ACT Science — data, design and two scientists

Thirteen questions on the three things this section actually asks — and almost none of them require you to know any science.

Reading results. Find a value, describe a trend, work out an average rate over a stated interval, recognize that a pressure and a volume whose product stays constant are inversely proportional. The arithmetic is small; the discipline is reading the units and the variable names before the numbers.

Judging how a study was built. Which variable was changed on purpose, which was measured, what a constant was holding steady and why, and why a student repeated each trial three times. These are worth learning as a short set of sentences, because they recur on every form and the wording barely changes.

Two scientists. Two short accounts of the same phenomenon that agree about most of it and part company on one step — a rock layer laid down fast by ash or slowly by marine sediment, atmosphere lost to solar wind or to gravitational escape, a mass extinction from an impact or from sustained volcanism. What would one of them predict; which described result would support or undermine one of them. The trap is always the same: the two agree about more than they disagree about, and a wrong option takes shared ground and presents it as the dispute.

The wrong answers misuse data that has been read correctly. That is the signature of this section, and it is why knowing the science can hurt you. An option compares the wrong pair, or asserts a trend two points cannot establish, or draws a cause from an observation, or takes the highest value that happened to be measured for the highest value there is.

That last one is worth dwelling on, and one question here is built entirely around it. Four pH values were sampled and the rate peaked at the third. The tempting answer is that the optimum is exactly that pH. It is not: nothing was measured between the sampled points, so the peak could sit anywhere in the gap. When a question asks what results support, the answer is the weakest claim the numbers actually carry — not the most interesting one.

What this cannot do, and it is half the section. The real ACT Science section is figures: graphs, tables, scatterplots, several at once, read fast. This material carries only a few drawings, five graphs and a pendulum, and every data set is also described in words with its units and its controls. That trains the judgment more than the eye.

So use these for the reasoning — what a trend supports, what a control was for, where two accounts diverge — and then go and read real figures under a clock on ACT's own free practice tests. And settle first whether science is even on your test: it no longer counts toward the Composite, and on a school administration the decision is not yours.

Every study described here is invented and framed as one experiment or one student's measurements, never as a published finding. Nothing is reproduced from any ACT publication.

• Identify the independent variable, the dependent variable and the purpose of a constant in a described study
• Say why a trial was repeated, and what repetition does and does not fix
• Compute an average rate of change over a stated interval from described values
• Recognize an inverse relationship from a product that stays constant
• Refuse a claim the sampled points cannot support — including treating the highest measured value as the true maximum
• Tell an observation from a cause, and refuse a causal conclusion drawn from neither
• Locate the single step on which two scientists' accounts differ, rather than the subject they share
• Say which described result would support or undermine one specific account

Practice material written for the ACT Science section, following the skills ACT publishes for it: interpreting data, understanding how a scientific investigation was designed, and evaluating competing models and explanations against evidence. Every study and dispute is invented; nothing is taken from any ACT publication or released test.