Science is a conversation
Science isn’t just doing experiments. It is a public conversation in which scientists make claims, share evidence, criticise each other’s work and slowly reach agreement. The tools of that conversation are scientific argument and peer review.
The structure of a scientific argument
A scientific argument is not a shouting match. It is a reasoned case built from evidence.
| Part | Purpose | Example |
|---|---|---|
| Claim | Answers the question | Microplastics are now found in most bottled water. |
| Evidence | Data supporting the claim | A study tested 259 bottles from 11 brands; 93% contained microplastic particles. |
| Reasoning | Links evidence to claim using scientific principles | Plastic caps and bottles shed particles through wear, so contamination during bottling is expected. |
| Counterargument | An alternative view | The particles might come from lab contamination. |
| Rebuttal | Answers the counterargument | The researchers ran blank controls and subtracted background particles. |
Strong arguments consider alternative explanations instead of ignoring them.
Judging the quality of evidence
Not all evidence is equally strong. Ask:
- Sample size: was it large enough?
- Controls: were confounding variables managed?
- Source: was it published in a peer-reviewed journal?
- Replication: have other teams found the same thing?
- Conflicts of interest: who funded it, and could that affect the results?
| Stronger evidence | Weaker evidence |
|---|---|
| Large, randomized, controlled trials | A single anecdote |
| Replicated by independent teams | One unrepeated study |
| Peer-reviewed publication | Press release or social media post |
| Systematic review of many studies | Opinion of one expert |
How peer review works
Before a study is published in a scientific journal, it usually goes through peer review:
- The authors submit their paper to a journal.
- An editor sends it to two or three independent experts (the "peers").
- Reviewers check the methods, analysis and conclusions, looking for errors or weak reasoning.
- They recommend acceptance, revision or rejection.
- The authors often revise the paper and resubmit.
Peer review catches many problems, but not all. It is a filter, not a guarantee.
Replication
Replication means repeating a study to see if the results hold up. A result that can’t be replicated is treated with suspicion. In recent years, scientists in psychology, medicine and other fields have found that some well-known results did not replicate. This "replication crisis" led to reforms such as pre-registering studies and sharing data openly.
Retractions and self-correction
Sometimes published papers turn out to be wrong or even fraudulent. Journals can retract them. A famous example is a 1998 paper that falsely linked the MMR vaccine to autism. It was retracted in 2010, and many large studies since have found no link. Self-correction is one of science’s strengths.
Scientific consensus
When many independent lines of evidence point the same way, experts reach a consensus. Consensus is not a vote or an opinion poll. It reflects the weight of evidence. Examples include the consensus that human activities are the main cause of recent global warming and that vaccines are safe and effective.
Common mistakes
- Treating one study as proof.
- Attacking the person instead of the evidence.
- Cherry-picking only the studies that support your view.
Try it yourself
Read a short science news article. Then answer:
- What is the main claim?
- What evidence is given? Was it peer reviewed?
- How large was the study, and was it replicated?
- Write one counterargument and a possible rebuttal.
Quick recap
- A scientific argument has a claim, evidence and reasoning, and addresses counterarguments.
- Evidence quality depends on design, size, source and replication.
- Peer review is expert checking before publication; it filters but doesn’t guarantee.
- Replication and retractions help science correct itself.
- Consensus reflects the overall weight of evidence.