All Skills
Thinking Skills

Scientific Thinking

How scientists think — forming hypotheses, testing ideas fairly, interpreting evidence honestly, and knowing what experiments can and cannot prove. Scientific thinking is not only for scientists. It is one of the most powerful and most reliable ways to find out what is true.

Key Ideas at This Level
1 We can find out if something is true by testing it.
2 A good test is fair — we change only one thing at a time.
3 What we observe with our senses is important evidence.
4 It is okay to be wrong — what matters is that we keep asking questions.
5 Science is a way of finding out things, not a collection of facts to memorise.
Teacher Background

Scientific thinking at Early Years level is about building the foundational disposition of inquiry — noticing, questioning, testing, and observing — before any scientific content is introduced. Young children are natural scientists: they test cause and effect constantly through play, they observe carefully when something interests them, and they form and revise theories about how the world works. The teacher's role is to name this as scientific thinking and to channel it towards more systematic observation and testing. The most important concept to establish at this level is fair testing: when we want to find out which of two things works better, we must change only one thing at a time — otherwise we cannot know which change made the difference. This principle is both the foundation of experimental design and one of the most important forms of causal reasoning. In communities with rich traditional knowledge of plants, animals, soil, and weather, children already possess valuable scientific observations. This knowledge should be honoured as genuine empirical knowledge — not dismissed as superstition — while also being subjected to the same questioning habits that scientific thinking develops. All activities use simple language at B1 CEFR level.

Skill-Building Activities
Activity 1 — I wonder and I notice: the beginning of scientific thinking
PurposeChildren practise the first two habits of scientific thinking — careful observation and genuine questioning — understanding that noticing leads to wondering, and wondering leads to finding out.
How to run itTake children outside to any natural space. Give them one instruction: spend five minutes looking at one thing very carefully. It can be anything — an ant, a crack in the ground, a plant, a stone, a puddle. After five minutes, ask each child: what did you notice? Collect all observations. Now ask: did anything you noticed make you wonder something? What questions did your observation create? Help children formulate genuine questions — ones they actually want answered and do not already know the answer to. Why does this ant carry something three times its size? What made this crack in the ground? Why is this plant growing here but not one metre away? Now sort the questions by type: questions we could answer by looking more carefully right now, questions we could answer by doing a test, questions we would need to ask an expert, questions that might be very hard to answer at all. Ask: which kind of question is easiest to investigate? Which is most interesting? Introduce the idea: science starts with noticing and wondering. The observation is the beginning. The question is the direction. The test is how we find out. You were doing all three of these things just now.
💡 Low-resource tipNo materials needed. Any outdoor or indoor environment provides enough for careful observation. The teacher should model genuine curiosity — pointing out things they have noticed and saying I wonder why — rather than only directing children's attention.
Activity 2 — Fair testing: changing only one thing
PurposeChildren understand the principle of fair testing — that to find out which of two things makes a difference, we must change only one thing at a time.
How to run itPresent a simple question: does a plant grow better with more water or less water? Ask children: how would we find out? Accept all suggestions. Now introduce the fair test problem through a thought experiment. Suppose we give Plant A a lot of water and leave it in the sun. We give Plant B very little water and leave it in the shade. After two weeks, Plant A grows much better. What does this prove? (Nothing certain — Plant A had more water AND more sun. We do not know which made the difference.) What would we need to do to find out? (Keep everything the same except the one thing we are testing — in this case, water. Same soil, same container, same sunlight, same temperature. Only change the amount of water.) Introduce the idea: a fair test is one where we change only one thing. We call this the variable we are testing. Everything else stays the same. Now try a simple fair test together with whatever is available — comparing two objects that fall at different rates, testing whether soil from two different places holds water differently, seeing whether a paper boat floats longer in fresh or salty water. Before the test, ask: what are we changing? What are we keeping the same? How will we observe the result? After the test: what did we find? Are we sure? What would make us more sure?
💡 Low-resource tipAny simple local materials work. The principle of fair testing is more important than the specific test. Even a thought experiment — imagining a test without doing it — establishes the concept. The question what are we keeping the same? is the most important one to ask repeatedly.
Activity 3 — What the evidence says: being honest about what we found
PurposeChildren practise interpreting evidence honestly — including saying when the evidence does not support their prediction, and understanding that being wrong is useful, not shameful.
How to run itSet up a simple prediction-and-test activity. Before the test, ask children to predict the result. Write all predictions on the board — or ask children to remember theirs. Do the test. Now look at the results honestly. Ask: what did we find? Does the result match your prediction? For children whose predictions were not supported, ask: was it bad that your prediction was wrong? Introduce the idea: in science, a prediction that turns out to be wrong is not a failure. It is information. It tells you something important about how the world works that you did not know before. Scientists are wrong very often. The most important thing is not to be right the first time but to look at the evidence honestly, whatever it shows. Now ask: what if I told you that your prediction was right, even though the evidence shows it was wrong? Would that be good? Introduce the idea: in science, we have to follow the evidence even when it does not show what we hoped. If we only accept results that confirm our predictions, we learn nothing. The hardest and most important part of scientific thinking is being honest when the evidence contradicts what you believed.
💡 Low-resource tipWorks with any simple test that has an observable result. The most powerful version uses a test where a majority of children predict the wrong outcome — the collective experience of being wrong together normalises it and reduces the shame. The teacher should model honest response to unexpected results: I predicted this and I was wrong — that is interesting. Why do you think the result was different?
Reflection Questions
  • Q1What is something you have noticed about the world around you that you do not fully understand? How could you find out more?
  • Q2Have you ever tested something to find out if it was true? What did you do?
  • Q3Have you ever been sure about something that turned out to be wrong? How did you find out you were wrong?
  • Q4Is there something that adults in your community believe that you have wondered whether it is really true?
  • Q5What is the difference between guessing and finding out through a test?
Practice Tasks
Drawing task
Draw something in nature that you have observed carefully. Write or say: I noticed __________, and this made me wonder __________. To find out, I could __________.
Skills: Practising the observe-wonder-test sequence that is the beginning of scientific thinking
Model Answer

A careful drawing of a specific observed natural phenomenon — an insect, a plant, a weather pattern, a soil type. The completion names a specific observation, a genuine question it raises, and a realistic investigation method — not I could look it up but I could watch more carefully, I could test, I could ask someone who knows.

Marking Notes

The to find out I could is the most important completion — it asks children to think about investigation methods, not just about questions. Celebrate any genuine investigative method, however simple. The habit of connecting observation to question to possible test is what matters.

Fair test design
Think of a question about the natural world that you could test. Write or say: my question is __________, in my test I will change __________, and I will keep everything else the same, especially __________. I will know the answer when I observe __________.
Skills: Designing a fair test — practising the identification of the variable, the controls, and the observable outcome
Model Answer

My question is whether plants grow faster in soil with animal droppings mixed in or in soil without them. In my test I will change whether droppings are in the soil or not, and I will keep everything else the same, especially the type of seed, the amount of water, and the amount of sunlight. I will know the answer when I observe which plant is taller after three weeks and measure the difference.

Marking Notes

Award marks for: a testable question (not too broad, not already answered); a clearly identified single variable being changed; specific controls that would actually matter to this test; and an observable outcome that is specific enough to distinguish the two conditions. The most common error at this level is changing more than one thing — celebrate test designs that correctly identify this problem.

Common Mistakes
Common misconception

Science is about finding right answers to put in books.

What to teach instead

Science is a process of inquiry — asking questions, designing tests, observing results, and revising understanding. The facts in science books are the current best answers to questions that scientists have investigated. But science is always open to revision when new evidence appears. Many things that were in science books one hundred years ago are now known to be wrong or incomplete. The most important thing science produces is not a collection of facts but a way of thinking that allows us to keep improving our understanding of how the world works.

Common misconception

Traditional knowledge and scientific knowledge are completely different — one is right and the other is wrong.

What to teach instead

Traditional ecological knowledge, medical knowledge, and agricultural knowledge represent genuine empirical observations accumulated over generations — people testing what works through experience. Much traditional knowledge has been scientifically validated. Much scientific knowledge was derived from or inspired by traditional knowledge. The relationship between the two is complementary, not competitive. Both involve careful observation and experience-based reasoning. The difference is primarily in the methods of validation — scientific methods use controlled experiments and systematic comparison; traditional methods use long experience and community validation. Both have genuine strengths and genuine limitations.

Common misconception

Scientists always know what they are looking for before they do an experiment.

What to teach instead

Many of the most important scientific discoveries happened when scientists were looking for something else entirely, or when they noticed something unexpected and followed their curiosity. Penicillin was discovered when Alexander Fleming noticed that mould was killing bacteria in a petri dish he had left out. X-rays were discovered accidentally. The structure of DNA was discovered partly through unexpected patterns in X-ray images. Scientific curiosity — the willingness to notice and follow the unexpected — is as important as systematic testing.

Key Ideas at This Level
1 The scientific method — from question to conclusion
2 Hypotheses — what a scientific prediction looks like
3 Variables and controls — what makes a test fair
4 Evidence and proof — what experiments can and cannot establish
5 Scientific consensus — how individual studies become accepted knowledge
6 Pseudoscience — how to tell the difference between science and imitation science
Teacher Background

Scientific thinking at primary level introduces students to the formal structure of scientific inquiry — hypothesis, experiment, evidence, and conclusion — and to the critically important skills of evaluating scientific claims. The scientific method: while real science is less linear than textbook diagrams suggest, the basic structure of scientific inquiry — observe, question, hypothesise, test, analyse, conclude, communicate, replicate — captures the key elements. The most important conceptual distinctions are between a hypothesis (a specific, testable prediction) and a theory (a well-supported, comprehensive explanation that has survived many tests), and between evidence (what the data shows) and conclusion (what this means for our understanding).

Hypotheses and falsifiability

Karl Popper's insight that scientific claims must be falsifiable — there must be some possible observation that could show them to be wrong — is one of the most important distinguishing features of science. A claim that is compatible with any possible observation is not scientific: if no possible evidence could change your mind, you are not thinking scientifically. This criterion also helps identify pseudoscience: pseudoscientific claims are typically designed so that no evidence can falsify them.

Scientific consensus

Individual studies can be wrong, biased, or misleading. Scientific knowledge is built through many studies on the same question, independent replication, peer review, and the accumulated weight of evidence. Understanding the difference between a single study (which might be wrong) and a scientific consensus (which represents the best current understanding from many studies) is essential for evaluating health, environmental, and policy claims.

Pseudoscience

Claims that look like science but do not follow scientific principles — homeopathy, astrology, many alternative medicine claims — are a genuine problem for public health decision-making. The hallmarks of pseudoscience include: not being falsifiable, relying on testimonials rather than controlled studies, explaining failures away rather than treating them as evidence, having no plausible mechanism, not being replicated in independent studies, and appealing to conspiracy theories to explain the absence of scientific support.

Key Vocabulary
Hypothesis
A specific, testable prediction about what will happen in an experiment — based on existing knowledge and stated in a way that could be proved wrong by evidence.
Variable
Something that can change in an experiment. The independent variable is what the experimenter deliberately changes. The dependent variable is what is measured to see if it changes as a result.
Control
A condition that stays the same throughout an experiment — allowing the effect of the independent variable to be isolated. Controls make a test fair.
Evidence
Data or observations collected during an experiment. Evidence is what the experiment actually shows — it does not always support the hypothesis.
Replication
Repeating an experiment — by the same researcher or by different ones — to check whether the result is consistent. Replication is essential for scientific knowledge: a result that cannot be replicated is not reliable.
Scientific theory
A well-tested, comprehensive explanation supported by a large body of evidence from many independent sources. In science, theory does not mean guess — it means the best supported explanation available.
Falsifiability
The property of a scientific claim that it could in principle be shown to be wrong by some possible evidence. A claim that cannot be falsified — that is compatible with any possible observation — is not a scientific claim.
Pseudoscience
A system of claims that looks like science but does not follow scientific principles — not falsifiable, not replicated, relying on testimonials rather than controlled evidence, and not accepted by the relevant scientific community.
Skill-Building Activities
Activity 1 — From question to conclusion: designing a complete investigation
PurposeStudents practise the complete scientific inquiry process — from question through hypothesis, fair test, data collection, and honest conclusion.
How to run itWalk students through the six stages of scientific inquiry for a simple, testable question relevant to local life. Use a genuinely useful local question — for example: does adding wood ash to soil increase the growth rate of a common local plant? Stage 1 — Question: what exactly do we want to know? Write it precisely. Stage 2 — Hypothesis: what do we predict will happen, and why? A hypothesis must be specific (ash will increase growth by a measurable amount) and falsifiable (we would change our view if growth does not differ). Stage 3 — Fair test design: what is the independent variable (ash/no ash), the dependent variable (plant height after three weeks), and the controls (same seed type, same soil volume, same water, same sunlight)? Stage 4 — Data collection: how will we measure and record results consistently? Stage 5 — Analysis: what did the data show? Were there differences? Were they large enough to be meaningful? Stage 6 — Conclusion: does the evidence support the hypothesis? Be honest — if it does not, say so. What would be the next question to investigate? Now run a simpler version of this process in one lesson with any locally available materials. Even a short investigation — does water with salt dissolve soap faster than fresh water? does a larger leaf catch more water than a smaller one? — produces the experience of the full process. Emphasise stage 6: the conclusion must follow the evidence, not the hypothesis.
💡 Low-resource tipWorks with any locally available natural materials. Seeds, soil, leaves, stones, and water are sufficient for many investigations. The process is more important than any specific investigation. Even designing an investigation without running it develops the thinking skills.
Activity 2 — Science versus pseudoscience: how to tell the difference
PurposeStudents learn to distinguish scientific claims from pseudoscientific ones — building the practical health literacy that protects against ineffective and harmful treatments.
How to run itIntroduce the problem: some health products, treatments, and claims look like science — they use scientific-sounding language, they claim to be proven — but they do not actually follow scientific principles. Learning to tell the difference can protect your health. Present a checklist of five questions for evaluating any health or scientific claim. Is it falsifiable? Is there some observation that could prove it wrong — and have people tried to find that observation? Has it been tested in controlled studies? Not just used by people, but tested in a fair experiment where some people received it and others did not? Have the results been replicated? Has more than one independent research group found the same result? Has it been reviewed by relevant scientific experts? Not just any expert — someone with knowledge of this specific field? Is there a plausible mechanism? Does it make sense that this would work given what we know about biology or physics? Now apply the checklist to three examples: a well-established medical treatment (a vaccine or an oral rehydration therapy), a common local traditional remedy, and a commercial product making health claims. Apply all five questions honestly to each. Discuss: what is the difference between the results? What does this tell us about how to evaluate health claims in our own lives?
💡 Low-resource tipNo materials needed. The examples should be genuinely local — the specific traditional remedies and commercial health products that circulate in students' communities are far more useful than abstract imported examples. Treat traditional medicine respectfully and honestly: some traditional remedies pass all five tests and some do not.
Activity 3 — Scientific consensus: when is the science settled?
PurposeStudents understand the difference between a single study and a scientific consensus — and develop the ability to assess the strength of evidence on questions of public importance.
How to run itIntroduce the problem: one study is not enough. A single study can be wrong — through poor design, statistical chance, researcher bias, or fraud. Scientific knowledge becomes reliable when many independent studies reach the same conclusion. Present three scientific questions at different stages of evidence development. Strong consensus — evolution, climate change, vaccine safety: thousands of studies from many independent research groups worldwide, all reaching consistent conclusions, all using different methods. The consensus is robust. Active research — the specific mechanisms of a complex disease, the best approach to a specific environmental problem: many studies exist but they do not all agree, the methods differ, and knowledge is genuinely evolving. Genuine uncertainty here is appropriate. Weak or no evidence — a specific new health supplement, a new treatment that has only been tested in one study: the evidence is insufficient to draw conclusions. Now discuss: how do you find out what the scientific consensus is on a topic? (Look for statements from major scientific organisations — the WHO, national health authorities, relevant scientific bodies — rather than individual studies.) How should you respond when someone tells you that scientists disagree on something where there is actually a strong consensus? (Ask what proportion of relevant scientists disagree and why — there is almost always some disagreement, but 97 versus 3 percent is very different from 50 versus 50 percent.)
💡 Low-resource tipWorks entirely through discussion. The three examples should include topics relevant to the students' community — local environmental questions, local health questions. The concept of checking what relevant scientific organisations say is the most practical takeaway and should be emphasised.
Reflection Questions
  • Q1Think of something you believe is true about the natural world. How do you know it is true? What evidence do you have?
  • Q2Have you ever changed your mind about something because of evidence? What happened?
  • Q3Is there a health claim in your community that you think should be tested more carefully? What would a good test look like?
  • Q4What is the difference between a scientific theory and a guess? Why does the word theory mean something different in science than in everyday speech?
  • Q5When is it reasonable to trust traditional knowledge without scientific testing? When is it important to test it?
  • Q6Can science answer all questions — or are there questions that science cannot address? What kinds of questions are those?
Practice Tasks
Task 1 — Investigate a local claim
Choose a claim about the natural world that people in your community make — about plants, animals, weather, soil, health, or farming. Write: (a) the claim; (b) what evidence currently supports it; (c) how you would design a fair test to investigate it; (d) what result would support the claim and what would not support it; (e) whether you currently believe the claim, and why. Write 4 to 6 sentences plus your test design.
Skills: Applying scientific thinking to a genuine local knowledge claim — practising respectful but rigorous evaluation of community knowledge
Model Answer

The claim I am investigating is that planting maize with beans nearby increases the yield of both crops compared to planting them separately. Current evidence supporting this comes from the experience of several older farmers in our community who have practised this for many years and report consistently better harvests. To test this fairly, I would plant six identical plots: three with maize and beans together and three with maize alone, using the same seed varieties, the same soil, the same amount of water and no other plants. The result that would support the claim is that the maize and bean plots produce measurably more of both crops than the separate plots after the same growing period. A result that does not support it would be no consistent difference, or a higher yield in the separate plots. I currently believe the claim is probably true because the farmers who report it have many years of careful observation, but I would like to test it under controlled conditions because other factors — different soil, different weather years — might explain the difference they have noticed.

Marking Notes

Award marks for: a genuine and specific local claim; honest assessment of current evidence without dismissing community knowledge; a test design that correctly identifies the variable, controls, and measurable outcome; a clear statement of what would and would not support the claim; and a personal assessment that balances respect for traditional knowledge with scientific openness. Strong answers will acknowledge that the absence of controlled testing does not make the community knowledge wrong — it makes it unverified.

Task 2 — Evaluate a health claim
Choose a health claim that is common in your community. Apply the five-question checklist: (a) is it falsifiable?; (b) has it been tested in controlled studies?; (c) have the results been replicated?; (d) has it been reviewed by relevant experts?; (e) is there a plausible mechanism? Write your verdict: scientific, pseudoscientific, or uncertain — with clear reasoning. Write 4 to 6 sentences.
Skills: Applying the pseudoscience checklist to a real local health claim — practising the critical evaluation skill that protects against harmful health decisions
Common Mistakes
Common misconception

A scientific theory is just a guess — it has not been proved yet.

What to teach instead

In everyday speech, theory means guess or speculation. In science, theory has a completely different meaning: a well-tested, comprehensive explanation that is supported by a large body of evidence from many independent sources. Evolution is a theory. Gravity is a theory. Germ theory of disease is a theory. These are not guesses — they are the best-supported explanations we have, each backed by decades or centuries of evidence from many different types of investigation. When someone says evolution is just a theory, they are using the everyday meaning of theory in a context where the scientific meaning applies.

Common misconception

Scientists agree on everything — if scientists disagree, then no one can know the truth.

What to teach instead

Scientific disagreement is normal and healthy in areas where evidence is still accumulating. Scientists debate the details, mechanisms, and implications of almost every active area of research. But disagreement about specific details does not mean there is no scientific consensus on broad conclusions. Scientists disagree about the precise mechanisms of climate change while overwhelmingly agreeing that it is happening and that human activity is the primary cause. Disagreement in science is usually about the frontier — the edge of current knowledge — not about well-established conclusions. Confusing frontier debate with the rejection of established conclusions is a common technique used to manufacture false doubt about scientific consensus.

Common misconception

If something has worked for many people, it is scientifically proven.

What to teach instead

Many people using something and reporting it worked is testimonial evidence — not the same as controlled experimental evidence. Testimonials are unreliable for several reasons: people tend to remember when something worked and forget when it did not; many conditions improve on their own regardless of treatment; people who believe a treatment will work often experience genuine improvement (placebo effect); people who do not improve may stop using the treatment and not report their experience. Controlled trials eliminate these biases by comparing groups who receive the treatment with groups who do not. Without this comparison, testimonial evidence cannot distinguish between a treatment that genuinely works and one that does not.

Common misconception

Science is a Western invention that does not apply to other cultures.

What to teach instead

Systematic empirical inquiry — observing carefully, testing ideas, revising understanding based on evidence — is a universal human capacity that has been practised in every culture. The formal scientific method, as taught in schools, was largely codified in Europe, but it built on knowledge systems from across the world: Arabic mathematics, Chinese technology, Indian medicine, African astronomical knowledge. Modern science is a global enterprise practised in every country. The methods of science — careful observation, controlled testing, honest reporting of evidence — are tools available to everyone, not the property of any culture.

Key Ideas at This Level
1 Philosophy of science — what science is and how it works
2 The replication crisis — why published science is sometimes wrong
3 Science and values — how social context shapes scientific questions and answers
4 The limits of science — what science can and cannot address
5 Science communication — how scientific knowledge reaches the public
6 Citizen science — how non-scientists contribute to scientific knowledge
Teacher Background

Scientific thinking at secondary level engages students with the philosophy of science — what distinguishes science from other forms of knowledge, how scientific knowledge changes over time, the social dimensions of science, and the genuine limits of scientific inquiry.

Philosophy of science

Thomas Kuhn's The Structure of Scientific Revolutions (1962) is the most influential account of how science actually develops — not through the steady accumulation of facts but through periods of normal science (working within a paradigm) punctuated by revolutionary paradigm shifts when accumulated anomalies force a fundamental rethinking. Kuhn's concept of paradigm shift is one of the most important ideas in twentieth-century intellectual history. Karl Popper's falsifiability criterion — discussed at primary level — is the most influential philosophical account of what distinguishes science from pseudoscience, though it has been significantly critiqued and developed by subsequent philosophers. Lakatos, Feyerabend, and Latour each offer important challenges to simple pictures of scientific method. The replication crisis: a significant proportion of published results in psychology, nutrition, medicine, and other fields cannot be replicated — meaning the original results were wrong or misleading. Causes include publication bias (positive results are more likely to be published than null results), small sample sizes, undisclosed flexibility in data analysis, and outright fraud. Understanding the replication crisis is essential for calibrated trust in scientific research: peer-reviewed publication is necessary but not sufficient for confident acceptance.

Science and values

The choice of what questions to study is not value-neutral — it reflects the interests of those who fund and conduct research. The historical exclusion of women, non-Western populations, and low-income communities from many research samples has produced knowledge that is not universally applicable. Climate science and COVID research have demonstrated how political interests can distort both the conduct and communication of science.

Science communication

The relationship between scientific research and public understanding is complex and often distorted — through media simplification, sensationalism, selective reporting, and deliberate manufacturing of doubt. Understanding how scientific knowledge travels from laboratory to public and how it can be distorted along the way is essential for scientific literacy.

Key Vocabulary
Paradigm
Kuhn's term for the framework of assumptions, methods, and standards that defines normal science in a field at a given time. A paradigm shift occurs when anomalies accumulate to the point where the existing framework is replaced.
Null hypothesis
The default assumption in a scientific test — typically that there is no effect or no difference. The experiment attempts to find evidence strong enough to reject the null hypothesis.
Publication bias
The tendency for positive results to be published and negative results to be unpublished — producing a distorted picture of the evidence, because studies that found no effect are hidden from view.
Replication crisis
The finding that a significant proportion of published scientific results cannot be reproduced by other researchers — indicating that many accepted findings may be false or overstated.
Peer review
The process by which a scientific paper is evaluated by other experts in the field before publication. Peer review improves quality but does not guarantee correctness — many flawed studies have passed peer review.
Confounding variable
A variable that affects both the independent and dependent variables in a study — making it appear that one thing causes another when both are actually caused by a third factor.
Science communication
The process of conveying scientific knowledge to non-specialist audiences — including journalism, social media, public health communication, and policy advice. Poor science communication can distort public understanding even when the underlying research is sound.
Citizen science
Scientific research conducted partly or wholly by non-professional scientists — members of the public who contribute observations, data, or analysis. Citizen science can extend the reach of research far beyond what professional scientists could achieve alone.
Manufactured doubt
A deliberate strategy — typically funded by commercial or political interests — to create the appearance of scientific uncertainty about well-established conclusions, in order to delay policy action or regulatory response.
Occam's razor
The principle that, when multiple explanations fit the evidence equally well, the simplest one should be preferred. Not an absolute rule, but a useful guide for scientific reasoning.
Skill-Building Activities
Activity 1 — Paradigm shifts: how science really changes
PurposeStudents understand Kuhn's account of how scientific knowledge actually develops — through periods of normal science and revolutionary paradigm shifts — replacing the naive picture of steady linear progress.
How to run itIntroduce Kuhn's framework in accessible terms. Normal science: most scientific work takes place within an established framework — a paradigm — that defines which questions are worth asking, which methods are acceptable, and what counts as a good answer. Scientists working within a paradigm do not question its basic assumptions — they apply it, extend it, and solve puzzles within it. Anomalies: over time, some observations do not fit the paradigm. Initially these are explained away, ignored, or treated as measurement errors. But as anomalies accumulate, the pressure for a new framework grows. Paradigm shift: eventually, a new framework is proposed that explains both the old observations and the anomalies. The shift from the old paradigm to the new one is not gradual but revolutionary — it changes not just what scientists believe but how they see their entire field. Present two examples in accessible form. The shift from an Earth-centred to a Sun-centred picture of the solar system: the old paradigm worked well for basic predictions but accumulated anomalies (the strange movements of planets) that required increasingly complex explanations. The new paradigm was simpler and explained everything better. The shift from the idea that stomach ulcers were caused by stress to the discovery that they are caused by the bacterium H. pylori: Barry Marshall could not get his colleagues to believe him — they were so committed to the stress paradigm that they dismissed his evidence. He famously drank a solution of the bacteria himself to prove it. Ask: what does this tell us about how scientific knowledge changes? Is the history of science a story of steady progress — or of revolution?
💡 Low-resource tipWorks entirely through discussion. The H. pylori example is particularly powerful because it is recent, because it shows how paradigms resist evidence, and because the heroic self-experiment is memorable. Use locally relevant examples of changing scientific understanding where possible.
Activity 2 — The replication crisis: why published science is sometimes wrong
PurposeStudents understand the replication crisis and its causes — developing calibrated trust in scientific research rather than either naive acceptance or wholesale cynicism.
How to run itPresent the problem: in the last two decades, researchers have tried to replicate hundreds of studies published in top scientific journals — and have found that many results could not be reproduced. In psychology, a large project found that only about 40 percent of published studies replicated successfully. In nutrition research, the situation is similar. Introduce four causes of the replication crisis. Publication bias: studies that find an effect are published; studies that find no effect are often not published. The published literature therefore looks more positive than the full picture would. P-hacking: researchers can adjust their analysis methods until they find a statistically significant result — a practice that produces apparently significant results that are actually due to chance. Small samples: studies with small numbers of participants are more likely to produce extreme results by chance — which look significant but do not replicate in larger studies. Fraud: a small proportion of published research involves fabricated or manipulated data. Now ask: does the replication crisis mean we should not trust science? Introduce the calibrated response: individual studies should be trusted cautiously, especially from small samples on surprising claims. Systematic reviews and meta-analyses that combine many studies are more reliable. The replication crisis shows that science is self-correcting — the problem was found and is being addressed. It does not show that all scientific claims are equally uncertain.
💡 Low-resource tipWorks entirely through discussion. The examples should include scientific claims that students have encountered — particularly in health and nutrition, where the replication crisis has had the most practical consequences. The message should be calibrated trust, not cynicism.
Activity 3 — Manufactured doubt: how science is distorted for political purposes
PurposeStudents understand the deliberate manufacture of scientific doubt as a strategy — equipping them to recognise it in current debates about climate, health, and environment.
How to run itIntroduce the concept of manufactured doubt — the deliberate strategy of creating the appearance of scientific uncertainty to delay policy action, even when the scientific consensus is clear. Present the tobacco example: in the 1950s and 1960s, as scientific evidence about the harm of smoking accumulated, tobacco companies funded research designed to create doubt, emphasised disagreements among scientists, and used the language of scientific uncertainty to argue that the evidence was not yet sufficient to justify regulation. The strategy worked for decades — tobacco companies used it to delay regulation while continuing to sell a product they knew was killing people. Now present parallel strategies in current debates. Climate change: the scientific consensus is overwhelming, but organised campaigns have successfully created public doubt by emphasising minor disagreements, funding sceptical scientists, and using the language of scientific uncertainty strategically. Vaccine safety: the consensus is clear, but a small number of retracted studies (including the fraudulent Wakefield MMR study) have been amplified to create ongoing hesitancy. Ask: how do you distinguish between genuine scientific uncertainty (where the evidence is genuinely unclear) and manufactured doubt (where a clear consensus is being strategically obscured)? Introduce four markers of manufactured doubt: funding by commercially interested parties, focus on minor disagreements rather than broad consensus, use of non-specialist scientists as spokespersons, and the claim that more research is always needed before any action can be taken.
💡 Low-resource tipWorks entirely through discussion. Use locally relevant examples of manufactured doubt where possible — the tobacco example is universal, but many industries and governments have used similar strategies on environmental, health, and food safety issues relevant to specific communities.
Reflection Questions
  • Q1Kuhn argues that scientific revolutions happen when anomalies accumulate to the point that the existing paradigm must be replaced. Does this mean that science is not a reliable path to truth — or does it mean something else?
  • Q2The replication crisis shows that many published scientific results are wrong. How should this change the way we trust scientific research — in health, in policy, in everyday life?
  • Q3Science communication often simplifies complex research in ways that distort the findings. Is this inevitable — or is it possible to communicate science accurately to non-specialist audiences?
  • Q4Science has historically been used to justify racism, colonialism, and the oppression of women. Does this history undermine science — or does it reveal something about how science should be reformed?
  • Q5Can science settle ethical and political questions — such as how to respond to climate change, or what the right immigration policy is? Or does science only provide part of what is needed to answer such questions?
  • Q6Is traditional ecological knowledge scientific? What makes it similar to and different from academic science?
Practice Tasks
Task 1 — Science in public life
Choose a current public debate in which scientific evidence is being discussed — climate change, vaccine policy, a local environmental issue, a health question. Write: (a) what the scientific consensus is and how strong it is; (b) what doubts or objections are being raised and by whom; (c) whether the doubt appears genuine or manufactured; (d) what you think the right response to the evidence is. Write 300 to 400 words.
Skills: Applying scientific thinking to a real public controversy — distinguishing genuine scientific uncertainty from manufactured doubt
Task 2 — Essay: science and knowledge
Choose ONE of the following questions and write a 400 to 600 word essay. (a) Kuhn argued that science does not progress steadily towards truth but through revolutions that replace one paradigm with another. Does this mean that scientific knowledge is just one perspective among many — or does it mean something less radical? (b) The replication crisis suggests that the scientific publication system is broken. What would a better system look like — and who would be responsible for creating it? (c) Science has limits — questions about values, meaning, and justice cannot be settled by scientific evidence alone. Does acknowledging these limits undermine the authority of science — or strengthen it?
Skills: Constructing a reasoned argument about the nature, limits, and authority of scientific knowledge
Common Mistakes
Common misconception

Peer review guarantees that a published study is correct.

What to teach instead

Peer review is an important quality check but not a guarantee. Reviewers cannot check the raw data, cannot always detect subtle methodological problems, and are sometimes subject to the same biases as the researchers whose work they review. A large proportion of peer-reviewed studies fail to replicate. Some peer-reviewed studies have turned out to involve fraud. Peer review is necessary for scientific publication but not sufficient for confident acceptance. Studies that replicate across multiple independent groups, with large samples, and are consistent with related evidence from other methods, deserve much more confidence than single peer-reviewed studies.

Common misconception

Science and religion are necessarily in conflict.

What to teach instead

The relationship between science and religion is complex and historically variable. Many scientists hold religious beliefs. Science and religion address different kinds of questions: science addresses empirical questions about how the natural world works; religion addresses questions of meaning, purpose, value, and transcendence. Conflicts arise when either domain makes claims in the other's territory — when religious authorities make specific empirical claims about the natural world, or when scientific findings are extended beyond evidence into philosophical or moral conclusions they do not support. The majority of working scientists in most fields do not experience their scientific and religious or spiritual commitments as incompatible.

Common misconception

Science is politically neutral — scientific facts speak for themselves.

What to teach instead

Science is conducted by human beings in specific social and political contexts, and those contexts shape what questions are asked, what populations are studied, whose priorities are reflected in research funding, and how results are communicated and used. This does not make scientific findings false — but it does mean that the claim of pure political neutrality is itself a political claim, one that has historically served to shield science from legitimate accountability. Feminist science scholars, postcolonial science studies, and the sociology of scientific knowledge have all documented systematic biases in what gets studied, who does the studying, and whose knowledge is validated.

Common misconception

Scientific thinking and spiritual or religious thinking are incompatible ways of seeing the world.

What to teach instead

Scientific thinking — careful observation, testing ideas, revising beliefs based on evidence — and spiritual or religious thought operate in different domains and with different goals. Most people, in most cultures, hold both simultaneously without experiencing them as fundamentally incompatible. The historical cases of apparent conflict — Galileo, evolution — typically involve specific institutional claims rather than the inherent incompatibility of scientific and religious thought. Many of the founders of modern science were devoutly religious. The existence of deep questions that science cannot answer — about consciousness, meaning, value, and the nature of existence — leaves genuine space for both.

Further Practice & Resources

Key texts and resources: Thomas Kuhn's The Structure of Scientific Revolutions (1962, University of Chicago Press) is the most important work in the philosophy of science of the 20th century — readable and genuinely mind-changing. Karl Popper's The Logic of Scientific Discovery (1959, Hutchinson) is the foundational text on falsifiability. For the replication crisis: Brian Nosek's Open Science Collaboration paper (Science, 2015) is the definitive published account — freely available online. Stuart Ritchie's Science Fictions (2020, Metropolitan Books) is the most accessible book-length treatment. For manufactured doubt: Naomi Oreskes and Erik Conway's Merchants of Doubt (2010, Bloomsbury) documents the deliberate manufacture of scientific uncertainty by tobacco, fossil fuel, and chemical industries — one of the most important books for scientific literacy. For science communication: Ben Goldacre's Bad Science (2008, Fourth Estate) remains the most entertaining and practical guide to evaluating health claims. For philosophy of science accessible to students: Peter Godfrey-Smith's Theory and Reality (2003, University of Chicago Press) is the best undergraduate-level introduction. For science and indigenous knowledge: the IPBES assessment on indigenous and local knowledge (ipbes.net) documents the scientific value of traditional ecological knowledge. For citizen science: SciStarter (scistarter.org) provides access to hundreds of citizen science projects globally. For African scientific traditions: the work of the African Institute for Mathematical Sciences (aims.ac.za) and historians of African science including Cheikh Anta Diop (featured in the thinker library) provide important context for the African contribution to scientific knowledge.