Experiment 11 · Famous experiments
The students who were labelled bloomers at random
A famous study said teachers' raised expectations made random pupils gain IQ points, but the effect turned out smaller and patchier than the headlines.
- Robert Rosenthal and Lenore Jacobson
- South San Francisco, California, United States
- 1968
- 3 min read
On this page
Before you readWhat would you do?
A colleague shares a slide saying teachers' expectations can raise a child's IQ by 20 points, citing a study from the 1960s. What is the fairest response?
Pick an answer now. The file tells you whether you were right.
01The file in one minute
1Random bloomers
Teachers were told some pupils were about to bloom. The pupils had been picked at random.
2Gains in young classes
The chosen children gained more IQ points on average, mostly in the first and second grades.
3Smaller than famous
Critics found test problems, and later reviews put typical expectation effects at small sizes.
4What to take from it
Labels can nudge how people are treated. Famous findings still need checking for size and repeats.
02Timeline
1964
Rosenthal and Jacobson test pupils and tell teachers which ones are about to bloom.
1968
Their book Pygmalion in the Classroom reports IQ gains for the chosen pupils.
1968 to 1970s
Reviewers question the test scores and later repeats often find little effect.
2005
Jussim and Harber review decades of studies and find real but usually small effects.
03The experiment
In 1964 the psychologist Robert Rosenthal and the school principal Lenore Jacobson gave the pupils at a California elementary school a test. Teachers were told it could spot children about to make a leap in ability. In fact the researchers picked about one in five children at random and told the teachers these were the ones to watch.
When the children were tested again, the chosen group had gained more IQ points on average than their classmates. The difference showed up mostly in the youngest classes, the first and second grades, and was small or absent in the older ones. The 1968 book Pygmalion in the Classroom made the result famous.
Critics pointed out problems soon after. Some of the test scores were implausible for young children, the test was not designed for their age, and the overall effect rested heavily on a small number of pupils. Later attempts to repeat it often found nothing or very little.
04What it showed
Reviews of hundreds of later studies, such as one by Lee Jussim and Kent Harber in 2005, concluded that teacher expectations do affect pupils, but the effect is usually small. It seems stronger for some groups, and when teachers know little about a pupil. Most of the time teacher expectations are fairly accurate, because they are based on what the teacher has seen.
Two lessons survive. Labels can shift how people treat someone, a little, and the shift can add up for students who are already behind. A famous finding also deserves the same checking as any other claim. This one became a staple of training courses long before the evidence settled.
Labels work on readers of AI output too. Tell yourself a tool is brilliant and you will read its answers more kindly.
The idea behind this caseLesson: Check what AI tells you →
05What it means when you use AI
- When you hear a famous study quoted, look for how big the effect was and whether it has been repeated.
- Be careful with labels such as "weak student" or "star hire". Judge new work on its own evidence.
- Check an AI tool's answers with the same care whether its brand impresses you or not.
- Give everyone the same clear, high but reachable standard, which helps without relying on a label.
07 — Sources
- Teacher expectations and self-fulfilling prophecies: Knowns and unknowns, resolved and unresolved controversies · Personality and Social Psychology Review · 2005
- Pygmalion effect · Wikipedia · 2026