How does adaptive module routing work on the Digital SAT?

The Digital SAT is what testing people call multistage adaptive. That is narrower than "adaptive" usually means, and the distinction matters.

A fully adaptive test — most computerised placement tests — picks each question based on how you answered the last one. The Digital SAT does not. It adapts once per section, between the two modules.

What happens, in order#Link to the section: What happens, in order

Reading and Writing module 1: 27 questions, 32 minutes. Everyone sits the same difficulty mix — a deliberate spread from easy to hard.

The test then scores that module and routes you. Do well and you get the harder second module; do less well and you get the easier one. You are not told which you received, and there is no reliable way to tell from inside it, because both contain a range of difficulties.

Reading and Writing module 2: 27 questions, 32 minutes, on whichever path you were routed to.

Then a ten-minute break, and Math repeats the structure: 22 questions in 35 minutes, routed, 22 more in 35 minutes. Math routing is decided entirely by your Math module 1. A strong Reading and Writing section does not route you into harder Math.

Why the two paths score differently#Link to the section: Why the two paths score differently

This is the part most practice material either gets wrong or does not model at all.

Getting 20 of 22 right on the harder Math module is not the same performance as 20 of 22 on the easier one, and the scoring cannot treat it as though it were. So each second module has its own conversion from raw correct count to section score, and the two curves differ.

The practical consequence: the easier path has a ceiling. A perfect score on the easier second module does not produce an 800 — it produces something meaningfully below it, because the test has already observed that you did not clear the bar module 1 set. That is not a punishment. It is the test declining to pretend it saw evidence it did not see.

What this means for practice#Link to the section: What this means for practice

First, a practice test that is not adaptive cannot tell you what you would score. It can tell you whether you know the content, which is worth knowing. But a flat 98-question test has no routing step, so it has no principled way to convert your raw score, and whatever number it reports has no mechanism behind it.

Second, module 1 carries more weight than its question count suggests. It sets part of your raw score and decides which ceiling you are scoring against. Rushing module 1 to bank time for module 2 is exactly backwards.

Third, and least intuitively: there is no strategic reason to underperform on module 1. Students sometimes ask whether an easier second module is an easier path to a good score. It is not — it is a lower maximum. Routing is a measurement, not a difficulty setting you would want to choose.

What routing does not do#Link to the section: What routing does not do

It does not change within a module. Once you are in module 2, the questions were fixed the moment you entered it. Answering three in a row correctly does not summon harder ones.

It does not carry across sections. Reading and Writing and Math are routed and scored independently, then added.

And it does not reset if you flag a question and come back to it. Flagging is navigation inside the module you are already in.

Our mocks implement all of this — the same module sizes, the same clock, one routing decision per section, and two separate conversion tables so the path you took is reflected in the score you get.