Bolt Series · Human Performance Calibration · Article 38
Yes.
A student can be right when a teacher is wrong.
A student can be right when a parent is wrong.
A student can even be right when several people disagree with them.
But there is a trap hidden inside that sentence.
Being the minority does not make you right.
Being brave enough to disagree does not make you right.
Feeling misunderstood does not make you right.
And being surrounded by authority does not make everyone else right either.
Agreement is evidence about what people believe. It is not automatically evidence about what the world will show.
Majorities are often useful — because independent errors can cancel
The “wisdom of crowds” is real under the right conditions.
When several people make reasonably independent estimates and their errors point in different directions, aggregating those estimates can produce a result more accurate than many individual judgements.
Research on collective judgement has repeatedly demonstrated this effect.
That means the fact that several informed observers agree should not be dismissed casually.
If three teachers independently notice the same reasoning weakness across different tasks, the agreement deserves weight.
But the word independently matters.
Ten people repeating one original judgement are not ten independent receipts
Imagine one teacher says:
“This student is weak at Mathematics.”
The comment enters a school record.
A second teacher reads the record before meeting the child.
A parent hears the same judgement.
The student begins repeating it.
Now four people agree.
But how many independent observations created the agreement?
Perhaps only one.
Research on judgement aggregation warns that correlated or interdependent opinions can create a misleading sense of consensus. People can become more confident when several opinions agree even when those opinions share the same source of error.
Count independent evidence, not merely the number of mouths repeating it.
Majority status can itself change confidence
Social psychology has found that majority positions tend to be expressed with greater confidence than minority positions.
Knowing that others agree can also increase the influence of an opinion.
That makes practical sense.
If everybody else says the answer is B and you say C, uncertainty rises.
But confidence generated by consensus is not identical to correctness.
Research on subjective confidence has shown that consistency and consensus can make an answer feel more reliable even under conditions where the favoured answer is wrong.
This matters enormously for a child challenging an adult judgement.
“Everyone disagrees with me” should lower certainty enough to recheck.
It should not force surrender before the evidence is examined.
Minority opinions can improve group thinking
Research on group decision-making shows that minority views can improve the quality of discussion and sometimes improve private judgement among majority members.
In one set of experiments, a minority perceived as expert caused majority members to process the problem more deeply and become more accurate in their private judgements, even when the public group decision itself did not improve.
Other work has found that serious discussion of minority opinions can be associated with better team decision quality.
The minority’s value therefore does not depend on being right every time.
A dissenting view can force the group to reveal assumptions it would otherwise leave untested.
What if the student really is right?
Suppose everybody believes the learner does not understand a topic.
The student says:
“I understand it. My problem is timing.”
Do not reward the student merely for asserting themselves.
Test the claim.
Give an unfamiliar problem without time pressure.
Ask for an explanation.
Then repeat under realistic timing.
If conceptual performance is repeatedly strong and only timed performance collapses, the student’s model was more accurate.
Now the adults should update.
That is not a victory over authority.
It is a successful calibration correction.
What if the student is wrong?
The same rule applies.
The learner says:
“Everybody underestimates me.”
A fair independent task is provided.
The predicted capability does not appear.
Then the learner has a responsibility to update too.
“Nobody believes in me” cannot become a shield against repeated evidence.
Otherwise dissent turns into the same self-sealing error as blind conformity.
The right to challenge the group comes with the obligation to let the world challenge you back.
Student self-assessment should not be judged only by agreement with the teacher
A critical review of student self-assessment makes an important methodological point.
Researchers often call student self-assessment “accurate” when it agrees with teacher assessment.
But teacher ratings are not perfect criteria either.
In many classroom settings, consistency between student and teacher judgements is a more defensible term than assuming the teacher’s rating is ground truth.
This does not weaken teachers.
Teacher judgement research shows that teachers often assess academic achievement substantially accurately.
It simply keeps the epistemic structure honest.
The learner is fallible.
The teacher is fallible.
The next performance can help distinguish them.
A strong learner needs independent judgement without reflexive contrarianism
Education should not produce children who automatically obey consensus.
Nor children who automatically distrust it.
A stronger learner can say:
“Several informed people disagree with me, so I should lower my confidence and re-examine my evidence. But I will not change the claim merely because it is unpopular if a fair test can resolve it.”
That is independence with calibration.
Why this matters for education
There will be moments in a child’s life when nearly everyone around them is wrong.
There will also be moments when nearly everyone around them is right and the child is wrong.
Education cannot tell them in advance which moment they are in.
It can teach a discipline for finding out.
Listen to consensus. Protect independent evidence. Distinguish repeated opinions from independent observations. Then let a fair receipt decide what should update.
The student does not win because they stood alone.
The student wins when the model that best survives contact with reality is allowed to replace the model that did not.
Evidence and further reading
- The Wisdom of Select Crowds
- Spurious Consensus and Opinion Revision
- Minority Expertise and Group Decision Accuracy
- Minority Opinion Expression and Team Decision Quality
- The Self-Consistency Model of Subjective Confidence
- Critical Review — Student Self-Assessment
- Meta-Analysis — Teachers’ Judgment Accuracy
