97 terms

Scientific Thinking Glossary

The key vocabulary of Scientific Thinking, each term defined in plain language. Start learning in the Scientific Thinking subject, or drill these terms as flashcards.

Abiogenesis: The natural process by which life is thought to have arisen from non-living chemistry.
Accuracy: How close measurements are to the true value.
Applied research: Research directed at solving a particular practical problem or creating a useful technology.
Atom: The basic unit of a chemical element, made of a nucleus surrounded by electrons.
Base rate: How common something is beforehand, which changes what a positive test result means.
Basic reproduction number (R0): The average number of people one infected case transmits to in a fully susceptible population.
Basic research: Curiosity-driven investigation aimed at understanding, without a specific application in mind.
Big Bang: The leading scientific model for the origin and expansion of the universe from a hot, dense early state.
Big science: Large-scale research requiring major funding, instruments, and collaboration across many people.
Blinding: Keeping participants — and in double-blinding, the researchers — unaware of group assignment.
Bradford Hill considerations: A structured set of criteria — strength, consistency, dose-response, temporality, plausibility — for building a causal case without experiment.
Calibrated confidence: Stating belief proportionate to evidence, with the reasons attached.
Causation: A claim that changing one variable would change the other.
Chaos theory: The study of systems so sensitive to initial conditions that they become effectively unpredictable.
Confounding variable: A factor associated with the exposure that also independently affects the outcome.
Connectome: A complete map of the connections between neurons in a brain or nervous system.
Conservation of energy: The principle that energy is neither created nor destroyed, only transformed.
Control group: A comparison group as similar as possible to the treatment group, differing only in the intervention.
Controlled variable: A factor deliberately held constant so it cannot affect the comparison.
Correlation: A measured association between two variables.
Cosmic microwave background: Faint radiation left over from the early universe, key evidence for the Big Bang.
CRISPR: A precise, low-cost gene-editing tool established by Doudna and Charpentier in 2012.
Cumulative knowledge: The way science builds on and corrects earlier work over generations.
Dark matter: Unseen matter inferred from its gravitational effects, whose nature remains unknown.
Dependent variable: What is measured to see whether it responded — the suspected effect.
DNA: The molecule carrying genetic instructions, structured as a double helix.
Effect size: How large an effect is, as distinct from how confident we are it exists.
Element: A substance made of only one kind of atom, defined by its number of protons.
Emergence: When large systems display properties their individual parts do not, such as mind from neurons.
Empiricism: The view that knowledge comes primarily from observation and experience.
Evidence-informed decision: A decision that takes the best available science seriously as an input, then weighs values and trade-offs.
Evolution: Change in the heritable characteristics of populations over successive generations.
Extremophile: An organism that thrives in extreme conditions, expanding our sense of where life is possible.
Fact-value distinction: The difference between what is the case (which science can investigate) and what ought to be done (a question of values).
False precision: Reporting more digits than the measurement justifies.
Falsifiability: The property of a claim that some possible observation would show it to be wrong.
Gene: A unit of heredity, a segment of DNA that carries instructions for a trait.
Germ theory: The discovery that many diseases are caused by microorganisms, a landmark of 19th-century science.
Greenhouse effect: The warming of a planet’s surface as certain gases trap outgoing heat in the atmosphere.
Hard problem of consciousness: The unresolved question of why and how physical brain processes give rise to subjective experience.
Heliocentrism: The model, advanced by Copernicus, placing the Sun rather than the Earth at the centre of the solar system.
Herd immunity: The slowing of a disease’s spread once enough of a population is immune.
Hype cycle: The common pattern of inflated expectations, disappointment, then realistic productivity for a new technology.
Hypothesis: A proposed explanation with consequences that can be checked.
Independent variable: What the researcher deliberately changes — the suspected cause.
Inference: An explanation built on top of an observation.
Keeling curve: The continuous record of rising atmospheric CO2 begun by Charles David Keeling at Mauna Loa in 1958.
Margin of error: A range within which the true value probably falls.
Measurement uncertainty: An honest statement of how much a measured value could reasonably differ from the true value.
Model: A deliberately simplified representation used to reason or predict.
Multiple comparisons: Testing many outcomes, which makes a chance 'significant' result likely.
n-of-1 experiment: A personal experiment on a single individual — useful, limited, and not generalizable.
Natural philosophy: The study of nature before modern science, relying more on reasoning than systematic experiment.
Natural selection: The process by which heritable traits that aid survival and reproduction become more common over generations.
Null hypothesis: The starting assumption that there is no effect, which evidence must argue against.
Observation: What can be detected: seen, heard, counted, or measured.
Observer bias: The way expectation shapes what an observer notices and records, without dishonesty.
Open question: A scientific problem that remains genuinely unresolved, marking a frontier of current knowledge.
p-value: The probability of data at least this extreme if there were no real effect.
Paradigm shift: A fundamental change in the basic concepts and practices of a scientific discipline.
Peer community: The network of researchers whose scrutiny, replication, and debate give scientific claims their credibility.
Peer review: Assessment of a submitted paper by researchers in the field before publication.
Placebo: An inert intervention giving the control group the experience of treatment without the active component.
Practical significance: An effect large enough to matter for a real decision.
Pre-registration: Declaring hypotheses and analysis plans before collecting data.
Precautionary principle: The idea that a lack of full certainty should not delay protective action against serious or irreversible harm.
Precision: How tightly repeated measurements cluster together.
Prediction: What should be observed if a hypothesis holds.
Preprint: A paper shared publicly before peer review.
Provisional knowledge: Scientific conclusions held as the best current account, open to revision with new evidence.
Pseudoscience: Claims adopting scientific presentation while avoiding conditions that would let them fail.
Publication bias: The tendency for positive and novel results to be published over null ones.
Quantum entanglement: A quantum link where measuring one particle instantly constrains a partner, however far apart.
Quantum superposition: The quantum principle that a system can be in a combination of states until it is measured.
Random error: Variation scattering either side of the true value, reduced by more measurements.
Randomization: Assigning participants to groups by chance, distributing unanticipated confounders evenly.
Replication: Independent researchers repeating a study and obtaining a comparable result.
Responsible innovation: Developing technology with attention to its wider social, ethical, and environmental effects.
Retraction: Formal withdrawal of a published paper, whether for error or misconduct.
Reverse causation: The possibility that the supposed effect is actually the cause.
Risky prediction: A prediction that many possible outcomes would have refuted, and so carries weight when it survives.
Sampling bias: A sample unrepresentative of the population, producing a wrong estimate at any size.
Science: A systematic method for building knowledge about the observable world by proposing explanations and testing them against evidence.
Science literacy: The ability to reason with evidence, weigh uncertainty, and separate facts from values.
Scientific consensus: The state where evidence from many independent lines has converged.
Scientific law: A reliable description of what happens, often mathematical, distinct from an explanation of why.
Scientific method: The iterative loop of question, hypothesis, prediction, test, and revision — not a fixed sequence.
Scientific misconduct: Deliberate fabrication, falsification, or plagiarism.
Scientific Revolution: The 16th-17th century shift to observation, mathematics, and experiment that founded modern science.
Scientific society: An organisation, such as the Royal Society, for sharing and scrutinising scientific work.
Simpson's paradox: When a pattern present in every subgroup reverses once the groups are pooled.
Statistical significance: A result unlikely to have arisen from noise alone, by a chosen threshold.
Systematic error: A consistent shift in one direction that more measurements cannot fix.
Theory: A broad explanation supported by a large converging body of evidence.
Theory of relativity: Einstein’s theories describing how space, time, and gravity behave, refining Newtonian physics at extremes.
Truncated axis: A chart axis not beginning at zero, magnifying small differences.
Turing test: Alan Turing’s 1950 proposal to judge machine intelligence by whether it is indistinguishable from a human in conversation.

← All subject glossaries