What Makes a Claim Scientific?

A scientific claim is an explanation or prediction about the natural world that can be investigated using observation, measurement, or experimentation. A claim does not become scientific merely because it uses technical language, includes a graph, or is presented by someone wearing a laboratory coat. What matters is whether the claim can be tested and whether the evidence supporting it can be examined by others.

Scientific claims generally share several important characteristics:

• They are testable. It must be possible to design an observation or experiment that could support or contradict the claim.
• They make measurable predictions. The expected outcome should be described clearly enough that it can be compared with actual data.
• They are based on evidence. Conclusions must be supported by observations or measurements rather than personal belief, tradition, or authority alone.
• They can be evaluated by others. Scientific methods and results should be described clearly enough for independent researchers to examine or repeat the work.
• They can be revised. Scientific explanations must change when new and reliable evidence shows that an earlier explanation is incomplete or incorrect.

Consider the following two claims:

“This medication lowers systolic blood pressure by an average of 10 mmHg when taken daily for eight weeks.”

“This product restores the body’s natural energy balance.”

The first claim identifies a measurable quantity, a treatment period, and an expected outcome. Researchers could test it by comparing blood-pressure measurements in treated and untreated groups. The second claim is much more difficult to evaluate because the terms energy balance and restore are not clearly defined or connected to measurable quantities.

Science Is Self-Correcting

Scientific knowledge changes over time. This does not mean that science is unreliable. Instead, the willingness to revise explanations is one of its greatest strengths.

Scientific explanations are based on the best available evidence. When new observations become available, scientists may refine an existing model, identify limits to a theory, or replace an explanation that no longer agrees with the evidence. Newton’s laws of motion, for example, remain extremely accurate for everyday objects, even though relativity provides a more complete description at very high speeds or in very strong gravitational fields.

The same process occurs in medicine. Treatments and clinical recommendations may change as larger studies are completed, new side effects are discovered, or improved diagnostic tools become available. A recommendation that changes in response to new evidence demonstrates that the scientific process is working as intended.

Science and Pseudoscience

Pseudoscience refers to claims, practices, or explanations that are presented as scientific but do not follow the methods and standards used in science. Pseudoscientific claims may use scientific terminology, refer to unpublished “studies,” or include personal testimonials. However, they often avoid meaningful testing or continue to be promoted even when reliable evidence contradicts them.

The difference between science and pseudoscience is not always determined by the subject of the claim. A claim about nutrition, herbal medicine, exercise, or another complementary treatment can be investigated scientifically. The important question is not whether the treatment is considered conventional or alternative. The important question is whether its benefits and risks have been evaluated using reliable evidence.

Common Warning Signs

No single warning sign automatically proves that a claim is pseudoscientific. However, several warning signs appearing together should lead to careful examination.

• Promises of a universal cure. A product is claimed to treat many unrelated conditions.
• Reliance on testimonials. Personal stories are presented in place of controlled evidence.
• Claims of hidden knowledge. Supporters say that scientists, physicians, governments, or companies are suppressing the truth.
• Vague scientific language. Technical terms are used without clear definitions or measurements.
• Guaranteed results. The claim leaves little room for uncertainty, individual variation, or treatment failure.
• Rejection of criticism. Negative results are blamed on the researcher, the patient, or improper belief in the treatment.
• Dependence on a single study. One unusual result is presented as conclusive while the larger body of evidence is ignored.
• Appeals to nature. A product is described as safe or effective simply because it is natural.

Evidence Is More Than Personal Experience

Personal experiences can be meaningful and may help researchers identify questions worth studying. However, an individual experience is not the same as scientific evidence. A personal account is called anecdotal evidence.

Suppose someone takes a supplement while recovering from an illness and feels better two days later. Several explanations are possible:

• The supplement may have helped.
• The illness may have improved naturally.
• Another treatment may have caused the improvement.
• The person may have changed diet, sleep, or activity at the same time.
• The expectation of improvement may have influenced how the symptoms were perceived.

Without a comparison group, it is difficult to determine which explanation is correct. Controlled studies are designed to separate the effect of a treatment from other influences.

Control Groups and Placebos

In a controlled study, researchers compare an experimental group receiving the treatment with a control group that does not receive the active treatment. In some studies, the control group receives a placebo, an inactive treatment made to resemble the treatment being tested.

A placebo helps researchers account for changes caused by expectations, natural recovery, attention from healthcare providers, or other factors unrelated to the active treatment.

Randomization and Blinding

In a randomized controlled trial, participants are assigned to groups using a random process. Random assignment helps distribute differences in age, health, behavior, and other characteristics across the groups.

In a blinded study, participants may not know which treatment they receive. In a double-blind study, neither the participants nor the researchers directly interacting with them know who receives the active treatment. Blinding reduces the risk that expectations will influence measurements or interpretation.

Peer Review, Reproducibility, and Scientific Consensus

Peer Review

Before scientific research is published in many scholarly journals, it is evaluated through peer review. Independent specialists examine the methods, reasoning, and interpretation of the results. They may identify errors, request clarification, or recommend additional analysis.

Peer review is not a guarantee that a study is correct. Reviewers can miss errors, and published research can later be challenged. However, peer review provides an important level of examination before research enters the scientific literature.

Reproducibility

A scientific result becomes more convincing when independent researchers obtain consistent findings. Reproducibility refers to the ability of other researchers to use the same data or methods and obtain consistent results. Closely related is replication, in which a new investigation is performed to determine whether a result appears again.

A single study may be affected by chance, measurement error, an unusual group of participants, or unrecognized bias. For this reason, strong conclusions rarely depend on only one experiment.

Scientific Consensus

A scientific consensus develops when many independent lines of evidence support the same general conclusion. Consensus does not mean that every scientist agrees about every detail. It means that the available evidence strongly favors one explanation over the alternatives.

Scientific consensus is especially important when making healthcare and public-policy decisions. A single study that disagrees with hundreds of other investigations may be worth examining, but it should not automatically be treated as equally reliable as the larger body of evidence.

Evaluating Health and Technology Claims

Advertisements and social media posts often present statistics without enough context. A statement such as “97% of users reported improvement” may sound impressive, but it leaves many questions unanswered.

When evaluating a claim, ask:

• How many people participated?
• How were participants selected?
• Was there a control or placebo group?
• Were participants assigned randomly?
• Were the outcomes measured objectively?
• How large was the improvement?
• Were side effects reported?
• Was the research peer reviewed?
• Have independent researchers obtained similar results?
• Who funded the research?
• Does the claim agree with the broader body of evidence?

Funding does not automatically invalidate research, but financial interests should be disclosed because they may introduce bias. Similarly, a prestigious title or institution does not replace the need for good methods and reliable evidence.

A Practical Checklist

Before accepting a scientific or medical claim, consider the following questions:

1. Is the claim clearly stated? The outcome and relevant quantities should be defined.
2. Can the claim be tested? There should be a possible observation that could support or contradict it.
3. What evidence is provided? Controlled measurements are stronger than testimonials.
4. Is the evidence publicly available? Methods and results should be open to examination.
5. Has the result been independently reproduced? Repeated findings are more reliable than isolated results.
6. Are uncertainty and limitations acknowledged? Reliable scientific reports rarely claim perfect certainty.
7. Does the claim fit the broader evidence? A conclusion should be compared with the full scientific literature.
8. Would the promoters change their position if contrary evidence appeared? A scientific claim must remain open to revision.

Science Does Not Require Absolute Certainty

Science rarely provides absolute proof. Instead, evidence may strongly support one explanation while making alternatives less likely. Scientific conclusions are often expressed in terms of probability, confidence, uncertainty, and the limits of measurement.

This does not mean that every explanation is equally likely. Some scientific conclusions are supported by enormous quantities of evidence and can be used with great confidence. For example, the germ theory of disease, the conservation of energy, and the relationship between smoking and lung cancer are supported by many independent lines of evidence.

Scientific uncertainty is not ignorance. It is an honest description of what is known, how precisely it is known, and what questions remain.

Check Your Understanding

A social media advertisement claims that wearing a magnetic bracelet increases blood oxygen concentration, improves circulation, reduces inflammation, strengthens the immune system, and increases energy. The advertisement includes many customer testimonials but provides no published research or measurable explanation. Which warning signs discussed in this chapter appear in the advertisement? What evidence would be needed to evaluate the claims scientifically?