CISSP CAT exam: how computerized adaptive testing works

How ability estimation works, in plain words

Think of a good interviewer. They open with a middling question. If you handle it well, the next one goes deeper; if you struggle, they step back and probe a little lower. After a while they have a fair idea of your level, and each extra question changes their view less.

The CAT does the same with numbers. Every question in the pool has a difficulty, set from how earlier candidates answered it while it was being pretested. Your ability is held as an estimate with a band of uncertainty around it. A correct answer nudges the estimate up, a wrong one nudges it down, and the band narrows as evidence builds.

What this means for how the exam feels

• You will get plenty wrong, even when you are passing. An adaptive test keeps aiming near the edge of your ability, because that is where an answer tells it the most. Feeling stretched is the design, not a warning.
• You cannot read your progress from the questions. Difficulty is a statistical property, not how obscure a question feels, and every domain still appears. ISC2 builds the CAT on the same content outline as its fixed-form exams, so you see all eight domains in line with their weights.
• Every scored answer counts. Your result rests on all your operational items, so a shaky start can be recovered.

When the exam stops: the three rules

ISC2 applies these rules in order.

The 25 unscored questions

Every sitting includes 25 pretest questions that ISC2 is still calibrating. They count towards the minimum of 100 but not towards your result. You cannot tell which ones they are, so give every question the same care.

No going back

Once you confirm an answer, it is final. There is no review screen, no flag for later and no skipping. Many people lose time by second-guessing in practice tests that allow review, then freeze on the day when they cannot. Train the habit you need:

1. Read the last line of the question first, so you know what is being asked (first, best, most, least).
2. Read the scenario for the facts that matter: role, goal, constraint.
3. Eliminate what is technically true but not the answer to this question.
4. Choose, confirm, and let it go.

Pacing

180 minutes for up to 150 questions gives you a little over a minute each at full length. Breaks come out of the same clock. A steady pace with no long stalls matters more than speed, because the run-out-of-time rule punishes an unfinished minimum hard.

Why stopping at 100 is not a bad sign

An exam that ends at the minimum means the evidence was already clear, in either direction. Strong candidates often finish there, and so do candidates who were clearly not ready. Running to 150 means your estimate sat close to the line for a long time, and that result can go either way too. The question count tells you nothing on its own, so do not spend the exam counting.

Myths that refuse to die

• "The exam is 125–175 questions in four hours." That was the older CAT format; some guides even quote six-hour sittings from the linear era. Neither applies now: it is 100–150 questions in 3 hours.
• "Harder questions mean you are passing." You cannot judge difficulty reliably from inside the exam, and the pool is mixed across domains.
• "Get the first ten right or you are finished." Early answers move the estimate more, but later ones keep correcting it.
• "You need a percentage to pass." There is no fixed percentage; see how the passing score works.

Next steps

Reading about adaptive testing is not the same as sitting one. Our CAT simulator runs our own practice questions under the same rules: a variable length, a clock and no going back. Then plan the rest of your preparation with how to pass the CISSP.