ACT® · The Enhanced ACT: Format, Scoring, and the Clock
How Field-Test Items Got Embedded
The exam used to give a whole separate unscored section to try out new questions. Now it hides those trial questions inside the real, scored sections so nobody can tell which ones don't count.
A labelled extra section had a problem: everyone knew it was unscored, so effort dropped and the data on new questions came out unreliable. Burying trial items among scored ones fixes that. A student facing a mixed section has no way to flag the fakes, so every question gets treated like it counts, and the people building future forms get honest difficulty data instead of noise.
The hiding method depends on what kind of question is being tested. English, reading and science questions are read in sets tied to a passage, so a trial question can't be tested alone without its passage context — pull the passage and the questions make no sense. That's why those sections smuggle in a whole extra passage with its own cluster of items. Math questions stand on their own, with no shared passage, so trial items there can be scattered one at a time among the scored ones without needing any extra scaffolding.
This connects directly to timing and format: since nobody can identify the trial material, the clock and question count for a section already bake in the extra passage or scattered items, which is part of why the sections feel long for what they score. It also explains why skipping effort anywhere is a bad plan — the whole system only works because test-takers can't play favorites among questions.
Key terms
- Field-test item
- A trial question embedded among scored ones to measure its difficulty before it appears on a real future form.
- Scored section
- The part of a test whose questions count toward the reported score.
- Passage-based item
- A question that depends on a shared reading passage, so it must be trialled together with that passage.
- Linear test
- A test where every taker gets the same fixed set of questions in the same order, unlike an adaptive test.
Watch it explained
- How it worksHow does burying trial items among scored ones produce better data than a labelled fifth test?
- How it worksWalk me through why English, reading and science field test a whole passage while math scatters items.
- Common mistakeWhy isn't it worth trying to save effort on questions you suspect will not be scored?
- Going furtherWhat happens to a future test form when students guess randomly on the trial questions?
Check yourself
End of topic.