Study guide
On the enhanced ACT rolled out beginning in April 2025, the science section is optional: it is a 40-question, 40-minute add-on that no longer counts toward your Composite score, which is computed from English, Math, and Reading alone. Despite its name, the section is really a test of reading data — charts, tables, and experimental descriptions — not of memorized science facts. This chapter covers the three question formats and the habits that let you extract answers from unfamiliar figures at speed.
An Optional Section: What Changed and Who Should Take It
Under the enhanced ACT — introduced for online national testing in April 2025, extended to paper national testing in September 2025, and reaching school-day testing in spring 2026 — science moved from required to optional, joining writing as an add-on. The core test is now English, Math, and Reading, and the Composite score (1-36) averages only those three sections. If you elect the science section, you receive a separate science score on the 1-36 scale, and ACT also reports a STEM score that averages your math and science results. Because science cannot lower your Composite, the decision to take it is strategic rather than defensive. Reasons to opt in: you plan to apply to STEM programs, engineering schools, or scholarships that value a science score; some institutions, scholarship programs, or state testing arrangements may still expect or require it, and requirements vary by college and by state, so verify the policies of every school on your list before registering. Reasons to skip it: your target schools do not use it, and the 40 minutes of additional testing and preparation time could go toward the core sections that actually drive your Composite. The section itself is 40 questions in 40 minutes — about a minute per question — built around six or so passages in three recurring formats: data representation, research summaries, and conflicting viewpoints. The skills involved overlap heavily with the reading section's evidence-first mindset, applied to figures instead of paragraphs.
Data Representation: Trends, Interpolation, and Extrapolation
Data representation passages hand you graphs and tables from a study with minimal text, then ask what the data show. The foundational skill is naming the trend between two variables. In a direct relationship, both move together: as temperature rises, so does the solubility of the salt. In an inverse relationship, one rises while the other falls: as altitude increases, air pressure drops. Some relationships change character partway — rising, peaking, then falling — and the test loves asking where the maximum occurs or over what interval the trend holds. Two skills get their own questions. Interpolation means estimating a value between measured data points. If a table shows a solution held 30 grams of solute at 20 degrees Celsius and 38 grams at 30 degrees, the value at 25 degrees is about 34 grams — read between the rows, assuming the trend continues smoothly. Extrapolation means extending the trend beyond the measured range: if each 10-degree increase adds roughly 8 grams, then 40 degrees predicts about 46 grams. Questions phrase these as 'the solubility at 25 degrees would most likely have been closest to...' Your job is to find the bracketing values and place the answer between or beyond them consistently with the trend. Precision matters less than direction and bracketing; the answer choices are usually spaced widely enough that a careful estimate lands cleanly. Always check units and axis scales before estimating, because a graph that jumps scales or switches from milliliters to liters is the classic ambush.
Research Summaries: Variables, Controls, and Experimental Design
Research summary passages describe two or three related experiments and ask you to reason about their design and results. Master the vocabulary first. The independent variable is what the experimenters deliberately change — the fertilizer concentration, the incline angle, the temperature. The dependent variable is what they measure in response — plant height, rolling distance, reaction time. Everything intentionally held the same across trials is a constant, and a control is a baseline trial or group used for comparison, often receiving none of the treatment: the plants given plain water while others get fertilizer solutions. Expect questions like 'In Experiment 2, which variable did the students intentionally vary?' — answered by scanning the setup paragraph — and 'What was the purpose of Trial 1?' — usually, to provide a baseline against which treatment effects can be measured. Design-extension questions ask how to test something new: 'Which change to Experiment 1 would best determine whether light intensity affects growth?' The correct answer varies only light intensity while keeping everything else identical, because a valid experiment changes one variable at a time. Also expect comparison questions across experiments — typically the experiments share a method but differ in one factor, and identifying that one difference unlocks the set. When results tables appear, questions blend design logic with the data-reading skills from the previous section: identify the trend within each experiment, then explain it using the variable that changed. Read the short setup text carefully once; nearly every design answer is sitting in it verbatim.
Conflicting Viewpoints: Managing the Debate Passage
One passage in the section presents two or more scientists, students, or hypotheses disagreeing about a phenomenon — why a lake's fish population crashed, what formed a crater, how a disease spreads. Unlike the other formats, this passage is text-heavy, often with few or no figures, making it the most time-consuming set; many test takers deliberately save it for last so a slow start cannot starve the quicker data passages. The strategy is bookkeeping. Read the introductory paragraph, which describes the phenomenon everyone agrees needs explaining. Then read Viewpoint 1 and compress it to a single sentence in your head or margin: 'Scientist 1 says the crash came from invasive predators.' Do the same for each subsequent viewpoint, noting explicitly where each agrees and disagrees with the others — often they accept the same observations but propose different mechanisms. Question types are predictable. Some target one viewpoint alone: 'According to Scientist 2, the decline began because...' — pure retrieval from that paragraph. Others test the mapping: 'Both scientists would agree that...' Still others introduce a new finding and ask whose position it strengthens or weakens: if researchers later find predator populations were flat during the crash, Scientist 1's explanation is weakened. Treat these like courtroom exhibits — ask which claim the new fact supports, contradicts, or leaves untouched. Wrong answers commonly attribute one scientist's claim to the other or overstate a position, the same trap family as paired reading passages. You never need to decide who is actually right; the test only asks what each viewpoint says and implies.
Why Outside Knowledge Is Rare, and Reading Tables Fast
Nearly every science question is answerable from the passage's figures and text alone — the section measures reasoning about presented evidence, not recall. A small handful of questions per test draw on basic background knowledge from typical school science, such as knowing that water freezes at 0 degrees Celsius and boils at 100, that plants take in carbon dioxide during photosynthesis, or that denser objects sink — but these are the exception. The practical consequence: do not panic when a passage discusses electrophoresis gels or stellar spectra you have never seen. Treat unfamiliar jargon as a label — 'Compound X,' 'Enzyme Q' — and follow the numbers, because the relationships in the data are what get tested. This mindset also dictates speed technique. Go to the figures first: read each figure's title, then both axis labels with their units, then the legend distinguishing multiple curves. Skim the setup text in seconds — it usually just explains how the data were collected — and head to the questions, returning to the text only when a question requires it. When a question names a specific variable, your first move is locating which table column or graph axis carries that exact label and unit; most wrong answers come from reading the neighboring column or the wrong curve. Watch for three recurring traps: mismatched units between question and table (milliliters versus liters), axes that do not start at zero, and figures where the variable of interest increases downward or leftward. At roughly one minute per question, this figures-first discipline is what makes the pace comfortable rather than frantic.
Key terms
- Optional science section
- — On the enhanced ACT (2025 onward), a 40-question, 40-minute add-on that yields its own score but does not affect the Composite.
- STEM score
- — A reported score averaging the math and science section scores for students who take the optional science section.
- Data representation
- — The passage format presenting graphs and tables with minimal text, testing pure data-reading skills.
- Research summaries
- — The passage format describing related experiments and testing reasoning about design, variables, and results.
- Conflicting viewpoints
- — The text-heavy passage format in which multiple scientists offer competing explanations of one phenomenon.
- Independent variable
- — The factor an experimenter deliberately changes between trials.
- Dependent variable
- — The outcome measured in response to changes in the independent variable.
- Control
- — A baseline trial or group, often untreated, used as the standard of comparison for experimental effects.
- Direct relationship
- — A pattern in which two variables increase or decrease together.
- Inverse relationship
- — A pattern in which one variable increases while the other decreases.
- Interpolation
- — Estimating a value that falls between measured data points by assuming the trend continues smoothly.
- Extrapolation
- — Extending an observed trend beyond the measured range to predict an unmeasured value.
Exam tips
- Confirm whether you need this section at all: it is optional and excluded from the Composite, and requirements vary by college, scholarship, and state — check each school on your list.
- Go figures-first: read titles, axis labels, units, and legends before the passage text, since most answers come straight from the data.
- Units and scales are the classic trap — verify that the question's units match the table's, and check whether an axis skips values or starts above zero.
- Consider saving the conflicting viewpoints passage for last; it is the most reading-heavy set, and stalling on it early can starve the faster data passages.
- Never let unfamiliar jargon rattle you — treat unknown terms as labels, follow the numeric relationships, and remember only a few questions per test need outside science knowledge.