NurseSavvy™
Log inSign Up

Free · No signup · Teaching model

NCLEX CAT Simulator: How the Exam Decides to Stop

This simulator shows the one idea behind NCLEX computer adaptive testing: an ability estimate that moves with every answer, and a confidence band around it that has to clear the passing line before the exam can stop. There are no questions here. You press buttons for right and wrong answers and watch what happens.

It is a simplified teaching model. It is not NCSBN’s scoring system, and it is not the engine behind NurseSavvy’s CAT readiness exam. Use it to understand why the length of an exam says so little about the result, not to predict your own.

Interactive · how the NCLEX decides
Ability estimate+0.00
−4.0passing standard+4.0

0

questions answered

At passing

difficulty of the next question

Locked until 85

stopping rule

This simplified model starts at the passing standard so the movement is easy to see. The official NCLEX FAQ says the real exam begins with a relatively low-difficulty item. Answer a few here and watch the estimate move.

Borderline = right on every question pitched below the passing line, wrong on every one at or above it.

This simplifies the public scoring logic for teaching. NurseSavvy’s production CAT also enforces the content blueprint, case studies, and format pacing. Take the full CAT free.

What the model does, and what it leaves out

  • The model starts with the ability estimate sitting on the passing standard so the movement is easy to see. That is a simplification: the official NCLEX FAQ says the real exam begins with a relatively low-difficulty item.
  • For each simulated answer, the model picks one of six difficulty tiers from the current estimate (a higher estimate draws a harder tier), compares your answer with what the estimate predicted, and moves the estimate up or down. The step shrinks as answers accumulate, so early answers move the estimate more than late ones.
  • The shaded band is a 95% confidence interval. In this model its width depends only on how many answers have been given, which is one of the simplifications.
  • The stopping rule stays locked until the real exam’s minimum number of questions. After that, the run ends as soon as the whole band sits above the passing line (shown as Would pass) or below it (Would fail). If the band still straddles the line at the real exam’s maximum length, the run ends and shows Max reached.
  • The buttons answer one right, one wrong, 20 right in a row, 20 wrong in a row, or 20 borderline: right on every question pitched below the passing line and wrong on every one at or above it, which keeps the estimate on the line and runs the model to the maximum. Reset starts over. Nothing is saved, and no account is needed.
  • It leaves out everything except that ability-and-confidence loop: there are no real questions, content areas, case studies, unscored pretest items, or time limit. NCSBN’s production system is proprietary, and NurseSavvy’s own production CAT also enforces the content blueprint, case studies, and format pacing. Would pass and Would fail describe the state of this model only.

The reasoning behind it

Questions people ask

Is this how the real NCLEX is scored?
No. It is a teaching model that isolates one idea: the exam stops when it is 95% confident your ability is above or below the passing standard. NCSBN’s actual system is proprietary and more complex, and this simulator is also simpler than the engine behind NurseSavvy’s practice CAT.
Does the NCLEX shutting off at the minimum number of questions mean I passed?
No. It means the computer became confident quickly, in one direction or the other. A band entirely above the passing line and a band entirely below it are equally valid reasons to stop, so a short exam is consistent with a clear pass and with a clear fail.
Why will the simulator not stop early even when every answer is right?
The stopping rule is locked until the minimum number of questions, the same as the real exam. The real NCLEX cannot stop before its minimum for anyone, no matter how well they are doing.
What makes an NCLEX run to the maximum number of questions?
An ability estimate that stays close to the passing line. The confidence band never fully clears the line, so the exam keeps asking for more evidence. At the maximum, the real exam decides using your final ability estimate.
Why do early answers move the estimate more than later ones?
The size of each adjustment shrinks as evidence accumulates. That is why a rocky start is recoverable, and why a rocky finish is more expensive.

Sources

How we source and review

Keep going

← All free nursing tools