Natural science: turn observations into questions that evidence can test
Natural science investigates the physical and living world through observation, measurement and explanations that can be examined. Biology, chemistry and physics contribute to medical understanding in different ways. A useful starting point is learning to distinguish what was measured from the explanation offered for it.
An observation becomes useful when the question is clear
An observation describes something noticed or measured. A hypothesis proposes an explanation that can be investigated. The two should remain distinct, because the same observation may fit more than one explanation. A scientific question should make clear what would be examined and what kind of evidence could support or challenge the proposed account.
Imagine a fictional set of plant-growth measurements. A difference between groups is an observation; an explanation involving light is a hypothesis. Ask what else differed between the groups before assigning the cause. This simple example transfers to more complicated work: naming a plausible mechanism does not establish that it produced the result in a particular study.
Values need units, methods and an account of uncertainty
A measurement is more informative when its unit and method are stated. Results can vary because of the system being studied and the measurement process. Precision describes the closeness of repeated measurements, while agreement with a suitable reference is a different concern. Several similar readings do not guarantee that the method is giving an accurate account of the quantity.
For revision, compare a fictional temperature recorded in two different units. Check that the values have been put on the same scale before interpreting a difference. Then ask whether the readings were taken under comparable conditions. A number copied without its unit or method can appear exact while losing the information needed to understand what it represents.
An explanation has assumptions and a range in which it is useful
A model represents selected features of a system to help describe or explain it. Models can be conceptual, physical or mathematical. Simplifying a system makes some relationships easier to examine but may leave out features that matter in another setting. A model’s usefulness therefore depends on its purpose, assumptions and how well its predictions fit relevant observations.
Take a fictional diagram of blood flow that shows only a few connections. Explain which relationship it helps you understand and what detail it omits. The omitted detail does not automatically make the diagram useless. It does mean that a conclusion requiring that detail needs a different level of evidence. Judge the representation against the question it is intended to answer.
Transparent methods make claims easier to examine
Research reports should explain methods and uncertainty so that others can assess the work. Terms for repeating analyses and studies differ between fields. One National Academies report uses reproducibility for obtaining consistent results from the same data and analysis, and replicability for consistency across studies collecting their own data. State the convention being used rather than assuming the words are universal.
For a teaching report, identify the question, measurements, method and conclusion. Ask whether the conclusion extends beyond the evidence collected and what further work would test it. A later result can lead to refinement of an explanation rather than making every earlier observation meaningless. Scientific confidence depends on the quality and consistency of evidence, with uncertainty made visible.
Check your understanding
A short quiz, with an explanation after each answer.
Is a plausible explanation identical to an observation?
Return to the section you found hardest and explain its example without looking at the page. Check the explanation, then repeat the quiz later. Being able to describe the reasoning is more useful than remembering the position of a correct answer.