or followers, with a small mean effect of r = 0.13. The authors explained the gap with the difference between results in controlled conditions and results in everyday work (Eden and colleagues, 2000).

Put together, a fair reading is this. Leader expectations can move performance, sometimes a lot, and the effect is most reliable where leaders have large influence over a person's tasks, feedback, and development. It isn't a guarantee, it isn't a motivational trick, and a one-day workshop won't install it.

Why It Matters for Managers

Two things follow.

First, expectations leak. You don't announce them, and people still pick them up from who gets the stretch project, who gets interrupted, and whose idea is taken seriously. If your expectations are inaccurate, you're distributing opportunity unevenly without noticing.

Second, accuracy matters as much as optimism. The Pygmalion effect isn't a license to inflate. Setting someone up for a goal they can't reach damages trust, which is the same problem expectancy theory describes when people stop believing effort leads to results. The useful version is calibrated high expectations paired with real support.

It also connects to how you view people in the first place. A manager who assumes people avoid work will treat them accordingly, which is the heart of McGregor's Theory X and Theory Y. Pygmalion is one way those assumptions become self-confirming.

How to Apply the Pygmalion Effect: Step by Step

1. Audit your expectations. List each direct report and write a one-line expectation of what they could deliver in the next six months. Next to each, note the evidence. Where you have none, that's a guess, not a judgment.
2. Look for the Golem pattern. Check which people you rate lowest and what's behind it. Is it a track record, or a first impression from a bad quarter two years ago?
3. Audit the four channels. For each person, count the last month's stretch assignments, specific feedback conversations, meeting contributions, and one-on-one time. Gaps show where your expectations are already acting.
4. Raise expectations with a reason. Pick one or two people you've been underestimating and set a goal one notch higher, tied to a skill they've shown. Pair it with a plan, not a pep talk. A goal-setting approach helps here: specific, challenging, and agreed.
5. Back it with input and feedback. Give the stretch task, then coach on it. This is where a coaching leadership style does the real work.
6. Make it safe to miss. Expectations land better when failure is survivable. Psychological safety is what turns a high bar into a challenge instead of a threat.
7. Say it out loud, once, specifically. "I'm giving you the renewal account because you handled the escalation well last quarter" does more than "I know you can do it."
8. Review and adjust. After 60 to 90 days, compare results with your starting notes. If your expectation was wrong, fix the expectation, not the person.

Template: Expectation Check-In

Copy this into your one-on-one notes and fill it in each quarter.

Examples by Business Function

These are illustrative scenarios, not case studies.

Do's and Don'ts

Limits and Criticisms

A few honest cautions.

• Accuracy can explain a lot. Part of the link between expectations and results is that good managers read their people correctly, per the Jussim and Harber review above.
• Effects vary by setting. The meta-analytic average comes mostly from military and training contexts, and the author flagged the need for more evidence from for-profit workplaces.
• Effects can fade. Classroom evidence suggests they don't reliably compound over time.
• Training doesn't guarantee transfer. The workshop studies found small effects.
• The negative direction is real but harder to study. Rosenthal and Jacobson only manipulated positive expectations, so the Golem side rests on other evidence.

None of this makes the idea worthless. It makes it a tool for examining your own behavior, not a promise that belief alone changes outcomes. For a related caution about a famous effect that turned out to be more complicated than its reputation, see the Hawthorne effect. To see how expectations shape the quality of a manager's relationship with each person, read leader-member exchange theory.

Expectations are one of the few management levers that cost nothing and are used constantly, mostly by accident. Write yours down, check them against evidence, and notice who gets the stretch work. Then adjust. For the motivation side of the same question, see Porter and Lawler's model.