ACT® · Science: Evaluating Scientific Arguments and Models with Evidence (8-13 items)
Testing a Hypothesis Against Two or More Presentations (EMI 501 and 701)
When a claim is checked against two figures instead of one, each figure has to be matched to exactly what it can and can't prove, because a claim with two parts, or one with hidden conditions, can look true in one figure and false in the next.
One graph can make almost any claim look reasonable, because it only shows one slice of the story. The moment a second figure sits next to it, that slice has to agree with a different slice taken under different conditions, and that is where weak claims fall apart. A claim that growth needs both light and nitrogen is really two claims stitched together. A figure that raises light while holding nitrogen fixed can only speak to light. A figure that raises nitrogen while holding light fixed can only speak to nitrogen. Neither one ever removes a factor, so neither can prove that factor is necessary, no matter how clean the rise looks.
The working method is simple: name what each figure actually varies and what it holds fixed, then ask whether the claim's full wording has been matched piece by piece. If a claim says rate rises with pressure, and it does at one temperature but reverses at another, the two figures together do not just contradict each other. They tell you the original claim is incomplete, true only under a condition it forgot to mention. That is a sharper, more useful conclusion than just picking a winner between the figures.
This connects directly to the unit's bigger theme of testing models against evidence: a model or claim is only as good as the conditions under which it was checked. It also sets up the unit case where two studies report the same quantity in different units, like g per m2 against mg per m2. Before any claim about agreement or disagreement can be judged, the numbers have to be converted onto one scale. Skipping that step makes any comparison meaningless, no matter how the figures look side by side.
Key terms
- Held-constant variable
- The factor a figure keeps fixed while it varies another, which means that figure says nothing about the fixed factor's role.
- Hidden condition
- A requirement, like a specific temperature, that a claim depends on but never states outright.
- Unit conversion
- Rescaling one measurement's units to match another's so the two values can be directly compared.
- Compound claim
- A claim that bundles two or more separate conditions together, such as needing both light and nitrogen.
Watch it explained
- How it worksHow do you check one prediction against two figures without losing track of which one supports what?
- Worked exampleFind whether a claim that growth needs both light and nitrogen survives one graph of each plotted alone.
- Going furtherWhat changes when two presentations record the same underlying quantity in different units?
- Common mistakeWhy isn't support from one figure enough when a claim makes two separate promises at once?
Check yourself
End of topic.