Golem Effect

The Golem Effect is a psychological phenomenon where low expectations placed on individuals lead to poorer performance. It’s a self-fulfilling prophecy: a belief that makes itself come true because it changes how people act. Negative beliefs about someone’s potential can hinder their actual achievement.

Key Takeaways

Negative Pygmalion Effect: An authority figure’s low expectation becomes a self-fulfilling prophecy that lowers a person’s performance (Babad, Inbar & Rosenthal, 1982).
Named From Folklore: Babad, Inbar, and Rosenthal (1982) coined the term after the golem, a clay creature of Jewish folklore that turns destructive on its maker.
How It Spreads: Differential treatment carries the effect: a colder climate, less input, fewer chances to respond, and thinner feedback (Rosenthal, 1973; Harris & Rosenthal, 1985).
Hard To Study: Inducing damaging low expectations is unethical, so evidence comes mainly from natural bias or from experiments that reduce an existing golem (Oz & Eden, 1994).
Beyond the Classroom: Contemporary research shows work teams with negative shared beliefs about “followers” perform worse, via members’ lower self-efficacy and effort (Leung & Sy, 2018).
Real but Modest: Like all expectancy effects, this one is real but modest on average, falling hardest on already-disadvantaged pupils and employees (Sorhagen, 2013).

In This Article:

Origins of the Golem Effect

The golem effect has two roots: a decades-long study of experimenter bias, and a single, famous field experiment in a real school. Both trace back to the psychologist Robert Rosenthal.

In the early 1960s, Rosenthal ran laboratory studies on expectation bias. The results were striking. “Maze-bright” rats outperformed “maze-dull” rats that had actually been assigned at random, for one simple reason: the experimenters who expected more handled and scored their animals differently.

The Pygmalion in the Classroom Study (Rosenthal & Jacobson, 1968)

Rosenthal then crossed into a real school. He worked with Lenore Jacobson. Together they produced Pygmalion in the Classroom, the study that founded this whole field.

Aim: To test whether teachers’ expectations, based on false information, could produce real changes in pupils’ intellectual performance in a genuine classroom.
Method: Every child at a Californian elementary school took an IQ test framed to staff as identifying children about to “bloom” intellectually. Teachers were then told a random 20% of pupils were “bloomers,” and the test was repeated a year later.
Results: The randomly chosen “bloomers” gained more IQ than their classmates, with the gains concentrated in the youngest children.
Conclusion: Since bloomer status was random, only the teachers’ belief could explain the difference, implying the converse too: an unjustified low expectation should be able to depress performance.

The study was hugely influential but drew sharp criticism. Later reviewers argued the IQ test was unreliable for young children. A meta-analysis found the effect real but small, strongest when a false expectation is planted before teachers meet the children (Thorndike, 1968; Raudenbush, 1984).

Elashoff and Snow (1971) went further. They argued the headline gains rested on just a handful of extreme scores. Their reanalysis became one of the most cited critiques of the whole expectancy literature.

That debate still shapes how the founding study is read today.

This traces to a bigger idea. The golem effect is one case of the self-fulfilling prophecy, named by sociologist Robert Merton. The theorem behind it is simple. If people define a situation as real, it becomes real in its consequences.

Here, the “definition” is one person’s low opinion of another’s ability. The “consequence” is that person’s reduced performance. The term for the negative case, golem, arrived a decade later, when Babad, Inbar, and Rosenthal (1982) named it directly.

What the Research Says

Babad, Inbar, and Rosenthal (1982) coined the term golem effect, testing how a teacher’s own bias shapes their treatment of pupils they expect little of.

Aim: To test whether teachers who are more prone to bias produce stronger expectancy effects, in both directions, than teachers who are less prone to it.
Method: Researchers first sorted teachers into high-bias and low-bias groups, then gave every teacher fabricated information labelling pupils as high- or low-potential and measured both the teachers’ behaviour and the pupils’ performance.
Results: High-bias teachers treated “low-potential” pupils more negatively and produced weaker task performance from them, while conferring a matching advantage on “high-potential” pupils. Low-bias teachers showed far weaker effects either way.
Conclusion: Golem effects are concentrated in a bias-prone minority of authority figures, not spread evenly across all of them; the danger lies in who holds an expectation and how rigidly.

Babad, Inbar, and Rosenthal named four mechanisms. Biased teachers create a colder climate for pupils they expect little of.

They also teach less and easier input, give fewer chances to respond, and offer thinner feedback (Rosenthal, 1973; Harris & Rosenthal, 1985). None of this requires the teacher to intend harm. The differential treatment is largely unconscious.

Implications for Real Life

The research discoveries concerning the Golem effect unveil to us the far-reaching consequences of the Golem effect.

These may readily manifest in everyday interactions between parents and children, teachers and students, and employers and employees in various settings (InnerDrive, 2022):

In Education

Teachers may impact students” performance in manifold ways. For instance, they may be nicer toward students upon whom they may have placed high expectations while paying less attention to those they have perceived as possessing less potential.

Such teachers may also treat these student populations vastly differently in providing constructive feedback and opportunities for growth, and bestowing praise (which could enhance the students’ self-esteem).

Students placed under high expectations may gain more confidence. Those who face negative expectations, however, can find themselves in a vicious cycle wherein low expectations engender lower levels of accomplishment, which in turn enhance such adverse expectations.

At Work

Employers often have varying levels of expectations for their employees based on their perceived skills, motivation and potential.

Consequently, supervisors may dismiss their ideas. They may also micro-manage tasks and lower the goals set for less-favoured employees.

Such treatment may in fact lead to employees engaging in conduct conforming to low managerial expectations. This could create a cycle of low productivity conducive neither to the future of the organization nor to the well-being of the employees.

In Sports

Finally, the Golem effect in sports cannot escape notice. Coaches must make crucial decisions about players’ talents and potential. These choices shape who gets picked for specific roles.

However, this can also readily result in diverse external displays of expectations on a coach’s part, over different individuals in a team (Mahoney, 2011). These expectations may in turn, significantly impact the way various members of the team perform.

Players from whom little accomplishment is anticipated might perform to conform to the stereotypes imposed upon them. Such outcomes, evidently, would adversely affect their athletic careers.

In Healthcare

Staff expectations shape patient care too. A clinician who expects little of a rehabilitation patient may set less ambitious targets and give thinner instruction.

This can translate into less rehabilitative effort and a slower recovery, which is part of the case for structured, criterion-based assessment and bias-awareness training in care settings.

Overcoming the Golem Effect

As serious as the Golem effect is, we are not without means to mitigate and overcome it. A two-step approach could be utilized for this purpose:

Cultivate Awareness

Individuals in supervisory roles should become aware of their capacity to shape the conduct of their subordinates. Parents, teachers, coaches and employers should realize this. Their interactions with those under their authority readily convey their expectations for such individuals.

Developing strategies to guard against biases in such expectations is an essential step toward mitigating the Golem effect.

The same goes for the other side. Children, students, athletes and employees should develop their own awareness of how those above them might be shaping their behavior.

This would enlighten their understanding of how others’ low expectations could hinder their capacity for accomplishment, and immensely aid them in cultivating resilience.

Change Expectations

Those occupying positions of authority can replace negative expectations with positive ones.

Instead of expecting their subordinates to constantly fail, supervisors can affirm their subordinates’ latent capabilities and set high standards of excellence for them, thereby aiding the improvement of their performance.

Likewise, those under authority may set higher expectations for themselves, and then strive to attain such standards, thereby raising their own performance.

Two field experiments show this works in practice. Oz and Eden (1994) found that when military instructors stopped treating low fitness scores as a sign of low ability, the expected golem effect shrank or disappeared.

Davidson and Eden (2000) went further. Raising the expectations that leaders held of disadvantaged women trainees, whose performance had been assigned at random, directly improved those trainees’ results.

One more risk remains. Authorities should also watch for the “sustaining-expectation effect,” failing to revise a low expectation even after a person’s performance genuinely improves (Rubie-Davies, Hattie, & Hamilton, 2006). Left unchecked, this can quietly hold someone back with no single dramatic incident of bias.

Structure helps here. Awareness alone will not fix an unconscious bias, so the strongest defences are structural. These include criterion-referenced rather than norm-referenced judgement, behaviourally anchored rating and feedback, and blind or anonymised assessment where it is feasible.

Critical Evaluation of the Golem Effect

Contemporary Research

Modern work asks two questions the classic studies left open: does the golem effect operate at the level of teams, not just one teacher and one pupil? And how large is it, really?

Aim: To test whether a work group’s shared, largely negative beliefs about “followers” function as expectations that create golem effects, and how any such effect reaches an individual member.
Method: Leung and Sy (2018) surveyed 202 followers and 101 leaders and tested whether negative group-level beliefs about followers predicted lower follower performance, through the followers’ own self-efficacy and effort.
Results: Negative group beliefs did predict lower performance, and this ran through reduced self-efficacy and effort. The more a group saw a “follower” as incompetent, the more members’ behaviour matched that image.
Conclusion: The golem effect is not only a classroom phenomenon; a group’s shared low expectations can shape individual underperformance too, through self-efficacy and effort.

This is cross-sectional evidence, so it cannot prove cause and effect the way a field experiment could. Around it, other research adds nuance.

Expectations often predict outcomes mainly because they are accurate, so the truly self-fulfilling slice is real but small (Jussim & Harber, 2005). The effects also tend to fade across successive teachers rather than build up (Smith, Jussim, & Eccles, 1999).

Yet its importance lies in who it hits hardest. Naturally low expectations fall most heavily on disadvantaged and stereotyped pupils and employees (Sorhagen, 2013; Madon, Jussim, & Eccles, 1997).

Rival Explanations

Rosenthal’s model is not the only explanation on offer. Two genuine rivals locate the cause differently.

Self-verification theory (Swann & Read, 1981) argues people are not simply shaped by however others treat them. They have their own role to play.

Instead, they actively seek out and interpret feedback that confirms their existing self-view. A stable self-concept feels more predictable than an unverified one. Applied here, someone who starts to see themselves as unpromising may keep gravitating toward cues that confirm it.

The second rival is not psychological at all. It is a statistical artifact: regression to the mean.

Where a test is imperfectly reliable, people who score at an extreme once tend to score less extremely next time, for purely statistical reasons (Nesselroade, Stigler, & Baltes, 1980). Critics used exactly this to attack the original Pygmalion study’s very young, unreliable IQ scores (Thorndike, 1968; Elashoff & Snow, 1971).

Taken together, the three accounts are complementary rather than competing. Rosenthal’s model explains how a golem effect spreads. Self-verification (self-verification theory) explains how it gets internalised. The regression critique is a reminder that not every apparent case is psychological at all.

Strengths and Limitations

Four points sum up the evidence.

Hard to Study: Deliberately inducing damaging low expectations is unethical, so evidence leans on natural bias and on experiments that prevent a golem rather than create one (Oz & Eden, 1994).
Real but Small: Expectations largely predict outcomes because they are accurate, so once accuracy is controlled for, the genuinely self-fulfilling slice is modest (Jussim & Harber, 2005).
Unequal Impact: The small average effect is still unjust: it concentrates on disadvantaged and stereotyped groups, and inaccurate early expectations can leave a measurable mark a decade later (Sorhagen, 2013).
Not Everyone Is Equally Movable: People with a strong internal locus of control can resist a low label, while the young and uncertain are more susceptible to one.

Positioned against neighbouring effects, the golem effect is the negative case of the self-fulfilling prophecy and the exact inverse of the Pygmalion effect. It differs from stereotype threat, where it is the person’s own awareness of a stereotype, not another’s expectation, that depresses performance.

Further Information

Cite this article

Pick a style, copy the reference, or download it to your reference manager.

Updated 22 Sep 2026

Perera, A. (2026). Golem Effect. Simply Psychology. https://www.simplypsychology.org/golem-effect.html

Download .ris Download .bib

References

Babad, E. Y., Inbar, J., & Rosenthal, R. (1982). Pygmalion, Galatea, and the Golem: Investigations of biased and unbiased teachers. Journal of Educational Psychology, 74(4), 459–474. https://doi.org/10.1037/0022-0663.74.4.459

Davidson, O. B., & Eden, D. (2000). Remedial self-fulfilling prophecy: Two field experiments to prevent Golem effects among disadvantaged women. Journal of Applied Psychology, 85(3), 386–398. https://doi.org/10.1037/0021-9010.85.3.386

Elashoff, J. D., & Snow, R. E. (1971). Pygmalion reconsidered: A case study in statistical inference. Charles A. Jones.

Harris, M. J., & Rosenthal, R. (1985). Mediation of interpersonal expectancy effects: 31 meta-analyses. Psychological Bulletin, 97(3), 363–386.

Jussim, L., & Harber, K. D. (2005). Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies. Personality and Social Psychology Review, 9(2), 131–155. https://doi.org/10.1207/s15327957pspr0902_3

Leung, A., & Sy, T. (2018). “I am as incompetent as the prototypical group member”: An investigation of naturally occurring Golem effects in work groups. Frontiers in Psychology, 9, 1581. https://doi.org/10.3389/fpsyg.2018.01581

Madon, S., Jussim, L., & Eccles, J. (1997). In search of the powerful self-fulfilling prophecy. Journal of Personality and Social Psychology, 72(4), 791–809. https://doi.org/10.1037/0022-3514.72.4.791

Mahoney, R. (2011). “Chris Bosh and the self-fulfilling prophecy”. The New York Times.

Nesselroade, J. R., Stigler, S. M., & Baltes, P. B. (1980). Regression toward the mean and the study of change. Psychological Bulletin, 88(3), 622–637. https://doi.org/10.1037/0033-2909.88.3.622

Oz, S., & Eden, D. (1994). Restraining the Golem: Boosting performance by changing the interpretation of low scores. Journal of Applied Psychology, 79(5), 744–754.

Raudenbush, S. W. (1984). Magnitude of teacher expectancy effects on pupil IQ as a function of the credibility of expectancy induction: A synthesis of findings from 18 experiments. Journal of Educational Psychology, 76(1), 85–97. https://doi.org/10.1037/0022-0663.76.1.85

Rosenthal, R. (1973). The Pygmalion effect lives. Psychology Today, 7(4), 56–63.

Rosenthal, R., & Jacobson, L. (1968). Pygmalion in the classroom: Teacher expectation and pupils’ intellectual development. Holt, Rinehart & Winston.

Rubie-Davies, C., Hattie, J., & Hamilton, R. (2006). Expecting the best for students: Teacher expectations and academic outcomes. British Journal of Educational Psychology, 76(3), 429–444. https://doi.org/10.1348/000709905X53589

Smith, A. E., Jussim, L., & Eccles, J. (1999). Do self-fulfilling prophecies accumulate, dissipate, or remain stable over time? Journal of Personality and Social Psychology, 77(3), 548–565. https://doi.org/10.1037/0022-3514.77.3.548

Sorhagen, N. S. (2013). Early teacher expectations disproportionately affect poor children’s high school performance. Journal of Educational Psychology, 105(2), 465–477.

Swann, W. B., Jr., & Read, S. J. (1981). Self-verification processes: How we sustain our self-conceptions. Journal of Experimental Social Psychology, 17(4), 351–372. https://doi.org/10.1016/0022-1031(81)90043-3

Thorndike, R. L. (1968). Review of Pygmalion in the classroom. American Educational Research Journal, 5(4), 708–711.