A student can know science without believing science belongs to them.
They may score well, follow experiments accurately and understand the chapter. Yet when someone asks, “Are you a science person?”, they say no.
Another student may love documentaries, ask endless questions about animals, fix electronics at home and follow space missions online — but perform modestly in formal science lessons and conclude that these interests do not count as “real science”.
A third learner may think science is fascinating but cannot imagine anyone from their family doing a scientific job. A fourth may see scientists as exceptionally gifted people who knew from childhood what career they wanted. A fifth may enjoy science until repeated difficulty turns “this topic is hard” into “I am not the kind of person who can do this”.
These are not only knowledge problems.
They are identity problems.
Science identity concerns the extent to which a learner sees science as connected to who they are, what they can do, what people like them can legitimately participate in, and what futures feel imaginable.
Current OECD work makes this especially timely. The PISA 2025 Assessment and Analytical Framework, published in May 2026, expands science assessment beyond knowledge and competency to include constructs around science identity. These include science capital, science self-concept, science self-efficacy, enjoyment, instrumental motivation, epistemic beliefs, agency and engagement. The framework reflects a broad international concern: knowing science matters, but so does whether learners recognise themselves as capable participants in science-related activity and decision-making.
That does not mean schools should try to persuade every child to become a scientist.
It means science education has a larger job than producing correct answers.
A learner should leave school able to use scientific knowledge in personal and civic life and able to choose further science without having been quietly excluded by identity assumptions.
The key distinction is therefore:
Science achievement asks, “What can I currently do?” Science identity also asks, “What does doing science say about who I am, and is this a world in which I can belong?”
The 50-second answer
Science identity develops through repeated interactions among at least five kinds of evidence.
- Competence evidence: Can I understand and solve scientific problems?
- Recognition: Do teachers, peers, family and institutions treat me as someone who can contribute to science?
- Interest and value: Does science connect to things I care about?
- Belonging and representation: Can I imagine people like me participating in this field without pretending to be someone else?
- Opportunity and capital: Do I know how science pathways work, have access to experiences and resources, and know people or institutions that make those pathways visible?
These signals accumulate.
A learner solves a difficult experiment independently. A teacher notices the quality of the reasoning rather than merely praising “talent”. A parent discusses science in ordinary life. A museum visit makes a scientific field concrete. A scientist’s biography complicates the stereotype of who scientists are. A failed test is interpreted as information about a method rather than a verdict on identity. The learner gradually updates:
“I can use this kind of thinking.”
The opposite can also happen.
Repeated difficulty, low expectations, narrow representation, inaccessible opportunities, peer stereotypes or hidden pathway information can accumulate into:
“Science is for other people.”
Identity is not fixed, and no single classroom activity creates it. It is an inference built across experiences.
First distinction: science identity is not the same as science ability
This distinction protects the whole article from confusion.
A learner can have high science attainment and weak science identity.
A learner can have strong science identity and uneven attainment.
A learner can be interested in science but doubt their competence.
A learner can feel competent but see little value in pursuing science further.
A learner can enjoy school science while believing scientific careers are socially inaccessible.
These combinations matter because educational decisions are not made from achievement alone.
Course selection, persistence through difficulty, career exploration and willingness to use science publicly can depend partly on how capability is interpreted.
Imagine two students who both score 78%.
Student A says, “Good. I know what I need to improve.”
Student B says, “See? Real science students would get 95. I should stop.”
The same evidence has entered two different identities.
This is why schools should avoid treating identity work as a substitute for strong science instruction. Learners need real competence. At the same time, real competence does not automatically become a durable identity if the learner explains every success away as luck and every difficulty as proof of unsuitability.
Stage 1: early experience teaches what counts as “science”
Children meet science before they meet the school subject.
They watch insects.
Mix ingredients.
Ask why the Moon changes shape.
Take apart toys.
Notice mould.
Ask why metal feels colder than wood.
Compare which paper aeroplane flies farther.
None of this automatically becomes “science identity”. The interpretation matters.
If science is presented mainly as textbook vocabulary and correct laboratory reports, students may fail to connect everyday curiosity with scientific practice.
A child who asks excellent causal questions may not recognise that questioning as scientific capability.
A useful classroom move is to make the disciplinary link visible:
“The question you just asked — what would happen if only one condition changed — is exactly the kind of question experimental science uses.”
Notice the precision.
The teacher is not saying, “You are a scientist!” after one question.
The teacher is naming a scientific practice.
Identity grows from credible evidence.
Stage 2: competence becomes identity only when the learner notices it
Students can improve without updating their self-concept.
A learner who begins the year unable to explain circuits may later produce a clear model of current, potential difference and resistance.
If the teacher says only “8/10,” the result records performance.
If the teacher says:
“At the start of the term you described current as something that gets used up. Here you correctly track charge through the whole circuit and explain why the current is the same in series. That conceptual model has changed.”
the learner receives identity-relevant evidence.
This connects with progress visibility.
People build identities partly from stories about change:
“I used to be bad at this, but now…”
Without comparison across time, improvement can disappear into the present.
Science identity therefore benefits from records that show developing capability rather than only rank.
This does not require inflated praise. In fact, inflated praise weakens credibility.
Name the actual scientific capability:
- you controlled variables;
- you revised the model after contradictory evidence;
- you distinguished observation from inference;
- you interpreted uncertainty;
- you connected mechanism to data;
- you asked a discriminating question.
Those are stronger identity signals than “you are naturally good at science”.
Stage 3: recognition tells learners whether competence is socially visible
Identity is partly relational.
It matters whether people around the learner recognise their scientific participation.
A student may know they understand, but repeated social messages can still shape whether they treat that understanding as legitimate.
Who gets asked the difficult science questions?
Who is chosen to set up equipment?
Whose explanation is repeated by the teacher?
Whose mistakes are interpreted as temporary?
Whose mistakes are treated as confirmation?
Who gets told to consider advanced science?
Who gets the benefit of the doubt?
These small recognition patterns can accumulate.
Recognition does not mean distributing empty compliments equally.
It means ensuring that actual scientific contributions are noticed across a wider range of learners.
A quiet student may produce the best graph interpretation in the room but never speak publicly. A multilingual learner may have excellent scientific reasoning but slower English production. A student with weaker calculation fluency may ask unusually powerful conceptual questions.
Teachers who look for multiple forms of scientific competence can make recognition more accurate.
Stage 4: stereotypes narrow the imagined membership of science
Ask young people to picture a scientist.
Many will still draw familiar stereotypes: lab coat, test tubes, perhaps male, perhaps isolated, perhaps unusually brilliant.
These images matter when learners compare themselves with the category.
If science is imagined as a world for solitary geniuses, a collaborative, socially oriented learner may not see fit.
If all examples of scientific expertise come from a narrow demographic range, students may infer that identity rather than historical selection created the pattern.
Representation can help, but representation must be more than poster replacement.
Show varied scientists doing varied work:
- field ecology;
- public health;
- engineering;
- climate modelling;
- veterinary science;
- data science;
- conservation;
- laboratory research;
- science communication;
- technical trades that use scientific knowledge.
Also show scientific work as social and imperfect.
Scientists collaborate.
They disagree.
They use instruments.
They make mistakes.
They revise models.
They write.
They negotiate uncertainty.
A richer occupational picture gives more learners a plausible point of entry.
Stage 5: science capital makes pathways visible
The OECD PISA 2025 framework includes science capital within the wider science-identity landscape.
Science capital is a useful way to think about the resources around a learner that make science more familiar and navigable.
These resources can include:
- science-related knowledge;
- family attitudes toward science;
- knowing people who use science;
- visits to museums, centres or natural environments;
- access to science media;
- awareness of education pathways;
- understanding what scientific careers actually involve;
- confidence interacting with institutions.
A learner may have high academic ability and low pathway visibility.
“I like chemistry” is not the same as knowing:
What subjects do I need next year?
What degrees exist?
What does a materials scientist do?
What is laboratory work actually like?
What careers use chemistry without being called “chemist”?
Families differ in how much of this information is already available at home.
Schools can reduce that inequality by making pathways public rather than assuming students will discover them through family networks.
Stage 6: self-efficacy is task-specific evidence
Science self-efficacy asks something like:
“Do I believe I can successfully perform this scientific task?”
That is more specific than identity.
A learner may have strong science identity overall but low self-efficacy for physics calculations.
Another may not think of themselves as a “science person” yet feel highly capable in biology experiments.
Self-efficacy grows through:
- successful performance;
- accurate feedback;
- observation of similar others succeeding;
- interpretation of difficulty and anxiety;
- manageable challenge.
This matters for identity because repeated task-level judgments can accumulate upward.
“I can interpret this graph.”
“I can design a fair test.”
“I can explain the mechanism.”
Eventually:
“I can do science.”
The direction can also reverse.
This is why broad encouragement is weaker than specific mastery evidence.
Stage 7: enjoyment and identity are related but not identical
A learner can enjoy science and still not imagine pursuing it.
A learner can identify strongly with science and still find some required tasks tedious.
A researcher can love the field and dislike data cleaning.
A doctor can value science and dislike certain laboratory exercises.
Education should therefore avoid making enjoyment the sole indicator of identity.
“Did you have fun?” is one question.
“Did you feel capable of making sense of the phenomenon?” is another.
“Could you imagine learning more about this?” is another.
“Did this feel relevant to something you care about?” is another.
Identity is a broader pattern of relationship to the field.
Stage 8: instrumental value can open a door without sustaining the whole journey
Some learners value science because of future usefulness.
Medicine.
Engineering.
Environmental work.
Technology.
Employment.
University pathways.
This instrumental motivation can be powerful.
It can also narrow science into credential acquisition.
Strong science education gives learners multiple reasons to engage:
- understanding the world;
- solving practical problems;
- participating in civic decisions;
- pursuing careers;
- experiencing intellectual beauty;
- making and testing explanations.
A learner does not need all six.
But if the only value is “this gets you into a good job,” identity may collapse when the career goal changes.
Stage 9: difficulty is where identity stories become visible
Easy success tells us little about how identity handles stress.
The important moment is:
What does the learner conclude when science becomes difficult?
Possible interpretations:
“This concept is difficult.”
“My current strategy is weak.”
“I am missing prerequisite mathematics.”
“I need more practice.”
or:
“I am not a science person.”
The last conclusion is global.
It converts one state into identity.
Teachers can intervene by localising the difficulty.
“Your model is correct. The calculation is the problem.”
“You understand the mechanism but the vocabulary is slowing you down.”
“Your data interpretation is strong; the conclusion overclaims.”
This turns identity threat into a specific learning target.
Stage 10: identity and performance can form feedback loops
Identity affects action.
Action affects performance.
Performance feeds identity.
A learner who believes science is “for me” may:
- choose harder courses;
- persist longer;
- seek help;
- join clubs;
- consume more science media;
- meet more science-interested peers.
Those experiences can build competence and capital.
The opposite loop can also occur.
Weak identity → fewer science choices → fewer experiences → less confidence → even weaker identity.
This does not mean identity causes every pathway difference.
Access, finances, school offerings, grades, family expectations, discrimination and many other factors matter.
Identity is one mechanism inside a larger opportunity system.
Science identity is not the same as “growth mindset”
Growth mindset concerns beliefs about whether abilities can develop.
Science identity asks whether the learner sees science as connected to self, recognition, belonging and possible participation.
A student can believe intelligence grows and still think science is “not for people like me”.
Another can identify strongly with science yet hold a fixed view of talent.
The constructs can interact.
They should not be collapsed.
Science identity is not simply belonging
Belonging asks:
“Do I feel accepted and legitimate in this environment?”
Science identity asks something broader:
“Is science part of who I am or could become?”
A learner may feel socially comfortable in science class but have no interest in science identity.
Another may identify strongly as a science learner while feeling isolated in a particular classroom.
eduKateSG already owns broader belonging mechanisms in its learning estate. This article remains bounded to identity in relation to science as a field and practice.
Science identity is not a career-recruitment campaign
The purpose of school science is not to maximise the number of scientists.
Citizens need science to:
- interpret health claims;
- reason about climate and energy;
- understand technology;
- evaluate evidence;
- make personal decisions;
- participate in democratic choices.
A healthy science identity can therefore include:
“I am not pursuing a science career, but I am capable of using scientific reasoning.”
That is a strong educational outcome.
Science identity is not solved by inspirational biographies alone
Role models can matter.
But one famous-scientist poster cannot overcome weak instruction, inaccessible pathways or repeated non-recognition.
Representation works best when embedded in ordinary curriculum.
Instead of one “women in science” week, include women’s scientific work where the relevant science is taught.
Instead of presenting only Nobel Prize winners, include technicians, engineers, field scientists, clinicians, data analysts, conservation workers and local science users.
Instead of portraying science as heroic individual discovery, show collaborative teams and routine scientific labour.
The aim is not inspiration theatre.
It is a more accurate map of the field.
Common failure mode: “You are a scientist” after any activity
A child mixes vinegar and bicarbonate.
The adult says, “You’re a scientist!”
Warm.
Potentially empty.
Better:
“You changed one variable and compared the result. That control is part of experimental reasoning.”
Now identity recognition is attached to a disciplinary practice.
The learner knows what behaviour made the label plausible.
Common failure mode: only recognising fast correctness
Science rewards more than rapid answers.
A slow learner may design excellent tests.
A student weak in recall may reason unusually well from data.
A quiet learner may notice anomalies others miss.
A learner with poor handwriting may build excellent models.
If teachers recognise only quick public correctness, science identity becomes attached to speed and confidence.
Broaden what counts as visible competence.
Common failure mode: advanced pathways that remain socially hidden
A school offers excellent science opportunities.
Only students who already know they exist take them.
Olympiads.
Research programmes.
Laboratory attachments.
Special courses.
Scholarships.
If access relies on family knowledge or informal teacher nomination, identity and opportunity can reinforce existing differences.
Publish the pathway.
Explain eligibility.
Invite widely.
Then select transparently where selection is necessary.
Common failure mode: assuming low interest is permanent
A fourteen-year-old says science is boring.
This is current data, not destiny.
Interest changes with topic, teacher, competence, relevance and experience.
Do not force passion.
But do not treat one statement as a lifelong identity contract.
Common failure mode: using identity language to mask weak attainment
“She feels like a scientist.”
Wonderful.
Can she interpret the data?
Explain the mechanism?
Use evidence?
Apply concepts?
Identity without competence becomes fragile.
The purpose of identity-supportive teaching is not to replace achievement with self-perception. It is to make accurate capability more likely to become durable participation.
Common failure mode: treating science as culturally neutral
Scientific methods aim at evidence that can be tested publicly.
Science education still takes place inside culture.
Which problems are studied?
Whose contributions are remembered?
How examples are framed?
Which language forms are treated as intelligent?
Who is expected to succeed?
Ignoring these questions does not remove culture.
It makes the dominant pattern harder to see.
If you are a learner
Separate science identity into smaller questions.
Do not ask only:
“Am I a science person?”
Ask:
- Which science tasks can I already do?
- Which ones am I learning?
- Which topics genuinely interest me?
- Which difficulties are scientific and which are mathematical or linguistic?
- What kinds of science work exist beyond the classroom?
- What pathways are available next?
Keep evidence.
Save one good explanation from the beginning and end of the term.
Compare them.
Track not only marks but capabilities.
If you think science is “not for people like me,” interrogate the sentence.
Who taught you that?
What evidence supports it?
Have you seen the actual range of people and jobs in science?
You do not owe science a career.
You do deserve an accurate choice.
If you are a parent
Parents influence science identity even without science expertise.
Avoid turning difficulty into identity:
“Science was never our family’s thing.”
“I was terrible at physics too.”
Those comments are usually meant to comfort.
They can also make struggle hereditary.
Better:
“That looks hard. Which part do you understand so far?”
Discuss science in ordinary life.
Food.
Health.
Weather.
Technology.
Animals.
Energy.
Materials.
Do not require the child to perform enthusiasm.
Make science familiar enough that interest can grow without becoming a family identity test.
When career decisions approach, investigate pathways together rather than relying on stereotypes.
If you are a teacher
Audit recognition.
Across two weeks, note whose scientific thinking you publicly acknowledge.
What did you recognise?
Speed?
Recall?
Questioning?
Evidence use?
Model revision?
Experimental design?
Try to make a broader range of authentic scientific practice visible.
Audit representation.
Who appears in your examples of science work?
Audit pathway information.
Do all students know how to access advanced opportunities?
Audit difficulty language.
When a learner fails, do comments localise the problem or globalise identity?
And protect the curriculum.
Identity grows most credibly when learners become genuinely more capable.
A classroom mini-sequence: identity from evidence, not slogan
At the end of a practical investigation, ask students to complete three lines.
One scientific decision I made today:
One piece of evidence that changed my explanation:
One scientific skill I can now use more independently:
Collect answers occasionally.
Use them in conferences.
Over time, students build a record of disciplinary participation.
This is different from saying “everyone is a scientist”.
It asks learners to identify the evidence on which a science identity could honestly be built.
A career-information mini-sequence
Take one curriculum topic.
Suppose the class is studying materials.
Show five uses:
- materials engineer;
- conservation scientist;
- medical-device designer;
- forensic analyst;
- semiconductor manufacturing specialist.
Do not turn the lesson into a career fair.
Ask:
Which scientific knowledge from this topic appears in each role?
What further education is typically required?
Which role surprises you?
This connects curriculum to possibility without demanding aspiration.
A Clementi-floor test: who gets to feel “scientific” by 10:30?
Imagine a normal science lesson.
The teacher asks a question.
The fastest student answers.
The teacher says, “Exactly.”
Then the class copies notes.
Who received science identity evidence?
One fast student.
Now redesign.
Everyone predicts.
Pairs compare.
The teacher asks one learner for evidence, another for a competing explanation, another for a flaw in the experiment, another for what should be measured next.
Different scientific practices become visible.
The class still learns the same science.
More learners get legitimate opportunities to see what participating in science actually means.
Frequently asked questions
What is science identity?
It refers to how learners see themselves in relation to science — including perceived competence, recognition, belonging, interest, value, opportunity and possible future participation.
Is science identity just confidence?
No. Confidence is one component. Identity also includes social recognition, belonging, value, science capital and how the learner connects science to self and future.
Can a student have strong science identity and low marks?
Yes. Identity and achievement overlap but are not identical. The learner may be highly interested and engaged but still need substantial knowledge or skill development.
Can a high-achieving student have weak science identity?
Yes. They may attribute success to luck, feel they do not fit the stereotype, or view science only as a school requirement.
Does every learner need a strong science identity?
Every learner needs enough scientific agency and confidence to use science in life and make informed pathway choices. Not everyone needs to make science central to personal identity or career.
What is science capital?
A way of describing science-related resources around a learner: knowledge, experiences, networks, attitudes, pathway information and familiarity with science-related institutions and careers.
Do role models work?
They can broaden representation and possibility, but role models alone are unlikely to overcome poor instruction, inaccessible pathways or repeated low expectations.
Should teachers tell children they are scientists?
It is usually stronger to recognise specific scientific practices: “You revised your explanation when the evidence changed.” Identity language then rests on real disciplinary behaviour.
How can parents help if they are not good at science?
Support curiosity, discuss everyday science, avoid hereditary-deficit language, investigate pathways together and ask the child to explain what they are learning.
What is the simplest classroom move?
Broaden what counts as visible scientific competence and name the practice precisely when you see it.
Evidence and caveats
The OECD PISA 2025 Assessment and Analytical Framework, published on 29 May 2026, explicitly includes science identity within the broader architecture of science education. The framework discusses constructs including science capital, self-concept, self-efficacy, enjoyment, instrumental motivation, epistemic beliefs, agency and engagement.
This makes science identity highly relevant to current international education research, but several caveats matter.
Identity is measured indirectly. Surveys rely on self-report and are sensitive to context and interpretation.
Identity and achievement influence one another. Strong performance may strengthen identity; stronger identity may increase participation; both may be shaped by family, school and opportunity.
Group patterns do not determine individuals. Research on demographic differences can identify inequities without permitting assumptions about one learner based on category.
Identity interventions vary. Role models, mentoring, inquiry learning, curriculum representation, pathway information and teacher recognition operate through different mechanisms.
Science identity is not a replacement for science knowledge. A learner needs real opportunity to acquire disciplinary competence.
The responsible conclusion is:
science learning is partly a problem of knowledge and skill, and partly a problem of whether learners experience enough competence, recognition, value and access to treat scientific participation as a legitimate option for themselves.
The deeper idea: education helps learners decide what worlds are open to them
Every subject teaches more than content.
It teaches an implicit answer to:
Who is this for?
A mathematics classroom teaches whether mathematical thinking belongs only to the fast.
An English classroom teaches whether literary interpretation belongs only to the articulate.
A science classroom teaches whether science belongs only to people who already look confident, know the right vocabulary and have families familiar with scientific pathways.
Or it can teach something more accurate.
Science is a disciplined way of building and correcting explanations about the natural world.
Those practices can be learned.
People enter through different routes.
Competence grows unevenly.
Careers are diverse.
Not everyone needs to choose one.
But nobody should leave because a stereotype made the decision before the evidence did.
Sources and further reading
- OECD. PISA 2025 Assessment and Analytical Framework. Published 29 May 2026. https://www.oecd.org/en/publications/pisa-2025-assessment-and-analytical-framework_01c7e80b-en.html
- OECD. Science Framework for PISA 2025. https://pisa-framework.oecd.org/science-2025/
- OECD. PISA 2025 Results (Volume I): Learning and Equity in a Changing World. Published 8 September 2026. https://www.oecd.org/en/publications/pisa-2025-results-volume-i_73451bc5-en.html
- Education Endowment Foundation. Science review. https://educationendowmentfoundation.org.uk/education-evidence/evidence-reviews/science
- Institute of Education Sciences. Science-related research and resources. https://ies.ed.gov/ncer