Identifying Scientifically Testable Claims
An ad says a new fertiliser boosts crop yield. A crystal seller says their stone carries healing energy. One of those claims can be tested with an investigation. The other never can, and knowing the difference is a core scientific skill.
Printable Worksheets
Print or save as PDF, or build a custom worksheet from any module's questions.
A friend reads you three claims: "Adding sugar to cut flowers makes them last longer", "This bracelet improves your luck", and "Drinking more water than anyone else is the healthiest thing you can do".
Which of these claims could you test with an investigation, and which could you not? What is it about a claim that lets science put it to the test?
A claim is a statement someone presents as true, such as "this fertiliser increases crop yield". A claim is scientifically testable when you can design an investigation, or a series of investigations, that could collect evidence to support it or show it is wrong. The key is that the claim must connect to something you can observe and measure, and the investigation must be able to come out either way.
That last point matters. A testable claim is one that an investigation could prove false if it really is false. "This fertiliser increases tomato yield" is testable because, if you grew tomatoes with and without it and counted no difference, the claim would be shown wrong. A claim no evidence could ever contradict is not scientifically testable, even if it sounds confident.
"This sunscreen blocks more UV than the leading brand" is a testable claim. You can put a UV sensor under each sunscreen, shine a known amount of ultraviolet light on both, and read off how much passes through. The numbers either support the claim or they do not.
Before a new sunscreen can be sold in Australia, the Therapeutic Goods Administration requires the SPF claim on the bottle to be backed by a standardised, testable investigation. The claim "SPF 50+" is only allowed because there is a defined experiment that can confirm or reject it.
A claim sounding confident or "sciencey" does not make it testable. "This water is energised at the quantum level" uses science words but names nothing you could ever measure, so no investigation could check it.
Know
- A claim is a statement presented as true, and only some claims can be tested with an investigation.
- A testable claim links to something you can observe and measure.
Understand
- Why a testable claim must be one that evidence could prove false.
- How a claim can be turned into a hypothesis you can investigate.
Can Do
- Sort claims into testable and untestable.
- Design one investigation, or a series, to test a chosen claim.
Wrong: If a claim sounds scientific, it must be testable.
Right: A claim is testable only if it links to something you can observe and measure.
Wrong: A claim no evidence could ever disprove is the strongest kind of claim.
Right: A scientific claim must be falsifiable, an investigation has to be able to show it wrong.
Wrong: Testing a claim once and finding support proves it is true forever.
Right: Evidence can support a claim, but a series of investigations builds far stronger confidence.
Wrong: A claim and a hypothesis are exactly the same thing.
Right: A hypothesis is the testable prediction you write to put a claim to the test.
A student lists three "testable" product claims to investigate. One can never be tested, click it.
- This toothpaste removes more stain than the leading brand.
- This bracelet surrounds you with a positive energy aura.
- This battery lasts longer in a torch than a standard battery.
Some claims can never be tested by an investigation, no matter how good your equipment. They tend to fall into a few groups. Vague or undefined claims use terms no one can measure, such as "this crystal has healing energy" or "this water is more pure than nature intended". There is no agreed way to measure "healing energy", so no fair test exists. Opinion claims ("this band makes the best music") state a preference, not a fact about the world. And some claims are written so they are impossible to prove false, such as "this charm protects you, but only when you truly believe", which always has an escape route if it fails.
An untestable claim is not always a dishonest one, but it does not belong to science. Science only deals with claims that an investigation could, at least in principle, support or knock down with evidence.
"Magnets in this insole boost your body's natural energy" cannot be tested, because "natural energy" is never defined. However, "wearing these insoles reduces reported foot pain after a week compared with plain insoles" is testable, because pain can be rated and the two groups compared.
Researchers at Australian universities such as UNSW often spend their first effort rewording a marketing claim into a testable one. "This supplement boosts wellbeing" becomes "does this supplement change a measured marker, such as blood vitamin D, over eight weeks?" Only the reworded version can enter a real study.
Beware claims with a built-in excuse, such as "this remedy works unless your energy is blocked". Whenever a claim explains away every possible failure in advance, it can never be shown false, which is exactly why it is not scientific.
To test a claim, you first turn it into a hypothesis: a clear, testable prediction that links a change you make to a result you can measure. A useful pattern is "if I change X, then Y will happen". Take the claim "this fertiliser increases tomato yield". The hypothesis becomes: "if tomato plants are given this fertiliser, then they will produce more fruit than plants given none". Now you have named what to change (the fertiliser) and what to measure (the number of tomatoes).
A good hypothesis is also falsifiable, your investigation must be able to show it wrong. If the fertilised and unfertilised plants produce the same amount of fruit, the hypothesis is not supported, and that is a valid, useful result. A claim you cannot turn into a falsifiable hypothesis is a sign the claim itself is not scientifically testable.
The claim "this stain remover gets out grass stains" becomes the hypothesis "if grass-stained cloth is washed with this remover, then less stain will remain than cloth washed with water alone". The change is the remover, the measure is how much stain is left, and the comparison makes the result clear.
When the CSIRO tested whether its wrinkle-resistant wool treatment really worked, it did not just repeat the claim. It wrote a hypothesis linking the treatment to a measurable outcome, the number of wrinkles after a standard wash, then ran the test. A claim becomes science only once it is a hypothesis you can check.
A hypothesis must predict something you can measure. "If I use this product, then I will feel more positive energy" is not a usable hypothesis, because "positive energy" has no measurable definition, so no result could ever support or reject it.
Once you have a testable hypothesis, you design an investigation to gather evidence: decide what to change, what to measure, and what to keep the same so the test is fair. Sometimes one investigation is enough. The claim "this fertiliser increases tomato yield" can be tested by one experiment with fertilised and unfertilised plants grown side by side.
Other claims need a series of investigations, because one cannot answer everything. Suppose the claim is "this fertiliser increases yield for all common vegetables". One experiment on tomatoes alone is not enough, you would also need to test it on beans, lettuce and carrots, then look at the results together. A series is also useful when you want to repeat a test to be sure the result was not a fluke, or to vary the dose to find how much fertiliser works best. Choosing between one investigation and a series depends on how broad the claim is.
The claim "this antiseptic kills more bacteria than soap" might start with one investigation comparing the two on a single bacterium. To support the broader claim, you would then run a series, testing several bacteria, repeating each test, and comparing all the results before deciding.
A broad claim tested by a single narrow experiment is a common mistake. Showing a fertiliser helps tomatoes does not prove it helps "all plants". Match how many investigations you run to how sweeping the claim actually is.
A company claims "this fertiliser increases the yield of every vegetable in your garden". You run one experiment on tomatoes and the yield goes up. Explain why a single experiment is not enough to support this particular claim.
How close was your prediction?
Nice, you saw that a broad claim needs a series of investigations, not one test.
Good to notice, one experiment only tests the narrow part of a broad claim.
Speed Round · 6 questions
True or false? Tap as fast as you can. Build a streak.
A scientifically testable claim must link to something you can observe and measure.
"This crystal carries healing energy" is a testable claim.
A good scientific claim must be falsifiable, an investigation has to be able to show it wrong.
A hypothesis is the testable prediction you write to put a claim to the test.
A broad claim, such as "this helps every vegetable", may need a series of investigations.
Once a claim is supported by one experiment, it is proven true forever.
How are you completing this lesson?
Think back to the three claims from the start: sugar making cut flowers last longer, a bracelet improving luck, and drinking the most water being healthiest.
Which claim is scientifically testable, and how would you turn it into a hypothesis you could investigate?
Quick Check · 5 questions
Check Your Understanding · 3 questions
1. Write one scientifically testable claim and one untestable claim about a food or drink, and explain what makes each one what it is.
2. Why is being falsifiable, able to be shown false, important for a claim to count as scientific? Give one example.
3. Turn the claim "this lamp helps seedlings grow" into a testable hypothesis, naming what you would change and what you would measure.
Show Your Working · 3 questions
SA1. Explain what makes a claim scientifically testable. Give one example of a testable claim and one example of an untestable claim, and say why each falls into its group.
SA2. A product is advertised with the claim "this crystal has healing energy". Explain why this claim cannot be scientifically tested, even though it is presented as a fact.
Hint: Think about measuring and about whether any evidence could ever prove it false.
SA3. A company claims "this fertiliser increases the yield of all common vegetables". Turn this into a testable hypothesis, then describe a series of investigations you could run to test the full claim, explaining why one experiment alone is not enough.
Quick Check
1. C. Mass lost to evaporation can be measured, so an investigation could support or disprove the claim.
2. D. A testable claim must be one an investigation can support or show false.
3. B. It names what to change (the soap) and what to measure (grease remaining) as an if-then prediction.
4. C. "Every common vegetable" is broad, so it needs a series of investigations across many vegetables.
5. C. Walking distance in 6 minutes is measurable, replacing the undefined term "inner energy".
Show Your Working Model Answers
SA1 (4 marks): A claim is testable when it links to something you can observe and measure [1] and an investigation could support it or show it false [1]. Testable example: "this fertiliser increases tomato yield", because you can count the fruit [1]. Untestable example: "this crystal has healing energy", because "healing energy" cannot be measured and no evidence could disprove it [1].
SA2 (4 marks): "Healing energy" has no agreed definition and no instrument can measure it [1], so there is no fair test you could design [1]. No possible result could ever show the claim false [1], which means it is not falsifiable and therefore not scientific [1].
SA3 (5 marks): Hypothesis: "if a vegetable is given this fertiliser, then it produces more yield than the same vegetable given none" [1]. Test it first on tomatoes with fertilised and unfertilised plants [1]. Because the claim covers all common vegetables, repeat the investigation on beans, lettuce and carrots [1], and repeat each test to check the result is reliable [1]. One experiment only supports the narrow claim about tomatoes, so a series is needed to test the broad claim [1].
Claim
A statement presented as true
Testable
Evidence could support or disprove it
Falsifiable
An investigation could show it wrong
Hypothesis
An if-then testable prediction
Untestable
Vague, opinion or never falsifiable
Series
Several tests for a broad claim
Put what you have learned to the test! Jump through the questions in game form.
Play GameYour Badges
0 of 6Mark lesson as complete
Tick when you've finished Learn, Practice and the game. Earns +85 XP and +25 coins.